0: print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) *\",\n \"init_reweighting_prec self.decreasing = decreasing self.final_reweighting_prec = final_reweighting_prec super(PolyatomicFWSolverForLasso, self).__init__(data, forwardOp,\",\n \"def combine_new_impulse(self): iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind is not None:\",\n \"triggered in practice print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue])\",\n \"from pycsou.func.penalty import L1Norm from pycsou.func.loss import SquaredL2Loss from pycsou.opt.proxalgs\",\n \"strategy must be in [\\\"relative_improvement\\\", \\\"certificate\\\"]') def restricted_support_lasso(self, active_indices: np.ndarray,\",\n \"self.last_weight = iterand[active_indices] overvalue = np.abs(iterand) > self.bound if overvalue.sum()\",\n \"at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound return {'iterand':\",\n \"2/(self.iter + 3) else: new_positions = self.old_iterand['positions'] if self.step_size ==\",\n \"be in ['optimal', 'regular']\\\") super(VanillaFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter,\",\n \"self.diagnostics.loc[ self.iter - 1, 'Objective Function'] if self.remember_iterand: self.iterand_history.append(self.iterand['iterand']) def\",\n \"if self.iter > 0 and self.remove_positions: active_indices = np.unique(np.hstack([iterand.nonzero()[0], self.new_ind]))\",\n \"new_positions else: new_positions = self.old_iterand['positions'] if self.iter > 0 and\",\n \"print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound return\",\n \"Function']) / \\\\ self.diagnostics.loc[ self.iter - 1, 'Objective Function'] if\",\n \"iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound if self.decreasing: self.reweighting_prec = self.init_reweighting_prec\",\n \"= remove_positions self.decreasing = decreasing self.dim = self.forwardOp.shape[1] self.x0 =\",\n \"self.diagnostics.loc[self.iter - 1, 'Dual Certificate Value'] return abs(abs(value) - 1)\",\n \"remove_positions self.decreasing = decreasing self.dim = self.forwardOp.shape[1] self.x0 = np.zeros(self.dim)\",\n \"new_positions if active_indices.shape[0] > 1: iterand[self.new_ind] = np.sign(self.dual_certificate_value) * self.last_weight\",\n \"gamma = np.dot(self.data_fidelity.gradient(iterand), iterand) + self.lambda_ * ( 1. *\",\n \"lambda_factor self.new_ind = None self.epsilon = None self.remember_iterand = remember_iterand\",\n \"= 'CD' tau = None elif self.reweighting == 'ista': tau\",\n \"(1 / 2) * SquaredL2Loss(dim=self.forwardOp.shape[0], data=self.data) self.data_fidelity = l22_loss *\",\n \"# print(\\\"Threshold: {} / {}\\\".format(threshold, maxi)) # print('Candidate indices: {}\\\\n'.format(indices.shape))\",\n \"{'iterand': iterand, 'positions': new_positions} class PolyatomicFWSolverForLasso(GenericFWSolverForLasso): def __init__(self, data: np.ndarray,\",\n \"combine_new_impulse(self) -> Any: pass def update_diagnostics(self): \\\"\\\"\\\" Dual ceritificate value\",\n \"Any: pass def update_diagnostics(self): \\\"\\\"\\\" Dual ceritificate value is computed\",\n \"= self.lambda_ * L1Norm(dim=self.dim) objective_functional = self.data_fidelity + self.penalty self.bound\",\n \"np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True) if self.verbose is not None: self.candidate_new.append(indices.shape[0]) self.actual_new.append(self.new_ind.size)\",\n \"self.diagnostics.loc[self.iter, 'Dual Certificate Value'] = self.dual_certificate_value # before iteration self.diagnostics.loc[self.iter,\",\n \"self.diagnostics = DataFrame( columns=['Iter', 'Relative Improvement Objective', 'Relative Improvement Iterand',\",\n \"final_reweighting_prec super(PolyatomicFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold,\",\n \"+ self.penalty self.bound = np.linalg.norm(self.data) ** 2 / (2 *\",\n \"bool = False, remove_positions: bool = False, reweighting_prec: float =\",\n \"'regular']\\\") super(VanillaFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold,\",\n \"x0=x0) if self.new_ind is not None: self.last_weight = iterand[self.new_ind] else:\",\n \"= None self.epsilon = None self.remember_iterand = remember_iterand self.iterand_history =\",\n \"self.dual_certificate_value = dual_certificate[self.new_ind] if self.new_ind in self.old_iterand['positions']: self.new_ind = None\",\n \"= 2/(self.iter + 3) if not 0 < gamma <\",\n \"== 0: self.diagnostics = DataFrame( columns=['Iter', 'Relative Improvement Objective', 'Relative\",\n \"return {'iterand': iterand, 'positions': new_positions} class FullyCorrectiveFWSolverForLasso(VanillaFWSolverForLasso): def __init__(self, data:\",\n \"= np.unique(np.hstack([iterand.nonzero()[0], self.new_ind])) else: active_indices = new_positions else: new_positions =\",\n \"None: print('Warning, dual certificate lower than 1 at iteration {}'.format(self.iter))\",\n \"Iterand', 'Dual Certificate Value', 'Objective Function']) self.diagnostics.loc[self.iter, 'Iter'] = self.iter\",\n \"= 1.9 / restricted_data_fidelity.diff_lipschitz_cst acceleration = None else: raise ValueError('Reweighting\",\n \"self.bound) gamma /= np.linalg.norm(self.forwardOp.mat[:, self.new_ind] * self.bound * np.sign( self.dual_certificate_value)\",\n \"min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, multi_spikes=True, multi_spikes_threshold=multi_spikes_threshold, reweighting='ista', t_max=t_max)\",\n \"np.ndarray, accuracy: float, x0: np.ndarray = None, d: float =\",\n \"active_indices, dtype=float).get_adjointOp() restricted_forward = pl.DenseLinearOperator( self.forwardOp.mat[:, active_indices]) restricted_forward.compute_lipschitz_cst(tol=1e-3) restricted_data_fidelity =\",\n \"self.lambda_ * np.linalg.norm(iterand, 1) gamma /= np.linalg.norm(self.forwardOp @ iterand, 2)\",\n \"self.old_iterand['iterand'] - self.iterand['iterand']) / np.linalg.norm( self.old_iterand['iterand']) self.diagnostics.loc[self.iter, 'Dual Certificate Value']\",\n \"= max(dual_certificate.min(), dual_certificate.max(), key=abs) else: self.new_ind = np.argmax(d) self.dual_certificate_value =\",\n \"False, remember_iterand: bool = False, decreasing: bool = False, multi_spikes_threshold:\",\n \"d = np.abs(dual_certificate) if self.multi_spikes: maxi = np.max(d) if self.iter\",\n \"= 'optimal', t_max: float = None): if step_size in ['optimal',\",\n \"if self.new_ind is not None: new_positions = np.unique(np.hstack([self.old_iterand['positions'], self.new_ind])) if\",\n \"'Dual Certificate Value'] = self.dual_certificate_value # before iteration self.diagnostics.loc[self.iter, 'Objective\",\n \"self.dim = self.forwardOp.shape[1] self.x0 = np.zeros(self.dim) self.dual_certificate_value = 1 /\",\n \"self.iter == 0: threshold = self.multi_spikes_threshold * maxi self.epsilon =\",\n \"max(dual_certificate.min(), dual_certificate.max(), key=abs) else: self.new_ind = np.argmax(d) self.dual_certificate_value = dual_certificate[self.new_ind]\",\n \"Improvement Iterand'] = np.linalg.norm( self.old_iterand['iterand'] - self.iterand['iterand']) / np.linalg.norm( self.old_iterand['iterand'])\",\n \"+= gamma * np.sign(self.dual_certificate_value) * self.bound return {'iterand': iterand, 'positions':\",\n \"(np.linalg.norm(column, 2) ** 2) self.last_weight = iterand[active_indices] overvalue = np.abs(iterand)\",\n \"== 'relative_improvement': return abs(self.diagnostics.loc[self.iter - 1, 'Relative Improvement Objective']) elif\",\n \"self.stopping_strategy == 'certificate': value = self.diagnostics.loc[self.iter - 1, 'Dual Certificate\",\n \"/ restricted_data_fidelity.diff_lipschitz_cst acceleration = None else: raise ValueError('Reweighting strategy must\",\n \"self.multi_spikes = multi_spikes self.multi_spikes_threshold = multi_spikes_threshold self.reweighting = reweighting self.remove_positions\",\n \"= self.lambda_ * L1Norm(dim=restricted_data_fidelity.shape[1]) if self.reweighting == 'fista': acceleration =\",\n \"'certificate', accuracy_threshold: float = 1e-4, verbose: Optional[int] = 10, remember_iterand:\",\n \"0 < gamma < 1: gamma = np.clip(gamma, 0., 1.)\",\n \"np.infty else: self.diagnostics.loc[self.iter, 'Relative Improvement Iterand'] = np.linalg.norm( self.old_iterand['iterand'] -\",\n \"else: active_indices = new_positions else: new_positions = self.old_iterand['positions'] if self.iter\",\n \"copy import deepcopy from abc import abstractmethod from utils import\",\n \"None self.epsilon = None self.remember_iterand = remember_iterand self.iterand_history = []\",\n \"> self.bound if overvalue.sum() > 0: print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue]))\",\n \"active_indices = new_positions else: new_positions = self.old_iterand['positions'] if self.iter >\",\n \"= reweighting_prec super(FullyCorrectiveFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy,\",\n \"np.sign(iterand[overvalue]) * self.bound return {'iterand': iterand, 'positions': new_positions} class PolyatomicFWSolverForLasso(GenericFWSolverForLasso):\",\n \"x0 is None: x0 = np.zeros(active_indices.shape) injection = pl.sampling.SubSampling(self.dim, active_indices,\",\n \"self.iterand_history = [] init_iterand = {'iterand': self.x0, 'positions': np.array([], dtype=int)}\",\n \"super(PolyatomicFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose,\",\n \"None if verbose is not None: self.candidate_new = [] self.actual_new\",\n \"dual_certificate.max(), key=abs) else: self.new_ind = np.argmax(d) self.dual_certificate_value = dual_certificate[self.new_ind] if\",\n \"if len(self.new_ind) == 0: self.new_ind = None self.dual_certificate_value = max(dual_certificate.min(),\",\n \"------- \\\"\\\"\\\" if self.iter == 0: self.diagnostics = DataFrame( columns=['Iter',\",\n \"if self.new_ind is not None: new_positions = np.hstack([self.old_iterand['positions'], self.new_ind]) if\",\n \"- 1, 'Objective Function'] if self.remember_iterand: self.iterand_history.append(self.iterand['iterand']) def print_diagnostics(self): print(dict(self.diagnostics.loc[self.iter]))\",\n \"self.data_fidelity.gradient(self.old_iterand['iterand']) / self.lambda_ d = np.abs(dual_certificate) if self.multi_spikes: maxi =\",\n \"iteration {}'.format(self.iter)) @abstractmethod def combine_new_impulse(self) -> Any: pass def update_diagnostics(self):\",\n \"'Objective Function']) self.diagnostics.loc[self.iter, 'Iter'] = self.iter if np.linalg.norm(self.old_iterand['iterand']) == 0:\",\n \"\\\"ista\\\"]') solver = APGD(dim=restricted_data_fidelity.shape[1], F=restricted_data_fidelity, G=restricted_regularization, x0=x0, tau=tau, acceleration=acceleration, verbose=None,\",\n \"500, stopping_strategy: str = 'certificate', accuracy_threshold: float = 1e-4, verbose:\",\n \"[] super(GenericFWSolverForLasso, self).__init__(objective_functional=objective_functional, init_iterand=init_iterand, max_iter=max_iter, min_iter=min_iter, accuracy_threshold=accuracy_threshold, verbose=verbose, t_max=t_max) def\",\n \"never been triggered in practice print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue]\",\n \"= False, step_size: str = 'optimal', t_max: float = None):\",\n \"= 10, remember_iterand: bool = False, remove_positions: bool = False,\",\n \"2) ** 2) overvalue = np.abs(iterand) > self.bound if overvalue.sum()\",\n \"self.multi_spikes_threshold * maxi self.epsilon = (1 - self.multi_spikes_threshold) * maxi\",\n \"acceleration=acceleration, verbose=None, accuracy_threshold=accuracy, d=d, max_iter=2000, min_iter=1) return injection(solver.iterate()[0]['iterand']) class VanillaFWSolverForLasso(GenericFWSolverForLasso):\",\n \"self.stopping_strategy = stopping_strategy self.accuracy_threshold = accuracy_threshold self.multi_spikes = multi_spikes self.multi_spikes_threshold\",\n \"= [] init_iterand = {'iterand': self.x0, 'positions': np.array([], dtype=int)} l22_loss\",\n \"verbose: Optional[int] = 10, remember_iterand: bool = False, remove_positions: bool\",\n \"self.compute_new_impulse() res = self.combine_new_impulse() return res def compute_new_impulse(self): dual_certificate =\",\n \"active_indices = np.unique(iterand.nonzero()[0]) else: active_indices = new_positions if active_indices.shape[0] >\",\n \"float = .7, multi_spikes: bool = True, reweighting: str =\",\n \"x0=x0) else: tmp = np.zeros(self.dim) tmp[active_indices] = 1. column =\",\n \"threshold = maxi - (1 / (self.iter + 2)) *\",\n \"self.actual_new.append(self.new_ind.size) if len(self.new_ind) == 0: self.new_ind = None self.dual_certificate_value =\",\n \"else: raise ValueError(\\\"Step size strategy must be in ['optimal', 'regular']\\\")\",\n \"self.data_fidelity + self.penalty self.bound = np.linalg.norm(self.data) ** 2 / (2\",\n \"np.sign( self.dual_certificate_value) - self.forwardOp @ iterand, 2) ** 2 else:\",\n \"t_max: float = None): if step_size in ['optimal', 'regular']: self.step_size\",\n \"= None): self.data = data self.forwardOp = forwardOp self.stopping_strategy =\",\n \"lambda_ is None: lambda_ = lambda_factor * np.abs(self.forwardOp.adjoint(self.data)).max() self.lambda_ =\",\n \"multi_spikes_threshold=multi_spikes_threshold, reweighting='ista', t_max=t_max) def combine_new_impulse(self): iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind\",\n \"multi_spikes=False, t_max=t_max) def combine_new_impulse(self) -> Any: iterand = deepcopy(self.old_iterand['iterand']) if\",\n \"than 1 at iteration {}'.format(self.iter)) @abstractmethod def combine_new_impulse(self) -> Any:\",\n \"gamma * np.sign(self.dual_certificate_value) * self.bound return {'iterand': iterand, 'positions': new_positions}\",\n \"self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand,\",\n \"self.iter > 0 and self.remove_positions: active_indices = np.unique(np.hstack([iterand.nonzero()[0], self.new_ind])) else:\",\n \"new_positions = np.unique(np.hstack([self.old_iterand['positions'], self.new_ind])) if self.iter > 0 and self.remove_positions:\",\n \"combine_new_impulse(self): iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind is not None: new_positions\",\n \"Iterand'] = np.infty else: self.diagnostics.loc[self.iter, 'Relative Improvement Iterand'] = np.linalg.norm(\",\n \"SquaredL2Loss(dim=restricted_forward.shape[0], data=self.data) \\\\ * restricted_forward # restricted_data_fidelity.lipschitz_cst = self.data_fidelity.lipschitz_cst #\",\n \"L1Norm(dim=restricted_data_fidelity.shape[1]) if self.reweighting == 'fista': acceleration = 'CD' tau =\",\n \"iterand, 'positions': new_positions} class PolyatomicFWSolverForLasso(GenericFWSolverForLasso): def __init__(self, data: np.ndarray, forwardOp:\",\n \"= 1. column = self.forwardOp(tmp) iterand[active_indices] = np.dot(self.data, column) /\",\n \"value = self.diagnostics.loc[self.iter - 1, 'Dual Certificate Value'] return abs(abs(value)\",\n \"self.restricted_support_lasso(active_indices, self.reweighting_prec, x0=x0) else: tmp = np.zeros(self.dim) tmp[active_indices] = 1.\",\n \"\\\"\\\"\\\" if self.iter == 0: self.diagnostics = DataFrame( columns=['Iter', 'Relative\",\n \"None else: raise ValueError('Reweighting strategy must be in [\\\"fista\\\", \\\"ista\\\"]')\",\n \"column) / (np.linalg.norm(column, 2) ** 2) self.last_weight = iterand[active_indices] overvalue\",\n \"/ np.linalg.norm( self.old_iterand['iterand']) self.diagnostics.loc[self.iter, 'Dual Certificate Value'] = self.dual_certificate_value #\",\n \"accuracy_threshold: float = 1e-4, verbose: Optional[int] = 10, remove_positions: bool\",\n \"- self.multi_spikes_threshold) * maxi else: threshold = maxi - (1\",\n \".7, multi_spikes: bool = True, reweighting: str = 'ista', t_max:\",\n \"# print('Candidate indices: {}\\\\n'.format(indices.shape)) self.new_ind = np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True) if\",\n \"None: x0 = np.zeros(active_indices.shape) injection = pl.sampling.SubSampling(self.dim, active_indices, dtype=float).get_adjointOp() restricted_forward\",\n \"self.last_weight = self.bound def combine_new_impulse(self) -> Any: iterand = deepcopy(self.old_iterand['iterand'])\",\n \"remove_positions self.reweighting_prec = reweighting_prec super(FullyCorrectiveFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter,\",\n \"accuracy_threshold self.multi_spikes = multi_spikes self.multi_spikes_threshold = multi_spikes_threshold self.reweighting = reweighting\",\n \"> self.bound if overvalue.sum() > 0: #Sanity check, never been\",\n \"verbose: Optional[int] = 10, remember_iterand: bool = False, step_size: str\",\n \"pl.sampling.SubSampling(self.dim, active_indices, dtype=float).get_adjointOp() restricted_forward = pl.DenseLinearOperator( self.forwardOp.mat[:, active_indices]) restricted_forward.compute_lipschitz_cst(tol=1e-3) restricted_data_fidelity\",\n \"print('Warning, dual certificate lower than 1 at iteration {}'.format(self.iter)) @abstractmethod\",\n \"self.restricted_support_lasso(active_indices, self.reweighting_prec, x0=x0) if self.new_ind is not None: self.last_weight =\",\n \"self.lambda_) self.start = None if verbose is not None: self.candidate_new\",\n \"self.forwardOp @ iterand, 2) ** 2 else: gamma = 2/(self.iter\",\n \"np.array([], dtype=int)} l22_loss = (1 / 2) * SquaredL2Loss(dim=self.forwardOp.shape[0], data=self.data)\",\n \"None self.remember_iterand = remember_iterand self.iterand_history = [] init_iterand = {'iterand':\",\n \"np.sign(self.dual_certificate_value) * self.last_weight x0 = iterand[active_indices] iterand = self.restricted_support_lasso(active_indices, self.reweighting_prec,\",\n \"float = None): self.remove_positions = remove_positions self.reweighting_prec = reweighting_prec super(FullyCorrectiveFWSolverForLasso,\",\n \"2) ** 2) self.last_weight = iterand[active_indices] overvalue = np.abs(iterand) >\",\n \"- 1, 'Relative Improvement Objective']) elif self.stopping_strategy == 'certificate': value\",\n \"* np.abs(self.forwardOp.adjoint(self.data)).max() self.lambda_ = lambda_ self.penalty = self.lambda_ * L1Norm(dim=self.dim)\",\n \"= np.unique(np.hstack([self.old_iterand['positions'], self.new_ind])) if self.iter > 0 and self.remove_positions: active_indices\",\n \"self.lambda_ = lambda_ self.penalty = self.lambda_ * L1Norm(dim=self.dim) objective_functional =\",\n \"None self.dual_certificate_value = max(dual_certificate.min(), dual_certificate.max(), key=abs) else: self.new_ind = np.argmax(d)\",\n \"from copy import deepcopy from abc import abstractmethod from utils\",\n \"import numpy as np from typing import Optional, Any from\",\n \"self.verbose is not None: self.candidate_new.append(indices.shape[0]) self.actual_new.append(self.new_ind.size) if len(self.new_ind) == 0:\",\n \"abs(self.dual_certificate_value) < 1.: if self.verbose is not None: print('Warning, dual\",\n \"min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, t_max=t_max) self.reweighting = reweighting\",\n \"def restricted_support_lasso(self, active_indices: np.ndarray, accuracy: float, x0: np.ndarray = None,\",\n \"not None: print('Warning, dual certificate lower than 1 at iteration\",\n \"2 else: gamma = 2/(self.iter + 3) if not 0\",\n \"= self.combine_new_impulse() return res def compute_new_impulse(self): dual_certificate = - self.data_fidelity.gradient(self.old_iterand['iterand'])\",\n \"restricted_support_lasso(self, active_indices: np.ndarray, accuracy: float, x0: np.ndarray = None, d:\",\n \"maxi)) # print('Candidate indices: {}\\\\n'.format(indices.shape)) self.new_ind = np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True)\",\n \"float = 1e-4, init_reweighting_prec: float = .2, decreasing: bool =\",\n \"= lambda_factor * np.abs(self.forwardOp.adjoint(self.data)).max() self.lambda_ = lambda_ self.penalty = self.lambda_\",\n \"= self.multi_spikes_threshold * maxi self.epsilon = (1 - self.multi_spikes_threshold) *\",\n \"reweighting: str = 'fista', t_max: float = None): self.remove_positions =\",\n \"= np.zeros(self.dim) tmp[active_indices] = 1. column = self.forwardOp(tmp) iterand[active_indices] =\",\n \"else: self.diagnostics.loc[self.iter, 'Relative Improvement Iterand'] = np.linalg.norm( self.old_iterand['iterand'] - self.iterand['iterand'])\",\n \"Function'] - self.diagnostics.loc[ self.iter, 'Objective Function']) / \\\\ self.diagnostics.loc[ self.iter\",\n \"lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, multi_spikes=False, t_max=t_max) def\",\n \"= multi_spikes_threshold self.reweighting = reweighting self.remove_positions = remove_positions self.decreasing =\",\n \"-> Any: iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind is not None:\",\n \"multi_spikes=True, multi_spikes_threshold=multi_spikes_threshold, reweighting='ista', t_max=t_max) def combine_new_impulse(self): iterand = deepcopy(self.old_iterand['iterand']) if\",\n \"active_indices = new_positions if active_indices.shape[0] > 1: x0 = iterand[active_indices]\",\n \"ValueError('Reweighting strategy must be in [\\\"fista\\\", \\\"ista\\\"]') solver = APGD(dim=restricted_data_fidelity.shape[1],\",\n \"super(FullyCorrectiveFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_, lambda_factor=lambda_factor, min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose,\",\n \"self.remove_positions = remove_positions self.reweighting_prec = reweighting_prec super(FullyCorrectiveFWSolverForLasso, self).__init__(data, forwardOp, lambda_=lambda_,\",\n \"<gh_stars>0 import numpy as np from typing import Optional, Any\",\n \"self.bound if overvalue.sum() > 0: #Sanity check, never been triggered\",\n \"VanillaFWSolverForLasso(GenericFWSolverForLasso): def __init__(self, data: np.ndarray, forwardOp: pcore.linop.LinearOperator, lambda_: Optional[float] =\",\n \"> 0 and self.remove_positions: active_indices = np.unique(iterand.nonzero()[0]) else: active_indices =\",\n \"= 'certificate', accuracy_threshold: float = 1e-4, verbose: Optional[int] = 10,\",\n \"- 1, 'Objective Function'] - self.diagnostics.loc[ self.iter, 'Objective Function']) /\",\n \"np.abs(dual_certificate) if self.multi_spikes: maxi = np.max(d) if self.iter == 0:\",\n \"float, x0: np.ndarray = None, d: float = 75.): if\",\n \"restricted_forward.compute_lipschitz_cst(tol=1e-3) restricted_data_fidelity = (1 / 2) * SquaredL2Loss(dim=restricted_forward.shape[0], data=self.data) \\\\\",\n \"self.data_fidelity = l22_loss * self.forwardOp if lambda_ is None: lambda_\",\n \"{}\\\\n'.format(indices.shape)) self.new_ind = np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True) if self.verbose is not\",\n \"= None # already present position if abs(self.dual_certificate_value) < 1.:\",\n \"Any: self.compute_new_impulse() res = self.combine_new_impulse() return res def compute_new_impulse(self): dual_certificate\",\n \"in [\\\"fista\\\", \\\"ista\\\"]') solver = APGD(dim=restricted_data_fidelity.shape[1], F=restricted_data_fidelity, G=restricted_regularization, x0=x0, tau=tau,\",\n \"step_size in ['optimal', 'regular']: self.step_size = step_size else: raise ValueError(\\\"Step\",\n \"self.iter == 0: self.diagnostics = DataFrame( columns=['Iter', 'Relative Improvement Objective',\",\n \"reweighting='ista', t_max=t_max) def combine_new_impulse(self): iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind is\",\n \"dual certificate lower than 1 at iteration {}'.format(self.iter)) @abstractmethod def\",\n \"min_iter=min_iter, max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, multi_spikes=False, t_max=t_max) def combine_new_impulse(self)\",\n \"combine_new_impulse(self) -> Any: iterand = deepcopy(self.old_iterand['iterand']) if self.new_ind is not\",\n \"None: new_positions = np.unique(np.hstack([self.old_iterand['positions'], self.new_ind])) if self.iter > 0 and\",\n \"+ self.lambda_ * np.linalg.norm(iterand, 1) gamma /= np.linalg.norm(self.forwardOp @ iterand,\",\n \"self.bound if self.decreasing: self.reweighting_prec = self.init_reweighting_prec / (self.iter + 1)\",\n \"None, d: float = 75.): if x0 is None: x0\",\n \"False, reweighting_prec: float = 1e-4, reweighting: str = 'fista', t_max:\",\n \"/= np.linalg.norm(self.forwardOp.mat[:, self.new_ind] * self.bound * np.sign( self.dual_certificate_value) - self.forwardOp\",\n \"max_iter=max_iter, stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, multi_spikes=True, multi_spikes_threshold=multi_spikes_threshold, reweighting='ista', t_max=t_max) def\",\n \"None: self.candidate_new.append(indices.shape[0]) self.actual_new.append(self.new_ind.size) if len(self.new_ind) == 0: self.new_ind = None\",\n \"self.diagnostics.loc[ self.iter, 'Objective Function']) / \\\\ self.diagnostics.loc[ self.iter - 1,\",\n \"if x0 is None: x0 = np.zeros(active_indices.shape) injection = pl.sampling.SubSampling(self.dim,\",\n \"self.new_ind in self.old_iterand['positions']: self.new_ind = None # already present position\",\n \"else: tmp = np.zeros(self.dim) tmp[active_indices] = 1. column = self.forwardOp(tmp)\",\n \"= .7, t_max: float = None): self.remove_positions = remove_positions self.reweighting_prec\",\n \"0: print(\\\"Overvalue at coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound\",\n \"> 1: x0 = iterand[active_indices] iterand = self.restricted_support_lasso(active_indices, self.reweighting_prec, x0=x0)\",\n \"@abstractmethod def combine_new_impulse(self) -> Any: pass def update_diagnostics(self): \\\"\\\"\\\" Dual\",\n \"np.infty elif self.stopping_strategy == 'relative_improvement': return abs(self.diagnostics.loc[self.iter - 1, 'Relative\",\n \"APGD(dim=restricted_data_fidelity.shape[1], F=restricted_data_fidelity, G=restricted_regularization, x0=x0, tau=tau, acceleration=acceleration, verbose=None, accuracy_threshold=accuracy, d=d, max_iter=2000,\",\n \"(self.iter + 1) self.reweighting_prec = max(self.reweighting_prec, self.final_reweighting_prec) return {'iterand': iterand,\",\n \"accuracy_threshold=accuracy_threshold, verbose=verbose, t_max=t_max) def update_iterand(self) -> Any: self.compute_new_impulse() res =\",\n \"stopping_strategy=stopping_strategy, accuracy_threshold=accuracy_threshold, verbose=verbose, remember_iterand=remember_iterand, multi_spikes=False, t_max=t_max) def combine_new_impulse(self) -> Any:\",\n \"self.epsilon = None self.remember_iterand = remember_iterand self.iterand_history = [] init_iterand\",\n \"None): self.data = data self.forwardOp = forwardOp self.stopping_strategy = stopping_strategy\",\n \"before iteration self.diagnostics.loc[self.iter, 'Objective Function'] = self.objective_functional(self.iterand['iterand']) if self.iter ==\",\n \"if self.decreasing: self.reweighting_prec = self.init_reweighting_prec / (self.iter + 1) self.reweighting_prec\",\n \"self.epsilon = (1 - self.multi_spikes_threshold) * maxi else: threshold =\",\n \"- 1.) * self.bound) gamma /= np.linalg.norm(self.forwardOp.mat[:, self.new_ind] * self.bound\",\n \"overvalue.sum() > 0: #Sanity check, never been triggered in practice\",\n \"/ (self.iter + 1) self.reweighting_prec = max(self.reweighting_prec, self.final_reweighting_prec) return {'iterand':\",\n \"pycsou.core as pcore import pycsou.linop as pl from pycsou.func.penalty import\",\n \"= decreasing self.dim = self.forwardOp.shape[1] self.x0 = np.zeros(self.dim) self.dual_certificate_value =\",\n \"== 0: threshold = self.multi_spikes_threshold * maxi self.epsilon = (1\",\n \"1. * np.linalg.norm(iterand, 1) + (np.abs(self.dual_certificate_value) - 1.) * self.bound)\",\n \"= multi_spikes self.multi_spikes_threshold = multi_spikes_threshold self.reweighting = reweighting self.remove_positions =\",\n \"(self.iter + 2)) * self.epsilon indices = np.where(d > max(threshold,\",\n \"(self.diagnostics.loc[ self.iter - 1, 'Objective Function'] - self.diagnostics.loc[ self.iter, 'Objective\",\n \"indices: {}\\\\n'.format(indices.shape)) self.new_ind = np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True) if self.verbose is\",\n \"np.sign(iterand[overvalue]) * self.bound if self.decreasing: self.reweighting_prec = self.init_reweighting_prec / (self.iter\",\n \"coordinates {}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound if self.decreasing: self.reweighting_prec\",\n \"1e-4, init_reweighting_prec: float = .2, decreasing: bool = False, multi_spikes_threshold:\",\n \"init_iterand = {'iterand': self.x0, 'positions': np.array([], dtype=int)} l22_loss = (1\",\n \"ValueError('Stopping strategy must be in [\\\"relative_improvement\\\", \\\"certificate\\\"]') def restricted_support_lasso(self, active_indices:\",\n \"float = .7, t_max: float = None): self.remove_positions = remove_positions\",\n \"self.dual_certificate_value = max(dual_certificate.min(), dual_certificate.max(), key=abs) else: self.new_ind = np.argmax(d) self.dual_certificate_value\",\n \"'Objective Function'] - self.diagnostics.loc[ self.iter, 'Objective Function']) / \\\\ self.diagnostics.loc[\",\n \"return injection(solver.iterate()[0]['iterand']) class VanillaFWSolverForLasso(GenericFWSolverForLasso): def __init__(self, data: np.ndarray, forwardOp: pcore.linop.LinearOperator,\",\n \"= None): if step_size in ['optimal', 'regular']: self.step_size = step_size\",\n \"self.decreasing = decreasing self.dim = self.forwardOp.shape[1] self.x0 = np.zeros(self.dim) self.dual_certificate_value\",\n \"print('Candidate indices: {}\\\\n'.format(indices.shape)) self.new_ind = np.setdiff1d(indices, self.old_iterand['positions'], assume_unique=True) if self.verbose\",\n \"from utils import TimedGenericIterativeAlgorithm import pycsou.core as pcore import pycsou.linop\",\n \"{}\\\".format(np.arange(overvalue.shape[0])[overvalue])) iterand[overvalue] = np.sign(iterand[overvalue]) * self.bound return {'iterand': iterand, 'positions':\",\n \"self.old_iterand['positions'] if self.step_size == 'optimal': gamma = np.dot(self.data_fidelity.gradient(iterand), iterand) +\",\n \"data self.forwardOp = forwardOp self.stopping_strategy = stopping_strategy self.accuracy_threshold = accuracy_threshold\",\n \".2, decreasing: bool = False, multi_spikes_threshold: float = .7, t_max:\",\n \"= pl.sampling.SubSampling(self.dim, active_indices, dtype=float).get_adjointOp() restricted_forward = pl.DenseLinearOperator( self.forwardOp.mat[:, active_indices]) restricted_forward.compute_lipschitz_cst(tol=1e-3)\",\n \"[\\\"relative_improvement\\\", \\\"certificate\\\"]') def restricted_support_lasso(self, active_indices: np.ndarray, accuracy: float, x0: np.ndarray\",\n \"pycsou.linop as pl from pycsou.func.penalty import L1Norm from pycsou.func.loss import\",\n \"0 and self.remove_positions: active_indices = np.unique(np.hstack([iterand.nonzero()[0], self.new_ind])) else: active_indices =\",\n \"TimedGenericIterativeAlgorithm import pycsou.core as pcore import pycsou.linop as pl from\",\n \"10, remove_positions: bool = False, remember_iterand: bool = False, decreasing:\",\n \"reweighting_prec: float = 1e-4, reweighting: str = 'fista', t_max: float\",\n \"self.objective_functional(self.iterand['iterand']) if self.iter == 0: self.diagnostics.loc[self.iter, 'Relative Improvement Objective'] =\",\n \"not None: iterand[self.new_ind] += gamma * np.sign(self.dual_certificate_value) * self.bound return\",\n \"= data self.forwardOp = forwardOp self.stopping_strategy = stopping_strategy self.accuracy_threshold =\",\n \"['optimal', 'regular']: self.step_size = step_size else: raise ValueError(\\\"Step size strategy\",\n \"lambda_ = lambda_factor * np.abs(self.forwardOp.adjoint(self.data)).max() self.lambda_ = lambda_ self.penalty =\",\n \"Certificate Value'] = self.dual_certificate_value # before iteration self.diagnostics.loc[self.iter, 'Objective Function']\",\n \"is None: lambda_ = lambda_factor * np.abs(self.forwardOp.adjoint(self.data)).max() self.lambda_ = lambda_\",\n \"self.remember_iterand: self.iterand_history.append(self.iterand['iterand']) def print_diagnostics(self): print(dict(self.diagnostics.loc[self.iter])) def stopping_metric(self): if self.iter ==\",\n \"= None self.remember_iterand = remember_iterand self.iterand_history = [] init_iterand =\",\n \"self.forwardOp.shape[1] self.x0 = np.zeros(self.dim) self.dual_certificate_value = 1 / lambda_factor self.new_ind\",\n \"= self.old_iterand['positions'] if self.step_size == 'optimal': gamma = np.dot(self.data_fidelity.gradient(iterand), iterand)\",\n \"Optional[int] = 10, remember_iterand: bool = False, remove_positions: bool =\",\n \"in self.old_iterand['positions']: self.new_ind = None # already present position if\",\n \"pandas import DataFrame from copy import deepcopy from abc import\"\n]","isConversational":false}}},{"rowIdx":675,"cells":{"ngram":{"kind":"list like","value":["g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix')","r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier =","isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a","# TODO: Fix once getGraph/getContext is in conjunctive graph if","unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal","from rdflib.term import URIRef, Literal from rdflib.graph import Graph class","URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label')","URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph()","import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass","None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)])","r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3","None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r =","= ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q","and that is the default one, but this is not","None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r","s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None,","bit ugly # we cannot match the bnode to the","2\"))) s2 = URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2,","Graph(identifier = s1) g1.add((r1, label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label,","g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in conjunctive graph","ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in","once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store,","this is a bit ugly # we cannot match the","is not always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if","ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o","# here I know there is only one anonymous graph,","pass def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2","= ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for","graph automagically # here I know there is only one","s2) g2.add((r2, label, Literal(\"label 3\"))) g = ConjunctiveGraph() for s,p,o","know there is only one anonymous graph, # and that","bnode to the right graph automagically # here I know","right graph automagically # here I know there is only","that is the default one, but this is not always","r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO","def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1 =","match the bnode to the right graph automagically # here","g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\"label 3\"))) g","# we cannot match the bnode to the right graph","rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self):","not always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if __name__=='__main__':","= g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r),","tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a bit ugly","g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 =","URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self):","rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph","# and that is the default one, but this is","automagically # here I know there is only one anonymous","only one anonymous graph, # and that is the default","conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode,","g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4)))","label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\"))) s2 =","testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2')","label = URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label,","for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3')","cannot match the bnode to the right graph automagically #","ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph import Graph","label, Literal(\"label 2\"))) s2 = URIRef('store:2') g2 = Graph(identifier =","StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): #","= URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\"label","= Graph(identifier = s1) g1.add((r1, label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1,","anonymous graph, # and that is the default one, but","python import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import","URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier","BNode, this is a bit ugly # we cannot match","URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1,","= Graph(identifier = s2) g2.add((r2, label, Literal(\"label 3\"))) g =","lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\"))) s2 = URIRef('store:2') g2 =","tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1')","Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\"))) s2 = URIRef('store:2')","Literal(\"label 3\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None, None,","= URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 =","format='trix') for q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext","one anonymous graph, # and that is the default one,","default one, but this is not always the case tg=g.default_context","one, but this is not always the case tg=g.default_context self.assertTrue(q[0:3]","import ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph import","URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\"label 1\",","g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for","here I know there is only one anonymous graph, #","= URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 =","g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)):","g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO: Fix once","TODO: Fix once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier,","s2 = URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label,","always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if __name__=='__main__': unittest.main()","to the right graph automagically # here I know there","we cannot match the bnode to the right graph automagically","g2.add((r2, label, Literal(\"label 3\"))) g = ConjunctiveGraph() for s,p,o in","this is not always the case tg=g.default_context self.assertTrue(q[0:3] in tg)","the bnode to the right graph automagically # here I","I know there is only one anonymous graph, # and","setUp(self): pass def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1')","= URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\"label","from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def","import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef,","None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3","1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\"))) s2 = URIRef('store:2') g2","the right graph automagically # here I know there is","#!/usr/bin/env python import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term","g1.add((r1, label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\"))) s2","label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO","graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this","URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a bit","g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from","for q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is","Graph(identifier = s2) g2.add((r2, label, Literal(\"label 3\"))) g = ConjunctiveGraph()","identifier=q[3].identifier) else: # BNode, this is a bit ugly #","is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else:","s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3,","URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\"label 3\")))","graph, # and that is the default one, but this","= s2) g2.add((r2, label, Literal(\"label 3\"))) g = ConjunctiveGraph() for","Fix once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef):","StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO: Fix","pass def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1","import URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def","in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label,","in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: #","def setUp(self): pass def tearDown(self): pass def testSerialize(self): s1 =","s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label","Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass","s1) g1.add((r1, label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label 2\")))","rdflib.term import URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase):","label, Literal(\"label 3\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None,","a bit ugly # we cannot match the bnode to","there is only one anonymous graph, # and that is","but this is not always the case tg=g.default_context self.assertTrue(q[0:3] in","in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None,","g1.add((r1, label, Literal(\"label 2\"))) s2 = URIRef('store:2') g2 = Graph(identifier","# BNode, this is a bit ugly # we cannot","the default one, but this is not always the case","is the default one, but this is not always the","= s1) g1.add((r1, label, Literal(\"label 1\", lang=\"en\"))) g1.add((r1, label, Literal(\"label","if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is","in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in conjunctive","from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal from","q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in","else: # BNode, this is a bit ugly # we","r1 = URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1","ugly # we cannot match the bnode to the right","g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\"label 1\", lang=\"en\")))","getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier)","g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix')","class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def testSerialize(self):","Literal(\"label 2\"))) s2 = URIRef('store:2') g2 = Graph(identifier = s2)","def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 =","Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def","import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO:","3\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)):","g = ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)])","Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import","is only one anonymous graph, # and that is the","= URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label =","is a bit ugly # we cannot match the bnode","for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in","from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)):","= URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier = s1)","TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def testSerialize(self): s1"],"string":"[\n \"g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix')\",\n \"r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier =\",\n \"isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a\",\n \"# TODO: Fix once getGraph/getContext is in conjunctive graph if\",\n \"unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal\",\n \"from rdflib.term import URIRef, Literal from rdflib.graph import Graph class\",\n \"URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label')\",\n \"URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph()\",\n \"import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass\",\n \"None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)])\",\n \"r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3\",\n \"None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r =\",\n \"= ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q\",\n \"and that is the default one, but this is not\",\n \"None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r\",\n \"s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None,\",\n \"bit ugly # we cannot match the bnode to the\",\n \"2\\\"))) s2 = URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2,\",\n \"Graph(identifier = s1) g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label,\",\n \"g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in conjunctive graph\",\n \"ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in\",\n \"once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store,\",\n \"this is a bit ugly # we cannot match the\",\n \"is not always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if\",\n \"ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o\",\n \"# here I know there is only one anonymous graph,\",\n \"pass def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2\",\n \"= ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for\",\n \"graph automagically # here I know there is only one\",\n \"s2) g2.add((r2, label, Literal(\\\"label 3\\\"))) g = ConjunctiveGraph() for s,p,o\",\n \"know there is only one anonymous graph, # and that\",\n \"bnode to the right graph automagically # here I know\",\n \"right graph automagically # here I know there is only\",\n \"that is the default one, but this is not always\",\n \"r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO\",\n \"def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1 =\",\n \"match the bnode to the right graph automagically # here\",\n \"g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\\\"label 3\\\"))) g\",\n \"# we cannot match the bnode to the right graph\",\n \"rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self):\",\n \"not always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if __name__=='__main__':\",\n \"= g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r),\",\n \"tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a bit ugly\",\n \"g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 =\",\n \"URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self):\",\n \"rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph\",\n \"# and that is the default one, but this is\",\n \"automagically # here I know there is only one anonymous\",\n \"only one anonymous graph, # and that is the default\",\n \"conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode,\",\n \"g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4)))\",\n \"label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\"))) s2 =\",\n \"testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2')\",\n \"label = URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label,\",\n \"for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3')\",\n \"cannot match the bnode to the right graph automagically #\",\n \"ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph import Graph\",\n \"label, Literal(\\\"label 2\\\"))) s2 = URIRef('store:2') g2 = Graph(identifier =\",\n \"StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): #\",\n \"= URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\\\"label\",\n \"= Graph(identifier = s1) g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1,\",\n \"anonymous graph, # and that is the default one, but\",\n \"python import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import\",\n \"URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier\",\n \"BNode, this is a bit ugly # we cannot match\",\n \"URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1,\",\n \"= Graph(identifier = s2) g2.add((r2, label, Literal(\\\"label 3\\\"))) g =\",\n \"lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\"))) s2 = URIRef('store:2') g2 =\",\n \"tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1')\",\n \"Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\"))) s2 = URIRef('store:2')\",\n \"Literal(\\\"label 3\\\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None, None,\",\n \"= URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 =\",\n \"format='trix') for q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext\",\n \"one anonymous graph, # and that is the default one,\",\n \"default one, but this is not always the case tg=g.default_context\",\n \"one, but this is not always the case tg=g.default_context self.assertTrue(q[0:3]\",\n \"import ConjunctiveGraph from rdflib.term import URIRef, Literal from rdflib.graph import\",\n \"URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\\\"label 1\\\",\",\n \"g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix') for\",\n \"here I know there is only one anonymous graph, #\",\n \"= URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 =\",\n \"g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)):\",\n \"g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO: Fix once\",\n \"TODO: Fix once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier,\",\n \"s2 = URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label,\",\n \"always the case tg=g.default_context self.assertTrue(q[0:3] in tg) if __name__=='__main__': unittest.main()\",\n \"to the right graph automagically # here I know there\",\n \"we cannot match the bnode to the right graph automagically\",\n \"g2.add((r2, label, Literal(\\\"label 3\\\"))) g = ConjunctiveGraph() for s,p,o in\",\n \"this is not always the case tg=g.default_context self.assertTrue(q[0:3] in tg)\",\n \"the bnode to the right graph automagically # here I\",\n \"I know there is only one anonymous graph, # and\",\n \"setUp(self): pass def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1')\",\n \"= URIRef('predicate:label') g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\\\"label\",\n \"from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def\",\n \"import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef,\",\n \"None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3\",\n \"1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\"))) s2 = URIRef('store:2') g2\",\n \"the right graph automagically # here I know there is\",\n \"#!/usr/bin/env python import unittest from rdflib.graph import ConjunctiveGraph from rdflib.term\",\n \"g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\"))) s2\",\n \"label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO\",\n \"graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this\",\n \"URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is a bit\",\n \"g.add((r3, label, Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from\",\n \"for q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is\",\n \"Graph(identifier = s2) g2.add((r2, label, Literal(\\\"label 3\\\"))) g = ConjunctiveGraph()\",\n \"identifier=q[3].identifier) else: # BNode, this is a bit ugly #\",\n \"is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else:\",\n \"s,p,o in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3,\",\n \"URIRef('store:2') g2 = Graph(identifier = s2) g2.add((r2, label, Literal(\\\"label 3\\\")))\",\n \"graph, # and that is the default one, but this\",\n \"= s2) g2.add((r2, label, Literal(\\\"label 3\\\"))) g = ConjunctiveGraph() for\",\n \"Fix once getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef):\",\n \"StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO: Fix\",\n \"pass def tearDown(self): pass def testSerialize(self): s1 = URIRef('store:1') r1\",\n \"import URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def\",\n \"in g2.triples((None, None, None)): g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label,\",\n \"in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: #\",\n \"def setUp(self): pass def tearDown(self): pass def testSerialize(self): s1 =\",\n \"s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label\",\n \"Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass\",\n \"s1) g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label 2\\\")))\",\n \"rdflib.term import URIRef, Literal from rdflib.graph import Graph class TestTrixSerialize(unittest.TestCase):\",\n \"label, Literal(\\\"label 3\\\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None,\",\n \"a bit ugly # we cannot match the bnode to\",\n \"there is only one anonymous graph, # and that is\",\n \"but this is not always the case tg=g.default_context self.assertTrue(q[0:3] in\",\n \"in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in g2.triples((None, None,\",\n \"g1.add((r1, label, Literal(\\\"label 2\\\"))) s2 = URIRef('store:2') g2 = Graph(identifier\",\n \"# BNode, this is a bit ugly # we cannot\",\n \"the default one, but this is not always the case\",\n \"is the default one, but this is not always the\",\n \"= s1) g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\"))) g1.add((r1, label, Literal(\\\"label\",\n \"if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode, this is\",\n \"in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in conjunctive\",\n \"from rdflib.graph import ConjunctiveGraph from rdflib.term import URIRef, Literal from\",\n \"q in g3.quads((None,None,None)): # TODO: Fix once getGraph/getContext is in\",\n \"else: # BNode, this is a bit ugly # we\",\n \"r1 = URIRef('resource:1') r2 = URIRef('resource:2') label = URIRef('predicate:label') g1\",\n \"ugly # we cannot match the bnode to the right\",\n \"g1 = Graph(identifier = s1) g1.add((r1, label, Literal(\\\"label 1\\\", lang=\\\"en\\\")))\",\n \"getGraph/getContext is in conjunctive graph if isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier)\",\n \"g.addN([(s,p,o,g2)]) r3 = URIRef('resource:3') g.add((r3, label, Literal(4))) r = g.serialize(format='trix')\",\n \"class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def testSerialize(self):\",\n \"Literal(\\\"label 2\\\"))) s2 = URIRef('store:2') g2 = Graph(identifier = s2)\",\n \"def testSerialize(self): s1 = URIRef('store:1') r1 = URIRef('resource:1') r2 =\",\n \"Graph class TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def\",\n \"import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)): # TODO:\",\n \"3\\\"))) g = ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)):\",\n \"g = ConjunctiveGraph() for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)])\",\n \"Literal(4))) r = g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import\",\n \"is only one anonymous graph, # and that is the\",\n \"= URIRef('store:1') r1 = URIRef('resource:1') r2 = URIRef('resource:2') label =\",\n \"is a bit ugly # we cannot match the bnode\",\n \"for s,p,o in g1.triples((None, None, None)): g.addN([(s,p,o,g1)]) for s,p,o in\",\n \"from StringIO import StringIO g3.parse(StringIO(r), format='trix') for q in g3.quads((None,None,None)):\",\n \"= URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier = s1)\",\n \"TestTrixSerialize(unittest.TestCase): def setUp(self): pass def tearDown(self): pass def testSerialize(self): s1\"\n]","isConversational":false}}},{"rowIdx":676,"cells":{"ngram":{"kind":"list like","value":["print \"\\033[35mRevision Updated in the database as %d\\033[0m\" % checkout_revision","updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the database %s\\033[0m\" % checkout_version","sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True)","checkout version, for example, 2.1 -->') if checkout_version != 'trunk':","= dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade)","x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions are 'trunk', '2.1', or '2.1.2'","__name__ == \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to","checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated","the database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in","checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in the database as %d\\033[0m\"","to checkout version, for example, 2.1 -->') if checkout_version !=","= int(raw_input('Revert to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version)","project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \"__main__\": updatelib_inst =","#!/usr/bin/env python from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade","print \"\\033[35mVersion Updated in the database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision)","dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__","from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata',","map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions are 'trunk', '2.1', or","database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in the","except: print \"valid versions are 'trunk', '2.1', or '2.1.2' etc\"","versions are 'trunk', '2.1', or '2.1.2' etc\" raise checkout_revision =","int(raw_input('Revert to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print","python from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade =","if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid","checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions","dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version","to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion","raw_input('Revert to checkout version, for example, 2.1 -->') if checkout_version","Updated in the database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision","= updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for example,","'2.1.2' etc\" raise checkout_revision = int(raw_input('Revert to checkout revision, for","drop=True) if __name__ == \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version =","checkout_revision = int(raw_input('Revert to checkout revision, for example, 16500 -->'))","try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions are 'trunk', '2.1',","!= 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions are","-->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print","== \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout","\"valid versions are 'trunk', '2.1', or '2.1.2' etc\" raise checkout_revision","revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in","\"\\033[35mVersion Updated in the database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print","'2.1', or '2.1.2' etc\" raise checkout_revision = int(raw_input('Revert to checkout","import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \"__main__\":","\"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version,","= raw_input('Revert to checkout version, for example, 2.1 -->') if","or '2.1.2' etc\" raise checkout_revision = int(raw_input('Revert to checkout revision,","in the database %s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated","% checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in the database as","updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for","updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for example, 2.1","example, 2.1 -->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2])","%s\\033[0m\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in the database","version, for example, 2.1 -->') if checkout_version != 'trunk': try:","print \"valid versions are 'trunk', '2.1', or '2.1.2' etc\" raise","example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the database","updatelib_inst.updateDatabaseRevision(checkout_revision) print \"\\033[35mRevision Updated in the database as %d\\033[0m\" %","raise checkout_revision = int(raw_input('Revert to checkout revision, for example, 16500","are 'trunk', '2.1', or '2.1.2' etc\" raise checkout_revision = int(raw_input('Revert","for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the","16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the database %s\\033[0m\"","-->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the database %s\\033[0m\" %","updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \"__main__\": updatelib_inst","for example, 2.1 -->') if checkout_version != 'trunk': try: map((lambda","checkout_version = raw_input('Revert to checkout version, for example, 2.1 -->')","2.1 -->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except:","import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if","etc\" raise checkout_revision = int(raw_input('Revert to checkout revision, for example,","'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \"valid versions are 'trunk',","'trunk', '2.1', or '2.1.2' etc\" raise checkout_revision = int(raw_input('Revert to","dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ ==","if __name__ == \"__main__\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert"],"string":"[\n \"print \\\"\\\\033[35mRevision Updated in the database as %d\\\\033[0m\\\" % checkout_revision\",\n \"updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in the database %s\\\\033[0m\\\" % checkout_version\",\n \"sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True)\",\n \"checkout version, for example, 2.1 -->') if checkout_version != 'trunk':\",\n \"= dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade)\",\n \"x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions are 'trunk', '2.1', or '2.1.2'\",\n \"__name__ == \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to\",\n \"checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated\",\n \"the database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in\",\n \"checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in the database as %d\\\\033[0m\\\"\",\n \"to checkout version, for example, 2.1 -->') if checkout_version !=\",\n \"= int(raw_input('Revert to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version)\",\n \"project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \\\"__main__\\\": updatelib_inst =\",\n \"#!/usr/bin/env python from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade\",\n \"print \\\"\\\\033[35mVersion Updated in the database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision)\",\n \"dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__\",\n \"from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata',\",\n \"map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions are 'trunk', '2.1', or\",\n \"database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in the\",\n \"except: print \\\"valid versions are 'trunk', '2.1', or '2.1.2' etc\\\"\",\n \"versions are 'trunk', '2.1', or '2.1.2' etc\\\" raise checkout_revision =\",\n \"int(raw_input('Revert to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print\",\n \"python from sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade =\",\n \"if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid\",\n \"checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions\",\n \"dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version\",\n \"to checkout revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion\",\n \"raw_input('Revert to checkout version, for example, 2.1 -->') if checkout_version\",\n \"Updated in the database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision\",\n \"= updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for example,\",\n \"'2.1.2' etc\\\" raise checkout_revision = int(raw_input('Revert to checkout revision, for\",\n \"drop=True) if __name__ == \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version =\",\n \"checkout_revision = int(raw_input('Revert to checkout revision, for example, 16500 -->'))\",\n \"try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions are 'trunk', '2.1',\",\n \"!= 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions are\",\n \"-->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print\",\n \"== \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout\",\n \"\\\"valid versions are 'trunk', '2.1', or '2.1.2' etc\\\" raise checkout_revision\",\n \"revision, for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in\",\n \"\\\"\\\\033[35mVersion Updated in the database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print\",\n \"'2.1', or '2.1.2' etc\\\" raise checkout_revision = int(raw_input('Revert to checkout\",\n \"import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \\\"__main__\\\":\",\n \"\\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version,\",\n \"= raw_input('Revert to checkout version, for example, 2.1 -->') if\",\n \"or '2.1.2' etc\\\" raise checkout_revision = int(raw_input('Revert to checkout revision,\",\n \"in the database %s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated\",\n \"% checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in the database as\",\n \"updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for\",\n \"updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert to checkout version, for example, 2.1\",\n \"example, 2.1 -->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2])\",\n \"%s\\\\033[0m\\\" % checkout_version updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in the database\",\n \"version, for example, 2.1 -->') if checkout_version != 'trunk': try:\",\n \"print \\\"valid versions are 'trunk', '2.1', or '2.1.2' etc\\\" raise\",\n \"example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in the database\",\n \"updatelib_inst.updateDatabaseRevision(checkout_revision) print \\\"\\\\033[35mRevision Updated in the database as %d\\\\033[0m\\\" %\",\n \"raise checkout_revision = int(raw_input('Revert to checkout revision, for example, 16500\",\n \"are 'trunk', '2.1', or '2.1.2' etc\\\" raise checkout_revision = int(raw_input('Revert\",\n \"for example, 16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in the\",\n \"16500 -->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in the database %s\\\\033[0m\\\"\",\n \"-->')) updatelib_inst.updateDatabaseVersion(checkout_version) print \\\"\\\\033[35mVersion Updated in the database %s\\\\033[0m\\\" %\",\n \"updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ == \\\"__main__\\\": updatelib_inst\",\n \"for example, 2.1 -->') if checkout_version != 'trunk': try: map((lambda\",\n \"checkout_version = raw_input('Revert to checkout version, for example, 2.1 -->')\",\n \"2.1 -->') if checkout_version != 'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except:\",\n \"import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if\",\n \"etc\\\" raise checkout_revision = int(raw_input('Revert to checkout revision, for example,\",\n \"'trunk': try: map((lambda x:int(x)),checkout_version.split('.')[:2]) except: print \\\"valid versions are 'trunk',\",\n \"'trunk', '2.1', or '2.1.2' etc\\\" raise checkout_revision = int(raw_input('Revert to\",\n \"dbconfig import updatelib project_dbupgrade = dbupgrade.DBUpgradeTools('projectdata', drop=True) if __name__ ==\",\n \"if __name__ == \\\"__main__\\\": updatelib_inst = updatelib.UpdateLib(project_dbupgrade) checkout_version = raw_input('Revert\"\n]","isConversational":false}}},{"rowIdx":677,"cells":{"ngram":{"kind":"list like","value":["from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y =","from sklearn.linear_model import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from","C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext)","C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear',","fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path)","LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary,","prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y)","save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y","import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg","penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext) save_regression_model(data_set_path,retain_reg,ext=c_ext)","fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def","import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path,","retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext","solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext) save_regression_model(data_set_path,retain_reg,ext=c_ext) save_dataset_predictions(data_set_path,retain_reg,X,ext=c_ext)","regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1',","X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True)","save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg =","= LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext =","from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions","sklearn.linear_model import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression","def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param,","= prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X,","import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import","LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param)","save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression(","prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param):"],"string":"[\n \"from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y =\",\n \"from sklearn.linear_model import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from\",\n \"C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext)\",\n \"C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear',\",\n \"fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path)\",\n \"LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary,\",\n \"prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y)\",\n \"save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y\",\n \"import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg\",\n \"penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext) save_regression_model(data_set_path,retain_reg,ext=c_ext)\",\n \"fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def\",\n \"import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path,\",\n \"retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext\",\n \"solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param) save_regression_summary(data_set_path,retain_reg,ext=c_ext) save_regression_model(data_set_path,retain_reg,ext=c_ext) save_dataset_predictions(data_set_path,retain_reg,X,ext=c_ext)\",\n \"regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1',\",\n \"X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True)\",\n \"save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg =\",\n \"= LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext =\",\n \"from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions\",\n \"sklearn.linear_model import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression\",\n \"def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param,\",\n \"= prepare_data(data_set_path) retain_reg = LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X,\",\n \"import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import\",\n \"LogisticRegression( C=C_param, penalty='l1', solver='liblinear', fit_intercept=True) retain_reg.fit(X, y) c_ext = '_c{:.3f}'.format(C_param)\",\n \"save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y = prepare_data(data_set_path) retain_reg = LogisticRegression(\",\n \"prepare_data, save_regression_model from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param):\"\n]","isConversational":false}}},{"rowIdx":678,"cells":{"ngram":{"kind":"list like","value":["3*6*16)) #%% [markdown] # ## ACP # Réduction de la","l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col)","#%% from IPython.display import Image #%% for cl in range(0,19):","# img = Tools.display_mosaic(image_array, nrow = 10) # fig =","Zoom sur le cluster 0 #%% res_tmp = select_cluster(clustering, 1)","range(0,19): print(\"Cluster %s\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp))","= select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54,","[markdown] # ## Visualisation des clusters #%% def select_cluster(clustering, id_cluster):","IPython.display import Image #%% for cl in range(0,19): print(\"Cluster %s\"","pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%%","Sauvegarde des modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical'])","#%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown]","autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss","[Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp[:100]] #","= select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size =","= Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496, nb_batch","## Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/',","#%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown]","## ACP # Réduction de la dimension #%% clustering =","image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for f in","0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] # ###","Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] #","mode = os.getenv(\"MODE\")) #%% [markdown] # ## Chargement du modèle","= (54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow","# ## ACP # Réduction de la dimension #%% clustering","#%% [markdown] # ## Chargement de la config #%% load_dotenv(find_dotenv())","modèle #%% ## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name']","os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] #","(54, 96)) for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array,","Tools.display_mosaic(image_array, nrow = 18) #%% col = [1 if l","# ## Chargement de la config #%% load_dotenv(find_dotenv()) cfg =","[markdown] # ## Chargement du modèle #%% ## charger le","= Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch","[markdown] # ## Résultats #%% [markdown] # ### Clusters intermediaires","cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for","= 10) # fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img,","== id_cluster] #%% from IPython.display import Image #%% for cl","ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans","96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18)","#%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,","#%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0],","# ### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7))","= 'auto') # plt.show() #%% [markdown] # ## Zoom sur","c = clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%%","= (27, 48), batch = 496, nb_batch = 10) #%%","# Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id,","figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') # plt.show() #%%","pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16))","[markdown] # ## import classique import os #%% %load_ext autoreload","col = [1 if l == 1 else 0 for","img = Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1,","from iss.tools import Config from iss.tools import Tools from iss.models","intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP #","os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%% [markdown] # ## Chargement du","= Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1, figsize=(12,","plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown] # ##","CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] #","from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv","= clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:,","# ## Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return","clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown] #","Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%% [markdown] # ##","= 18) #%% col = [1 if l == 1","clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers clusters #%%","config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode =","[markdown] # ### Sauvegarde des modèles #%% clustering.save() #%% #","(cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size","dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%%","== 1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0],","Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0]","#%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f,","import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering","= SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%% filenames","Config from iss.tools import Tools from iss.models import SimpleConvAutoEncoder from","SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from","Résultats #%% [markdown] # ### Clusters intermediaires #%% fig =","Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0],","clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH #","= ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des clusters","de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"),","# ## import classique import os #%% %load_ext autoreload %autoreload","96)) for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow","figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%%","sur le cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%%","Clustering classique #%% [markdown] # ## import classique import os","'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images","from IPython.display import Image #%% for cl in range(0,19): print(\"Cluster","0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array =","os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%% from iss.tools","iss #%% from iss.tools import Config from iss.tools import Tools","in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col =","1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96))","= 496, nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs,","clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] # ### Sauvegarde","ACP # Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'],","0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c","(27, 48), batch = 496, nb_batch = 10) #%% pictures_id,","res[2] == id_cluster] #%% from IPython.display import Image #%% for","select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96))","from iss.tools import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering","clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] # ### Clusters","classique import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%%","for res in clustering.get_zip_results() if res[2] == id_cluster] #%% from","filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48),","# fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect =","= 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown]","Import iss #%% from iss.tools import Config from iss.tools import","in range(0,19): print(\"Cluster %s\" % (cl)) res_tmp = select_cluster(clustering, cl)","[Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp] #%%","import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown]","model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%%","#%% [markdown] # ## Visualisation des clusters #%% def select_cluster(clustering,","%load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import","clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds clusters","1], c = clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux","Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496, nb_batch =","Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] #","## Résultats #%% [markdown] # ### Clusters intermediaires #%% fig","# Clustering classique #%% [markdown] # ## import classique import","## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown]","#%% [markdown] # ## Import iss #%% from iss.tools import","print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for f","target_size = (54, 96)) for f in res_tmp[:100]] # img","[1 if l == 1 else 0 for l in","clustering.load() #%% [markdown] # ## Visualisation des clusters #%% def","ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv","#%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown]","charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model","## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown]","res[0] + '.jpg') for res in clustering.get_zip_results() if res[2] ==","cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%% [markdown]","from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering","du modèle #%% ## charger le modèle model_type = 'simple_conv'","= clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles #%%","select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in","np #%% [markdown] # ## Chargement de la config #%%","target_size = (54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array,","des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size","cl in range(0,19): print(\"Cluster %s\" % (cl)) res_tmp = select_cluster(clustering,","import find_dotenv, load_dotenv import numpy as np #%% [markdown] #","### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:,","#%% [markdown] # ## Résultats #%% [markdown] # ### Clusters","import numpy as np #%% [markdown] # ## Chargement de","classique #%% [markdown] # ## import classique import os #%%","48), batch = 496, nb_batch = 10) #%% pictures_id, pictures_preds","### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =","iss.tools import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import","'auto') # plt.show() #%% [markdown] # ## Zoom sur le","## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs =","= pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP # Réduction","model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ##","for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow =","= [Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp]","# ## Résultats #%% [markdown] # ### Clusters intermediaires #%%","#%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\"))","model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%%","clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] #","plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') # plt.show()","# ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c","print(\"Cluster %s\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array","[markdown] # ## Chargement de la config #%% load_dotenv(find_dotenv()) cfg","0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] # ###","id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results()","clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] #","# ## Zoom sur le cluster 0 #%% res_tmp =","res in clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display","plt.show() #%% [markdown] # ## Zoom sur le cluster 0","la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode","#%% from iss.tools import Config from iss.tools import Tools from","clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des","#%% [markdown] # ### Sauvegarde des modèles #%% clustering.save() #%%","fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto')","= [1 if l == 1 else 0 for l","1], c = clustering.final_labels) #%% [markdown] # ### Sauvegarde des","load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%%","plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) ==","= Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown]","### Sauvegarde des modèles #%% clustering.save() #%% # clustering =","#%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP","plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] #","l == 1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:,","(54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow =","intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers","#%% [markdown] # ## Chargement du modèle #%% ## charger","% (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f,","for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%%","for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =","[markdown] ## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs","#%% [markdown] # ## Kmeans # Premiers clusters #%% clustering.compute_kmeans()","from dotenv import find_dotenv, load_dotenv import numpy as np #%%","os.getenv(\"MODE\")) #%% [markdown] # ## Chargement du modèle #%% ##","clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%%","iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy","AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy as np","#%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown]","# Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ##","= col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) == 1,","select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54,","import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering","fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c","finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%%","= os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%% [markdown] # ## Chargement","for cl in range(0,19): print(\"Cluster %s\" % (cl)) res_tmp =","des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] +","## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab'","SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%% filenames =","clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg')","res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col = [1","cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array","#%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for","#%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for","Image #%% for cl in range(0,19): print(\"Cluster %s\" % (cl))","# ## Chargement du modèle #%% ## charger le modèle","batch = 496, nb_batch = 10) #%% pictures_id, pictures_preds =","[markdown] # ## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers()","clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0],","'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ##","if l == 1 else 0 for l in clustering.kmeans_labels]","Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels)","Chargement du modèle #%% ## charger le modèle model_type =","= Config(project_dir = os.getenv(\"PROJECT_DIR\"), mode = os.getenv(\"MODE\")) #%% [markdown] #","#%% clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers clusters","id_cluster] #%% from IPython.display import Image #%% for cl in","= (54, 96)) for f in res_tmp[:100]] # img =","#%% for cl in range(0,19): print(\"Cluster %s\" % (cl)) res_tmp","[markdown] # # Clustering classique #%% [markdown] # ## import","from iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv import","f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col","target_size = (27, 48), batch = 496, nb_batch = 10)","in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10) #","c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) ==","[markdown] # ## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels()","as np #%% [markdown] # ## Chargement de la config","[os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results() if res[2]","clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles #%% clustering.save()","if res[2] == id_cluster] #%% from IPython.display import Image #%%","import AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy as","+ '.jpg') for res in clustering.get_zip_results() if res[2] == id_cluster]","[markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],","plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels)","des modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load()","find_dotenv, load_dotenv import numpy as np #%% [markdown] # ##","## Chargement du modèle #%% ## charger le modèle model_type","1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,","# ## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%%","clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown] # ###","res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size =","#%% [markdown] # ## Zoom sur le cluster 0 #%%","# plt.show() #%% [markdown] # ## Zoom sur le cluster","Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output)","col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) == 1, 1])","10) # fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect","clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col)","#%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27,","# ### Sauvegarde des modèles #%% clustering.save() #%% # clustering","load_dotenv import numpy as np #%% [markdown] # ## Chargement","#%% [markdown] # ### Clusters intermediaires #%% fig = plt.figure(1,","#%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ##","Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from","#%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ##","10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output =","clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0],","in clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display import","res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size","#%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],","2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%% from","## Zoom sur le cluster 0 #%% res_tmp = select_cluster(clustering,","pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP # Réduction de","in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%%","= [Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp[:100]]","clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display import Image","# Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ##","#%% [markdown] # ## CAH # Seconds clusters #%% clustering.compute_cah()","## Chargement de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir","= 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des","# clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation","Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats","#%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds","## Import iss #%% from iss.tools import Config from iss.tools","#%% [markdown] ## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/')","import Config from iss.tools import Tools from iss.models import SimpleConvAutoEncoder","iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import","#%% ## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] =","= ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ##","# plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown] #","= plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') #","aspect = 'auto') # plt.show() #%% [markdown] # ## Zoom","0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1,","de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%%","clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds clusters #%%","res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10) # fig","[markdown] # ### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12,","le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model =","def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res","#%% col = [1 if l == 1 else 0","nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%%","7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown]","f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10)","# ## Import iss #%% from iss.tools import Config from","c = clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles","Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1, figsize=(12, 7))","images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size =","intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,","nrow = 10) # fig = plt.figure(1, figsize=(12, 7)) #","[markdown] # ## ACP # Réduction de la dimension #%%","= plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =","Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames,","import classique import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work')","pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans #","1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col)","18) #%% col = [1 if l == 1 else","= os.getenv(\"MODE\")) #%% [markdown] # ## Chargement du modèle #%%","Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH","modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%%","7)) # plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown]","## import classique import os #%% %load_ext autoreload %autoreload 2","return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results() if","#%% [markdown] # # Clustering classique #%% [markdown] # ##","#%% Tools.display_mosaic(image_array, nrow = 18) #%% col = [1 if","import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import find_dotenv,","Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch =","ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des clusters #%%","import Image #%% for cl in range(0,19): print(\"Cluster %s\" %","else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],","iss.tools import Config from iss.tools import Tools from iss.models import","dotenv import find_dotenv, load_dotenv import numpy as np #%% [markdown]","cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement","#%% [markdown] # ## import classique import os #%% %load_ext","%autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%%","# ## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%%","[markdown] # ## Zoom sur le cluster 0 #%% res_tmp","iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import","la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca()","'.jpg') for res in clustering.get_zip_results() if res[2] == id_cluster] #%%","# # Clustering classique #%% [markdown] # ## import classique","Chargement de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir =","%s\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array =","modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type])","[markdown] # ## Import iss #%% from iss.tools import Config","plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] #","le cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp))","#%% [markdown] # ## ACP # Réduction de la dimension","generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496,","nrow = 18) #%% col = [1 if l ==","numpy as np #%% [markdown] # ## Chargement de la","496, nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model)","= 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output"],"string":"[\n \"3*6*16)) #%% [markdown] # ## ACP # Réduction de la\",\n \"l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col)\",\n \"#%% from IPython.display import Image #%% for cl in range(0,19):\",\n \"# img = Tools.display_mosaic(image_array, nrow = 10) # fig =\",\n \"Zoom sur le cluster 0 #%% res_tmp = select_cluster(clustering, 1)\",\n \"range(0,19): print(\\\"Cluster %s\\\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp))\",\n \"= select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54,\",\n \"[markdown] # ## Visualisation des clusters #%% def select_cluster(clustering, id_cluster):\",\n \"IPython.display import Image #%% for cl in range(0,19): print(\\\"Cluster %s\\\"\",\n \"pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%%\",\n \"Sauvegarde des modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical'])\",\n \"#%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown]\",\n \"autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss\",\n \"[Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp[:100]] #\",\n \"= select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size =\",\n \"= Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496, nb_batch\",\n \"## Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/',\",\n \"#%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown]\",\n \"## ACP # Réduction de la dimension #%% clustering =\",\n \"image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for f in\",\n \"0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] # ###\",\n \"Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] #\",\n \"mode = os.getenv(\\\"MODE\\\")) #%% [markdown] # ## Chargement du modèle\",\n \"= (54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow\",\n \"# ## ACP # Réduction de la dimension #%% clustering\",\n \"#%% [markdown] # ## Chargement de la config #%% load_dotenv(find_dotenv())\",\n \"modèle #%% ## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name']\",\n \"os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] #\",\n \"(54, 96)) for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array,\",\n \"Tools.display_mosaic(image_array, nrow = 18) #%% col = [1 if l\",\n \"# ## Chargement de la config #%% load_dotenv(find_dotenv()) cfg =\",\n \"[markdown] # ## Chargement du modèle #%% ## charger le\",\n \"= Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch\",\n \"[markdown] # ## Résultats #%% [markdown] # ### Clusters intermediaires\",\n \"cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for\",\n \"= 10) # fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img,\",\n \"== id_cluster] #%% from IPython.display import Image #%% for cl\",\n \"ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans\",\n \"96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18)\",\n \"#%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,\",\n \"#%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0],\",\n \"# ### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7))\",\n \"= 'auto') # plt.show() #%% [markdown] # ## Zoom sur\",\n \"c = clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%%\",\n \"= (27, 48), batch = 496, nb_batch = 10) #%%\",\n \"# Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id,\",\n \"figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') # plt.show() #%%\",\n \"pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16))\",\n \"[markdown] # ## import classique import os #%% %load_ext autoreload\",\n \"col = [1 if l == 1 else 0 for\",\n \"img = Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1,\",\n \"from iss.tools import Config from iss.tools import Tools from iss.models\",\n \"intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP #\",\n \"os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%% [markdown] # ## Chargement du\",\n \"= Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1, figsize=(12,\",\n \"plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown] # ##\",\n \"CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] #\",\n \"from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv\",\n \"= clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:,\",\n \"# ## Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return\",\n \"clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown] #\",\n \"Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%% [markdown] # ##\",\n \"= 18) #%% col = [1 if l == 1\",\n \"clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers clusters #%%\",\n \"config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode =\",\n \"[markdown] # ### Sauvegarde des modèles #%% clustering.save() #%% #\",\n \"(cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size\",\n \"dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%%\",\n \"== 1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0],\",\n \"Visualisation des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0]\",\n \"#%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f,\",\n \"import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering\",\n \"= SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%% filenames\",\n \"Config from iss.tools import Tools from iss.models import SimpleConvAutoEncoder from\",\n \"SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from\",\n \"Résultats #%% [markdown] # ### Clusters intermediaires #%% fig =\",\n \"Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0],\",\n \"clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH #\",\n \"= ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des clusters\",\n \"de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"),\",\n \"# ## import classique import os #%% %load_ext autoreload %autoreload\",\n \"96)) for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow\",\n \"figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%%\",\n \"sur le cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%%\",\n \"Clustering classique #%% [markdown] # ## import classique import os\",\n \"'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images\",\n \"from IPython.display import Image #%% for cl in range(0,19): print(\\\"Cluster\",\n \"0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array =\",\n \"os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%% from iss.tools\",\n \"iss #%% from iss.tools import Config from iss.tools import Tools\",\n \"in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col =\",\n \"1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96))\",\n \"= 496, nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs,\",\n \"clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] # ### Sauvegarde\",\n \"ACP # Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'],\",\n \"0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c\",\n \"(27, 48), batch = 496, nb_batch = 10) #%% pictures_id,\",\n \"res[2] == id_cluster] #%% from IPython.display import Image #%% for\",\n \"select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96))\",\n \"from iss.tools import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering\",\n \"clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] # ### Clusters\",\n \"classique import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%%\",\n \"for res in clustering.get_zip_results() if res[2] == id_cluster] #%% from\",\n \"filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48),\",\n \"# fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect =\",\n \"= 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown]\",\n \"Import iss #%% from iss.tools import Config from iss.tools import\",\n \"in range(0,19): print(\\\"Cluster %s\\\" % (cl)) res_tmp = select_cluster(clustering, cl)\",\n \"[Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp] #%%\",\n \"import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown]\",\n \"model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%%\",\n \"#%% [markdown] # ## Visualisation des clusters #%% def select_cluster(clustering,\",\n \"%load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import\",\n \"clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds clusters\",\n \"1], c = clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux\",\n \"Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496, nb_batch =\",\n \"Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] #\",\n \"## Résultats #%% [markdown] # ### Clusters intermediaires #%% fig\",\n \"# Clustering classique #%% [markdown] # ## import classique import\",\n \"## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown]\",\n \"#%% [markdown] # ## Import iss #%% from iss.tools import\",\n \"print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for f\",\n \"target_size = (54, 96)) for f in res_tmp[:100]] # img\",\n \"[1 if l == 1 else 0 for l in\",\n \"clustering.load() #%% [markdown] # ## Visualisation des clusters #%% def\",\n \"ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv\",\n \"#%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown]\",\n \"charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model\",\n \"## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown]\",\n \"res[0] + '.jpg') for res in clustering.get_zip_results() if res[2] ==\",\n \"cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%% [markdown]\",\n \"from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering\",\n \"du modèle #%% ## charger le modèle model_type = 'simple_conv'\",\n \"= clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles #%%\",\n \"select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in\",\n \"np #%% [markdown] # ## Chargement de la config #%%\",\n \"target_size = (54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array,\",\n \"des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size\",\n \"cl in range(0,19): print(\\\"Cluster %s\\\" % (cl)) res_tmp = select_cluster(clustering,\",\n \"import find_dotenv, load_dotenv import numpy as np #%% [markdown] #\",\n \"### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:,\",\n \"#%% [markdown] # ## Résultats #%% [markdown] # ### Clusters\",\n \"import numpy as np #%% [markdown] # ## Chargement de\",\n \"classique #%% [markdown] # ## import classique import os #%%\",\n \"48), batch = 496, nb_batch = 10) #%% pictures_id, pictures_preds\",\n \"### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =\",\n \"iss.tools import Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import\",\n \"'auto') # plt.show() #%% [markdown] # ## Zoom sur le\",\n \"## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs =\",\n \"= pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP # Réduction\",\n \"model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ##\",\n \"for f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow =\",\n \"= [Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp]\",\n \"# ## Résultats #%% [markdown] # ### Clusters intermediaires #%%\",\n \"#%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\"))\",\n \"model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%%\",\n \"clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] #\",\n \"plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') # plt.show()\",\n \"# ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c\",\n \"print(\\\"Cluster %s\\\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array\",\n \"[markdown] # ## Chargement de la config #%% load_dotenv(find_dotenv()) cfg\",\n \"0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] # ###\",\n \"id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results()\",\n \"clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] #\",\n \"# ## Zoom sur le cluster 0 #%% res_tmp =\",\n \"res in clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display\",\n \"plt.show() #%% [markdown] # ## Zoom sur le cluster 0\",\n \"la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode\",\n \"#%% from iss.tools import Config from iss.tools import Tools from\",\n \"clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des\",\n \"#%% [markdown] # ### Sauvegarde des modèles #%% clustering.save() #%%\",\n \"fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto')\",\n \"= [1 if l == 1 else 0 for l\",\n \"1], c = clustering.final_labels) #%% [markdown] # ### Sauvegarde des\",\n \"load_dotenv(find_dotenv()) cfg = Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%%\",\n \"plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) ==\",\n \"= Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown]\",\n \"### Sauvegarde des modèles #%% clustering.save() #%% # clustering =\",\n \"#%% intermediate_output = pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP\",\n \"plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown] #\",\n \"l == 1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:,\",\n \"(54, 96)) for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow =\",\n \"intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers\",\n \"#%% [markdown] # ## Chargement du modèle #%% ## charger\",\n \"% (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f,\",\n \"for f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%%\",\n \"for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =\",\n \"[markdown] ## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs\",\n \"#%% [markdown] # ## Kmeans # Premiers clusters #%% clustering.compute_kmeans()\",\n \"from dotenv import find_dotenv, load_dotenv import numpy as np #%%\",\n \"os.getenv(\\\"MODE\\\")) #%% [markdown] # ## Chargement du modèle #%% ##\",\n \"clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%%\",\n \"iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy\",\n \"AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy as np\",\n \"#%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown]\",\n \"# Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ##\",\n \"= col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) == 1,\",\n \"select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54,\",\n \"import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering\",\n \"fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c\",\n \"finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%%\",\n \"= os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%% [markdown] # ## Chargement\",\n \"for cl in range(0,19): print(\\\"Cluster %s\\\" % (cl)) res_tmp =\",\n \"des clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] +\",\n \"## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab'\",\n \"SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des images #%% filenames =\",\n \"clusters #%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg')\",\n \"res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col = [1\",\n \"cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array\",\n \"#%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size = (54, 96)) for\",\n \"#%% def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for\",\n \"Image #%% for cl in range(0,19): print(\\\"Cluster %s\\\" % (cl))\",\n \"# ## Chargement du modèle #%% ## charger le modèle\",\n \"batch = 496, nb_batch = 10) #%% pictures_id, pictures_preds =\",\n \"[markdown] # ## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers()\",\n \"clustering.kmeans_labels) #%% [markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0],\",\n \"'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ##\",\n \"if l == 1 else 0 for l in clustering.kmeans_labels]\",\n \"Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels)\",\n \"Chargement du modèle #%% ## charger le modèle model_type =\",\n \"= Config(project_dir = os.getenv(\\\"PROJECT_DIR\\\"), mode = os.getenv(\\\"MODE\\\")) #%% [markdown] #\",\n \"#%% clustering.compute_pca() #%% [markdown] # ## Kmeans # Premiers clusters\",\n \"id_cluster] #%% from IPython.display import Image #%% for cl in\",\n \"= (54, 96)) for f in res_tmp[:100]] # img =\",\n \"#%% for cl in range(0,19): print(\\\"Cluster %s\\\" % (cl)) res_tmp\",\n \"[markdown] # # Clustering classique #%% [markdown] # ## import\",\n \"from iss.clustering import AdvancedClustering from dotenv import find_dotenv, load_dotenv import\",\n \"f in res_tmp] #%% Tools.display_mosaic(image_array, nrow = 18) #%% col\",\n \"target_size = (27, 48), batch = 496, nb_batch = 10)\",\n \"in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10) #\",\n \"c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) ==\",\n \"[markdown] # ## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels()\",\n \"as np #%% [markdown] # ## Chargement de la config\",\n \"[os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results() if res[2]\",\n \"clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles #%% clustering.save()\",\n \"if res[2] == id_cluster] #%% from IPython.display import Image #%%\",\n \"import AdvancedClustering from dotenv import find_dotenv, load_dotenv import numpy as\",\n \"+ '.jpg') for res in clustering.get_zip_results() if res[2] == id_cluster]\",\n \"[markdown] # ### Clusters finaux #%% plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],\",\n \"plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels)\",\n \"des modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load()\",\n \"find_dotenv, load_dotenv import numpy as np #%% [markdown] # ##\",\n \"## Chargement du modèle #%% ## charger le modèle model_type\",\n \"1 else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,\",\n \"# ## Kmeans # Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%%\",\n \"clustering.compute_cah_labels() #%% [markdown] # ## Résultats #%% [markdown] # ###\",\n \"res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size =\",\n \"#%% [markdown] # ## Zoom sur le cluster 0 #%%\",\n \"# plt.show() #%% [markdown] # ## Zoom sur le cluster\",\n \"Réduction de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output)\",\n \"col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col) == 1, 1])\",\n \"10) # fig = plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect\",\n \"clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col)\",\n \"#%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27,\",\n \"# ### Sauvegarde des modèles #%% clustering.save() #%% # clustering\",\n \"load_dotenv import numpy as np #%% [markdown] # ## Chargement\",\n \"#%% [markdown] # ### Clusters intermediaires #%% fig = plt.figure(1,\",\n \"#%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ##\",\n \"Tools from iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from\",\n \"#%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ##\",\n \"10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output =\",\n \"clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0],\",\n \"in clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display import\",\n \"res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp)) image_array = [Tools.read_np_picture(f, target_size\",\n \"#%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],\",\n \"2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%% from\",\n \"## Zoom sur le cluster 0 #%% res_tmp = select_cluster(clustering,\",\n \"pictures_preds.reshape((pictures_preds.shape[0], 3*6*16)) #%% [markdown] # ## ACP # Réduction de\",\n \"in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col) #%%\",\n \"= [Tools.read_np_picture(f, target_size = (54, 96)) for f in res_tmp[:100]]\",\n \"clustering.get_zip_results() if res[2] == id_cluster] #%% from IPython.display import Image\",\n \"# Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ##\",\n \"#%% [markdown] # ## CAH # Seconds clusters #%% clustering.compute_cah()\",\n \"## Chargement de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir\",\n \"= 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement des\",\n \"# clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation\",\n \"Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%% [markdown] # ## Résultats\",\n \"#%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds\",\n \"## Import iss #%% from iss.tools import Config from iss.tools\",\n \"#%% [markdown] ## Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/')\",\n \"import Config from iss.tools import Tools from iss.models import SimpleConvAutoEncoder\",\n \"iss.models import SimpleConvAutoEncoder from iss.clustering import ClassicalClustering from iss.clustering import\",\n \"#%% ## charger le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] =\",\n \"= ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ##\",\n \"# plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown] #\",\n \"= plt.figure(1, figsize=(12, 7)) # plt.imshow(img, aspect = 'auto') #\",\n \"aspect = 'auto') # plt.show() #%% [markdown] # ## Zoom\",\n \"0], clustering.pca_reduction[:, 1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1,\",\n \"de la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%%\",\n \"clustering.compute_kmeans_centers() #%% [markdown] # ## CAH # Seconds clusters #%%\",\n \"res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10) # fig\",\n \"[markdown] # ### Clusters intermediaires #%% fig = plt.figure(1, figsize=(12,\",\n \"le modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model =\",\n \"def select_cluster(clustering, id_cluster): return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res\",\n \"#%% col = [1 if l == 1 else 0\",\n \"nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%%\",\n \"7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.kmeans_labels) #%% [markdown]\",\n \"f in res_tmp[:100]] # img = Tools.display_mosaic(image_array, nrow = 10)\",\n \"# ## Import iss #%% from iss.tools import Config from\",\n \"c = clustering.final_labels) #%% [markdown] # ### Sauvegarde des modèles\",\n \"Tools.display_mosaic(image_array, nrow = 10) # fig = plt.figure(1, figsize=(12, 7))\",\n \"images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size =\",\n \"intermediaires #%% fig = plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:,\",\n \"nrow = 10) # fig = plt.figure(1, figsize=(12, 7)) #\",\n \"[markdown] # ## ACP # Réduction de la dimension #%%\",\n \"= plt.figure(1, figsize=(12, 7)) plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c =\",\n \"Chargement des images #%% filenames = Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames,\",\n \"import classique import os #%% %load_ext autoreload %autoreload 2 os.chdir('/home/jovyan/work')\",\n \"pictures_id, intermediate_output) #%% clustering.compute_pca() #%% [markdown] # ## Kmeans #\",\n \"1], c = col) #%% plt.scatter(clustering.pca_reduction[np.array(col) == 1, 0], clustering.pca_reduction[np.array(col)\",\n \"18) #%% col = [1 if l == 1 else\",\n \"= os.getenv(\\\"MODE\\\")) #%% [markdown] # ## Chargement du modèle #%%\",\n \"Premiers clusters #%% clustering.compute_kmeans() clustering.compute_kmeans_centers() #%% [markdown] # ## CAH\",\n \"modèles #%% clustering.save() #%% # clustering = ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%%\",\n \"7)) # plt.imshow(img, aspect = 'auto') # plt.show() #%% [markdown]\",\n \"## import classique import os #%% %load_ext autoreload %autoreload 2\",\n \"return [os.path.join('data/processed/models/autoencoder/train/k/', res[0] + '.jpg') for res in clustering.get_zip_results() if\",\n \"#%% [markdown] # # Clustering classique #%% [markdown] # ##\",\n \"#%% Tools.display_mosaic(image_array, nrow = 18) #%% col = [1 if\",\n \"import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import find_dotenv,\",\n \"Tools.list_directory_filenames('data/processed/models/autoencoder/train/k/') generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch =\",\n \"ClassicalClustering(cfg.get('clustering')['classical']) clustering.load() #%% [markdown] # ## Visualisation des clusters #%%\",\n \"import Image #%% for cl in range(0,19): print(\\\"Cluster %s\\\" %\",\n \"else 0 for l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1],\",\n \"iss.tools import Config from iss.tools import Tools from iss.models import\",\n \"dotenv import find_dotenv, load_dotenv import numpy as np #%% [markdown]\",\n \"cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type]) #%% [markdown] ## Chargement\",\n \"#%% [markdown] # ## import classique import os #%% %load_ext\",\n \"%autoreload 2 os.chdir('/home/jovyan/work') #%% [markdown] # ## Import iss #%%\",\n \"# ## CAH # Seconds clusters #%% clustering.compute_cah() clustering.compute_cah_labels() #%%\",\n \"[markdown] # ## Zoom sur le cluster 0 #%% res_tmp\",\n \"iss.clustering import ClassicalClustering from iss.clustering import AdvancedClustering from dotenv import\",\n \"la dimension #%% clustering = ClassicalClustering(cfg.get('clustering')['classical'], pictures_id, intermediate_output) #%% clustering.compute_pca()\",\n \"'.jpg') for res in clustering.get_zip_results() if res[2] == id_cluster] #%%\",\n \"# # Clustering classique #%% [markdown] # ## import classique\",\n \"Chargement de la config #%% load_dotenv(find_dotenv()) cfg = Config(project_dir =\",\n \"%s\\\" % (cl)) res_tmp = select_cluster(clustering, cl) print(len(res_tmp)) image_array =\",\n \"modèle model_type = 'simple_conv' cfg.get('models')[model_type]['model_name'] = 'model_colab' model = SimpleConvAutoEncoder(cfg.get('models')[model_type])\",\n \"[markdown] # ## Import iss #%% from iss.tools import Config\",\n \"plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = clustering.final_labels) #%% [markdown] #\",\n \"le cluster 0 #%% res_tmp = select_cluster(clustering, 1) #%% print(len(res_tmp))\",\n \"#%% [markdown] # ## ACP # Réduction de la dimension\",\n \"generator_imgs = Tools.generator_np_picture_from_filenames(filenames, target_size = (27, 48), batch = 496,\",\n \"nrow = 18) #%% col = [1 if l ==\",\n \"numpy as np #%% [markdown] # ## Chargement de la\",\n \"496, nb_batch = 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model)\",\n \"= 10) #%% pictures_id, pictures_preds = Tools.encoded_pictures_from_generator(generator_imgs, model) #%% intermediate_output\"\n]","isConversational":false}}},{"rowIdx":679,"cells":{"ngram":{"kind":"list like","value":["handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) #","'%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s '","= logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create","__author__ = \"<NAME>\" __email__ = \"<EMAIL>\" import errno import logging","%(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info)","location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related to","'/info.log' error_log_file = log_folder + '/error.log' # check if exists","= log_folder + '/info.log' error_log_file = log_folder + '/error.log' #","datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s ' '[","%H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s","import os from logging.config import fileConfig # runtime location this_folder","log_folder + '/error.log' # check if exists or create log","'[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error =","create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError:","from logging.config import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__))","# set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add","and add to handlers format_info = \\ logging.Formatter('%(asctime)s %(levelname)s '","logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger","add to handlers format_info = \\ logging.Formatter('%(asctime)s %(levelname)s ' '[","levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to handlers","ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file))","' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ]","def get_logger(name: str = __name__): logger = logging.getLogger(name) # add","= logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if","files ini_file = 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file","log_folder + '/info.log' error_log_file = log_folder + '/error.log' # check","+ '/info.log' error_log_file = log_folder + '/error.log' # check if","] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s","folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder)","' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name:","if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise #","handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger = logging.getLogger(name) #","----------------------------------------------------------- __author__ = \"<NAME>\" __email__ = \"<EMAIL>\" import errno import","= os.path.dirname(os.path.abspath(__file__)) # define log folder related to location log_folder","linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error)","'debug.ini' info_log_file = log_folder + '/info.log' error_log_file = log_folder +","'/error.log' # check if exists or create log folder try:","linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter(","utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>,","= \\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ]","config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info","# runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder","'%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread:","' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error","-*- coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke","\"<EMAIL>\" import errno import logging import os from logging.config import","= log_folder + '/error.log' # check if exists or create","= logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels","# -*- coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021,","info_log_file = log_folder + '/info.log' error_log_file = log_folder + '/error.log'","else: raise # setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file,","ini and log files ini_file = 'debug.ini' info_log_file = log_folder","check if exists or create log folder try: os.makedirs(log_folder, exist_ok=True)","%(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S')","created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # -----------------------------------------------------------","os.makedirs(log_folder) except OSError as exc: # Python >2.5 if exc.errno","create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file))","# create formatters and add to handlers format_info = \\","<NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \"<NAME>\" __email__","# ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL> #","exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup",">2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise","except TypeError: try: os.makedirs(log_folder) except OSError as exc: # Python","ini_file = 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file =","error_log_file = log_folder + '/error.log' # check if exists or","https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \"<NAME>\" __email__ = \"<EMAIL>\" import","define ini and log files ini_file = 'debug.ini' info_log_file =","= \\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] '","location log_folder = os.path.join(this_folder, '../logs') # define ini and log","fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log","os.path.join(this_folder, '../logs') # define ini and log files ini_file =","configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers","= \"<EMAIL>\" import errno import logging import os from logging.config","add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ == '__main__':","this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related to location","handlers format_info = \\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s","logging.config import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) #","logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters","log folder related to location log_folder = os.path.join(this_folder, '../logs') #","%(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger =","handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ == '__main__': pass","# add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ ==","# create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder,","as exc: # Python >2.5 if exc.errno == errno.EEXIST and","= 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file = log_folder","] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def","os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder,","%(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s","format_info = \\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ]","import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define","coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke #","# define ini and log files ini_file = 'debug.ini' info_log_file","folder related to location log_folder = os.path.join(this_folder, '../logs') # define","exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as","== errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup configuration","%(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__):","import errno import logging import os from logging.config import fileConfig","exists or create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2","import logging import os from logging.config import fileConfig # runtime","exc: # Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder):","os.path.dirname(os.path.abspath(__file__)) # define log folder related to location log_folder =","and os.path.isdir(log_folder): pass else: raise # setup configuration config_file =","' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s","try: os.makedirs(log_folder) except OSError as exc: # Python >2.5 if","and log files ini_file = 'debug.ini' info_log_file = log_folder +","try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except","%(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\","\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] '","__name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return","# Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as exc:","runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related","TypeError: try: os.makedirs(log_folder) except OSError as exc: # Python >2.5","raise # setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True)","fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error","log_folder = os.path.join(this_folder, '../logs') # define ini and log files","logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d","if exists or create log folder try: os.makedirs(log_folder, exist_ok=True) #","02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__","Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as exc: #","setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create","errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup configuration config_file","# Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass","create formatters and add to handlers format_info = \\ logging.Formatter('%(asctime)s","= os.path.join(this_folder, '../logs') # define ini and log files ini_file","__email__ = \"<EMAIL>\" import errno import logging import os from","logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__","# setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) #","to location log_folder = os.path.join(this_folder, '../logs') # define ini and","= \"<NAME>\" __email__ = \"<EMAIL>\" import errno import logging import","'../logs') # define ini and log files ini_file = 'debug.ini'","format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s","= __name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error)","#!/usr/bin/python3 # -*- coding: utf-8 -*- # ----------------------------------------------------------- # created","<EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \"<NAME>\" __email__ =","tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ =","# ----------------------------------------------------------- __author__ = \"<NAME>\" __email__ = \"<EMAIL>\" import errno","\"<NAME>\" __email__ = \"<EMAIL>\" import errno import logging import os","# https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \"<NAME>\" __email__ = \"<EMAIL>\"","----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh","'[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str","to handlers format_info = \\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s","# check if exists or create log folder try: os.makedirs(log_folder,","-*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL>","# define log folder related to location log_folder = os.path.join(this_folder,","os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError","errno import logging import os from logging.config import fileConfig #","# <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \"<NAME>\"","= os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info =","pass else: raise # setup configuration config_file = os.path.join(this_folder, ini_file)","handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to handlers format_info","log files ini_file = 'debug.ini' info_log_file = log_folder + '/info.log'","define log folder related to location log_folder = os.path.join(this_folder, '../logs')","info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR)","+ '/error.log' # check if exists or create log folder","logging import os from logging.config import fileConfig # runtime location","thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str =","os from logging.config import fileConfig # runtime location this_folder =","formatters and add to handlers format_info = \\ logging.Formatter('%(asctime)s %(levelname)s","] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger","logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[","logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO)","set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to","except OSError as exc: # Python >2.5 if exc.errno ==","Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else:","related to location log_folder = os.path.join(this_folder, '../logs') # define ini","disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error =","OSError as exc: # Python >2.5 if exc.errno == errno.EEXIST","or create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except","os.path.isdir(log_folder): pass else: raise # setup configuration config_file = os.path.join(this_folder,","str = __name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info)","error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and","'[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s')","get_logger(name: str = __name__): logger = logging.getLogger(name) # add handler","# created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh #","log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try:","handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set","handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) #","\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s',","handler_error.setLevel(logging.ERROR) # create formatters and add to handlers format_info =","handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger = logging.getLogger(name)"],"string":"[\n \"handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) #\",\n \"'%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\ logging.Formatter( '%(asctime)s %(levelname)s '\",\n \"= logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create\",\n \"__author__ = \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import errno import logging\",\n \"%(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info)\",\n \"location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related to\",\n \"'/info.log' error_log_file = log_folder + '/error.log' # check if exists\",\n \"= log_folder + '/info.log' error_log_file = log_folder + '/error.log' #\",\n \"datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[\",\n \"%H:%M:%S') format_error = \\\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s\",\n \"import os from logging.config import fileConfig # runtime location this_folder\",\n \"log_folder + '/error.log' # check if exists or create log\",\n \"'[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error =\",\n \"create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError:\",\n \"from logging.config import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__))\",\n \"# set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add\",\n \"and add to handlers format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s '\",\n \"logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger\",\n \"add to handlers format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[\",\n \"levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to handlers\",\n \"ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file))\",\n \"' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ]\",\n \"def get_logger(name: str = __name__): logger = logging.getLogger(name) # add\",\n \"= logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if\",\n \"files ini_file = 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file\",\n \"log_folder + '/info.log' error_log_file = log_folder + '/error.log' # check\",\n \"+ '/info.log' error_log_file = log_folder + '/error.log' # check if\",\n \"] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\ logging.Formatter( '%(asctime)s\",\n \"folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder)\",\n \"' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name:\",\n \"if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise #\",\n \"handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger = logging.getLogger(name) #\",\n \"----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import errno import\",\n \"= os.path.dirname(os.path.abspath(__file__)) # define log folder related to location log_folder\",\n \"linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error)\",\n \"'debug.ini' info_log_file = log_folder + '/info.log' error_log_file = log_folder +\",\n \"'/error.log' # check if exists or create log folder try:\",\n \"linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\ logging.Formatter(\",\n \"utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>,\",\n \"= \\\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ]\",\n \"config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info\",\n \"# runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder\",\n \"'%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread:\",\n \"' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error\",\n \"-*- coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke\",\n \"\\\"<EMAIL>\\\" import errno import logging import os from logging.config import\",\n \"= log_folder + '/error.log' # check if exists or create\",\n \"= logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels\",\n \"# -*- coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021,\",\n \"info_log_file = log_folder + '/info.log' error_log_file = log_folder + '/error.log'\",\n \"else: raise # setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file,\",\n \"ini and log files ini_file = 'debug.ini' info_log_file = log_folder\",\n \"check if exists or create log folder try: os.makedirs(log_folder, exist_ok=True)\",\n \"%(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S')\",\n \"created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # -----------------------------------------------------------\",\n \"os.makedirs(log_folder) except OSError as exc: # Python >2.5 if exc.errno\",\n \"create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file))\",\n \"# create formatters and add to handlers format_info = \\\\\",\n \"<NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__\",\n \"# ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL> #\",\n \"exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup\",\n \">2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else: raise\",\n \"except TypeError: try: os.makedirs(log_folder) except OSError as exc: # Python\",\n \"ini_file = 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file =\",\n \"error_log_file = log_folder + '/error.log' # check if exists or\",\n \"https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import\",\n \"define ini and log files ini_file = 'debug.ini' info_log_file =\",\n \"= \\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] '\",\n \"location log_folder = os.path.join(this_folder, '../logs') # define ini and log\",\n \"fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log\",\n \"os.path.join(this_folder, '../logs') # define ini and log files ini_file =\",\n \"configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers\",\n \"= \\\"<EMAIL>\\\" import errno import logging import os from logging.config\",\n \"add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ == '__main__':\",\n \"this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related to location\",\n \"handlers format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s\",\n \"logging.config import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) #\",\n \"logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters\",\n \"log folder related to location log_folder = os.path.join(this_folder, '../logs') #\",\n \"%(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger =\",\n \"handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ == '__main__': pass\",\n \"# add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__ ==\",\n \"# create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder,\",\n \"as exc: # Python >2.5 if exc.errno == errno.EEXIST and\",\n \"= 'debug.ini' info_log_file = log_folder + '/info.log' error_log_file = log_folder\",\n \"] ' '[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def\",\n \"os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder,\",\n \"%(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s\",\n \"format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ]\",\n \"import fileConfig # runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define\",\n \"coding: utf-8 -*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke #\",\n \"# define ini and log files ini_file = 'debug.ini' info_log_file\",\n \"folder related to location log_folder = os.path.join(this_folder, '../logs') # define\",\n \"exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as\",\n \"== errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup configuration\",\n \"%(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__):\",\n \"import errno import logging import os from logging.config import fileConfig\",\n \"exists or create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2\",\n \"import logging import os from logging.config import fileConfig # runtime\",\n \"exc: # Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder):\",\n \"os.path.dirname(os.path.abspath(__file__)) # define log folder related to location log_folder =\",\n \"and os.path.isdir(log_folder): pass else: raise # setup configuration config_file =\",\n \"' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\ logging.Formatter( '%(asctime)s %(levelname)s\",\n \"try: os.makedirs(log_folder) except OSError as exc: # Python >2.5 if\",\n \"and log files ini_file = 'debug.ini' info_log_file = log_folder +\",\n \"try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except\",\n \"%(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\\\\",\n \"\\\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] '\",\n \"__name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return\",\n \"# Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as exc:\",\n \"runtime location this_folder = os.path.dirname(os.path.abspath(__file__)) # define log folder related\",\n \"TypeError: try: os.makedirs(log_folder) except OSError as exc: # Python >2.5\",\n \"raise # setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True)\",\n \"fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error\",\n \"log_folder = os.path.join(this_folder, '../logs') # define ini and log files\",\n \"logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s', datefmt='%Y-%m-%d\",\n \"if exists or create log folder try: os.makedirs(log_folder, exist_ok=True) #\",\n \"02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__\",\n \"Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError as exc: #\",\n \"setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create\",\n \"errno.EEXIST and os.path.isdir(log_folder): pass else: raise # setup configuration config_file\",\n \"# Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass\",\n \"create formatters and add to handlers format_info = \\\\ logging.Formatter('%(asctime)s\",\n \"= os.path.join(this_folder, '../logs') # define ini and log files ini_file\",\n \"__email__ = \\\"<EMAIL>\\\" import errno import logging import os from\",\n \"logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error) return logger if __name__\",\n \"# setup configuration config_file = os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) #\",\n \"to location log_folder = os.path.join(this_folder, '../logs') # define ini and\",\n \"= \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import errno import logging import\",\n \"'../logs') # define ini and log files ini_file = 'debug.ini'\",\n \"format_error = \\\\ logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s\",\n \"= __name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info) logger.addHandler(handler_error)\",\n \"#!/usr/bin/python3 # -*- coding: utf-8 -*- # ----------------------------------------------------------- # created\",\n \"<EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__ =\",\n \"tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ =\",\n \"# ----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import errno\",\n \"\\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\" import errno import logging import os\",\n \"# https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \\\"<NAME>\\\" __email__ = \\\"<EMAIL>\\\"\",\n \"----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh\",\n \"'[ thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str\",\n \"to handlers format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s\",\n \"# check if exists or create log folder try: os.makedirs(log_folder,\",\n \"-*- # ----------------------------------------------------------- # created 02.02.2021, tkaulke # <NAME>, <EMAIL>\",\n \"# define log folder related to location log_folder = os.path.join(this_folder,\",\n \"os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try: os.makedirs(log_folder) except OSError\",\n \"errno import logging import os from logging.config import fileConfig #\",\n \"# <NAME>, <EMAIL> # https://github.com/kaulketh # ----------------------------------------------------------- __author__ = \\\"<NAME>\\\"\",\n \"= os.path.join(this_folder, ini_file) fileConfig(config_file, disable_existing_loggers=True) # create handlers handler_info =\",\n \"pass else: raise # setup configuration config_file = os.path.join(this_folder, ini_file)\",\n \"handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to handlers format_info\",\n \"log files ini_file = 'debug.ini' info_log_file = log_folder + '/info.log'\",\n \"define log folder related to location log_folder = os.path.join(this_folder, '../logs')\",\n \"info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR)\",\n \"+ '/error.log' # check if exists or create log folder\",\n \"logging import os from logging.config import fileConfig # runtime location\",\n \"thread: %(threadName)s ] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str =\",\n \"os from logging.config import fileConfig # runtime location this_folder =\",\n \"formatters and add to handlers format_info = \\\\ logging.Formatter('%(asctime)s %(levelname)s\",\n \"] %(message)s') handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger\",\n \"logging.Formatter( '%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[\",\n \"logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO)\",\n \"set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and add to\",\n \"except OSError as exc: # Python >2.5 if exc.errno ==\",\n \"Python >2.5 if exc.errno == errno.EEXIST and os.path.isdir(log_folder): pass else:\",\n \"related to location log_folder = os.path.join(this_folder, '../logs') # define ini\",\n \"disable_existing_loggers=True) # create handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error =\",\n \"OSError as exc: # Python >2.5 if exc.errno == errno.EEXIST\",\n \"or create log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except\",\n \"os.path.isdir(log_folder): pass else: raise # setup configuration config_file = os.path.join(this_folder,\",\n \"str = __name__): logger = logging.getLogger(name) # add handler logger.addHandler(handler_info)\",\n \"error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) # create formatters and\",\n \"'[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '[ thread: %(threadName)s ] %(message)s')\",\n \"get_logger(name: str = __name__): logger = logging.getLogger(name) # add handler\",\n \"# created 02.02.2021, tkaulke # <NAME>, <EMAIL> # https://github.com/kaulketh #\",\n \"log folder try: os.makedirs(log_folder, exist_ok=True) # Python>3.2 except TypeError: try:\",\n \"handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set\",\n \"handlers handler_info = logging.FileHandler(os.path.join(this_folder, info_log_file)) handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) #\",\n \"\\\\ logging.Formatter('%(asctime)s %(levelname)s ' '[ %(module)s.%(funcName)s linenr.%(lineno)s ] ' '%(message).180s',\",\n \"handler_error.setLevel(logging.ERROR) # create formatters and add to handlers format_info =\",\n \"handler_info.setFormatter(format_info) handler_error.setFormatter(format_error) def get_logger(name: str = __name__): logger = logging.getLogger(name)\"\n]","isConversational":false}}},{"rowIdx":680,"cells":{"ngram":{"kind":"list like","value":["True, use_propensity = True, test_size = 0) uplift_model.best_params_net y_test, x_test,","rmse_tolerance = .05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1],","t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t =","= np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert","= param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity =","= 1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1,","get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha","x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat","unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0]","== 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff =","axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000])","= get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1],","( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape ==","[0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 =","copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid","get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality","correct_tmts_3 = np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==","test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t","MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss","uplift_model.get_erupt_curves(x = x, y = y, t = t) assert","== (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE =","= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1","uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis =","test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05","y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert","uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice =","= MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid,","missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs,","1 assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE","activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model","param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity = True,","1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance","= get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1)","1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in","batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift()","mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys","= get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis =","np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert","10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise,","uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self):","y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = False)","import numpy as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data,","= optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1","param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts()","y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y,","optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff =","= [.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y,","1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1],","10000 TOLERANCE = .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs)","uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True)","x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t,","dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times","np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments)","assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self):","= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t","= dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model =","= get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512],","as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1","= 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test,","y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y,","= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise","n_jobs=1, optimized_loss = True, use_propensity = True, test_size = 0)","uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def","== 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1)","= t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t","optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape == (","experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff","missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs,","np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3","True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] >","in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3","dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs)","get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[","get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1)","activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model","= get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1,",".05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model","y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights","epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise,","[1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 =","3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) ==","t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1,","num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1],","= np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,","= param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity =","assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t","1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000])","test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t","n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid","= x, y = y, t = t) assert uplift_model_named.get_erupt_curves()","y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in","assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 == test_2_values)","n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x)","[.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1,","test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],","= param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns","= param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs","oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] >","t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'],","= np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid =","np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0,","y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y,","= uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance","uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1,","activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y,","get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],","== -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0],","in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid =","1000 y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1),","= True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert","= ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean()","y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs =","x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])]","= param_grid, optimized_loss = True, use_propensity = True) varimp_propensity =","]).mean() correct_tmts_2 = np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups","masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1]))","correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups ==","masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape","MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1 =","optimized_loss = True, use_propensity = True, test_size = 0) uplift_model.best_params_net","np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean()",") == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert","use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs","num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30],","[.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t,","uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30],","copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1),","test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs)","= param_grid, n_jobs=1, optimized_loss = True, use_propensity = True, test_size","= ['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy() y_1","tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1)","= [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1),","np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y,","= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss","pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1 =","= .05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],","= y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x","= np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments =","t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8],","= True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0]","t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1],","varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y,","== test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0],","def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)","test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t","= get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values =","t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1)","uplift_model.get_erupt_curves(x = x, y = y, t = t) def","MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss =","y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for","True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis =","uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid =","assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000 param_grid","== ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape","np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) ==","n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) <","utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape ==","def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =","def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],","1)], axis=1) unique_treatments = np.unique(t, axis = 0) masks =","activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1])","y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True)","varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],","tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice","param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns =","def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =","t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid","range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def","(missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0,","[0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 =","missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape ==","True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE","= True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis","for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups","correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is","correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert","param_grid = param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer =","np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0,","0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff","assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False assert correct_tmts_no_cutoff>TOLERANCE","= np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks,","use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1]","= MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer","param_grid = param_grid, optimized_loss = True, use_propensity = True) varimp_propensity","= get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)],","np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha","1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis","in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1","0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups =","* .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert","uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x =","epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid =","uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =","batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift()","x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] =","batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model = MRUplift()","= x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1)","((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff","axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat,","param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x,","import prepare_data_optimized_loss import sys import pandas as pd class TestMRUplift(object):","MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas","MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp","rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]]","n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity = True)","= np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values)",".5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis = 0)","= 1000 y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1,","num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1])","x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for","= x, y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self):","= 3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]),","x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert","= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert","[1, .5]]) rmse_tolerance = .05 num_obs = 10000 param_grid =","y, t, param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1","activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1])","y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True,","assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert(","num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid","prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs,","assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for","= [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid =","oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000 param_grid =","= x, y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self):","from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift,","num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1)","axis=1) unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments))","unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x,","assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000","1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1))","optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat =","numpy as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data,","t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x,","= t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y,","= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5],","y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity","np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[","get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid","np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8],","x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1,","== (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs =","1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1","def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y, x,","t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000","uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape ==",".98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8],","x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],","True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7,","test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 ==","y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid","dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift()","0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights","]).mean() correct_tmts_3 = np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups","def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1 =","t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments","2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for x in","= get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] =","tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1),","1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) == 1","x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in","y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1,","pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import","np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from","for x in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1)","test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05","correct_tmts_4 = np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==","get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1,","x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid =","dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times =","= 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False)","axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs,","assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE","= 10000 TOLERANCE = .98 y, x, t, rule_assignment =","param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity = True)","batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise","x, y = y, t = t) assert uplift_model_named.get_erupt_curves() def","= np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs =","test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1,","uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x =","((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE","param_grid, n_jobs=1, optimized_loss = True, use_propensity = True, test_size =","get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import","= uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0","= x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x","= MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid,","param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha =","y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True,","== 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE","= ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1,","dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times","y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])]","= 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks,","==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]])","weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape","= np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]]","1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1],","uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape ==","prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0],","10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha","param_grid, optimized_loss = True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights","epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity =","uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1)","true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1,","np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean()","(num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in","batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,","test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t","= True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert","= uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x","< rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t","= np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights,","y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs =","uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1)","optimized_loss = True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights =","dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times =","y, t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity","np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000","= True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs *","MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape","use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat =","as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2","True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs =","param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise,","varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def","= 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups","= np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs =","y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y,","== 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for x","(missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000","= [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1,","= 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks,","tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity","assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98","t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100],","assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False","uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2","= uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self):","test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values","TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1","= uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def","param_grid, n_jobs=1, optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape","x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments)","import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from","= y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs","param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0],","masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights =","t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1","True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test =","= pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1","= True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self):","0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for x in","x in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns","in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2]","= get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values =","alpha = [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x,","optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for","3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2)","t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x,","weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape ==","in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4","np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility,","num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y =","uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t =","oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance","1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for x in","y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift()","pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10","def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)","np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks","= param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8],","optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1)","1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity =","in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]]","1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) ==","def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)","== x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE","epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model =","[.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t,","y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == (","= 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100],","uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs",".5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'],","<gh_stars>10-100 import numpy as np import pytest from mr_uplift.dataset.data_simulation import","1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha","= y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000","0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights","= 1) correct_tmts_1 = np.array([x in [0,1] for x in","uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =","= prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape ==","num_obs = 10000 TOLERANCE = .98 y, x, t, rule_assignment","get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss","x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in","False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat","= 1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1)","-true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x,","assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x","import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import","MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss","0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise,","= uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],","class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3","10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y,","for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t,","optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise,","y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1],","x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in","param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs *","== 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for x","MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer =","num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise,","param_grid = param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity","= [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()","from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as pd","1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments","-999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x,","axis=1) assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 ==","np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean()","x, y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs","1) correct_tmts_1 = np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups","oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs","= np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1)","= False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -","= t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t","dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times","[1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid,","y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs =","= [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid =","= t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1],","test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)],","t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks","t, param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1 =","alpha = [.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x,","MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss =","in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns =","num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model","uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss = True,","pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named =","True, use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0],","for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean()","(num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) ==","uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1,","= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99],","correct_tmts_1 = np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==","in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0]","y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments =","get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1,","== -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y,","x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert","tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re","n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7,","varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE =","== 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 )","uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid","= np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]]","= [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y,","experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]]","]).mean() correct_tmts_4 = np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups","0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000 param_grid","num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity","x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff","uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp =","= True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1))","= prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs,","np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 == test_2_values) ==","def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)","False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2))","0 def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1],","get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0","[1, .5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise,","t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity","assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE =","unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x,","[1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid,","dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x,","rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t =","optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1]","[0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 =","assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE =","= 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x","range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid,","uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss = True,","((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs","np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean()","True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1]","use_propensity = True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test,","['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy() y_1 =","= optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1] for","= t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks =","uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid =","optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1)","10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2 =","uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =","= MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss","x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid","optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for","epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift()","= True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs","[2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,","t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],","batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x,","test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t",") def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance","= get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1,","1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1),","dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)","= MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid,","np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility,","for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x","true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs","get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid","> 0 def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8],","t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity =","mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data","epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()","rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20],","1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert",") def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t =","t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights =","epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model =","optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ==","1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1","y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]):","True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1,","y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert","param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1],","10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,","assert uplift_model.get_erupt_curves(x = x, y = y, t = t)","utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape","[.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model","n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts()","== 1 assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self):","y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]]","get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as","= [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)","param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha =","import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift","np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) == 1 assert","in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2","= pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named","x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups ==","test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05","- true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0],","= True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test","MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss","param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert","from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import","= 10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2","= x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test,","= np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]]","test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1 = get_t_data(0,","param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model","alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise =","t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1],","num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model =","== 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for x","pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 =","1), n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice =","num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis = 0) masks","0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000","axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat,","t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t,","dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise =","= np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]]","3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff","in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 )","= uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis","correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False assert","uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1 = x.copy()","MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape","assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self):","= 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff =","for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x","masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0],","= MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert","param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha =","experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff =","optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for","batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid,","num_obs_1 = 10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1)","experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x =","== x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0])","test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y, x, t,","masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights =","param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1)","t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1),","= 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100])","mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as pd class","prepare_data_optimized_loss import sys import pandas as pd class TestMRUplift(object): def","1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2))","assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1])",",unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1]))","dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times =","= param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True)","= np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] =","use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff =","1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1)","x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape","= True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise,","optimized_loss = True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert","]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups","np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000","num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =","np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0],","unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1]))","= uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1,","np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE","test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values","param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid =","'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t,","np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t,","import pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 =","y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == (","== test_1_values) == 1 assert np.mean(test_2 == test_2_values) == 1","t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis","correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE","== (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x,","q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1","num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1,","1), n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re =","n_jobs=1, optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape ==","uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y, t =","[2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid","['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1,","True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0]","num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t =","optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 =","uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid =","rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise =",".05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],","x, y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs","y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise,","assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat)","x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy()","num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model","num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0,","= np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100],","uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert","> 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs =","def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =","import sys import pandas as pd class TestMRUplift(object): def test_get_t_data(self):","num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =","test_1_values) == 1 assert np.mean(test_2 == test_2_values) == 1 def",".7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves()","TOLERANCE = .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid","= param_grid, n_jobs=1, optimized_loss = True, use_propensity = True) assert","oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) <",".5]]) rmse_tolerance = .05 num_obs = 10000 param_grid = dict(num_nodes=[8],","t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t =","def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)","= MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid)","t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x,","= False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True)","oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1),","y = y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self):","= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5],","x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5,","param_grid = param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity","(np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y","assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self):","= True, use_propensity = True, test_size = 0) uplift_model.best_params_net y_test,","assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE","== 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]])","= .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)","get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice","range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x)","= t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1],","uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def","alpha = [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x,","= .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid =",".5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise","= y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000","1)], axis=1) assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2","x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in","== (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q","n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer","copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid","correct_tmts_2 = np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==","in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2]","-999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1,","assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y, t","sys import pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1","activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2])","optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1] for x","= True, use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t,","copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =","==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t","1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x","rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance","= 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()","use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0","for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x","True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert","param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift()","== (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0)","missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1]))","uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid =","dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t =","y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] ==","np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis =","= np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights,","get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0)","< rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]])","x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1,"],"string":"[\n \"True, use_propensity = True, test_size = 0) uplift_model.best_params_net y_test, x_test,\",\n \"rmse_tolerance = .05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t =\",\n \"= np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert\",\n \"= param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity =\",\n \"= 1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1,\",\n \"get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha\",\n \"x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat\",\n \"unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0]\",\n \"== 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff =\",\n \"axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000])\",\n \"= get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape ==\",\n \"[0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 =\",\n \"copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid\",\n \"get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality\",\n \"correct_tmts_3 = np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==\",\n \"test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t\",\n \"MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss\",\n \"uplift_model.get_erupt_curves(x = x, y = y, t = t) assert\",\n \"== (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE =\",\n \"= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1\",\n \"uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis =\",\n \"test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05\",\n \"y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert\",\n \"uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice =\",\n \"= MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid,\",\n \"missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs,\",\n \"1 assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE\",\n \"activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model\",\n \"param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity = True,\",\n \"1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance\",\n \"= get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1)\",\n \"1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in\",\n \"batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift()\",\n \"mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys\",\n \"= get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis =\",\n \"np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert\",\n \"10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise,\",\n \"uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self):\",\n \"y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = False)\",\n \"import numpy as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data,\",\n \"= optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1\",\n \"param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts()\",\n \"y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y,\",\n \"optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff =\",\n \"= [.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y,\",\n \"1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"10000 TOLERANCE = .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs)\",\n \"uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True)\",\n \"x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t,\",\n \"dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times\",\n \"np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments)\",\n \"assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self):\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model =\",\n \"= get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512],\",\n \"as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1\",\n \"= 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test,\",\n \"y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y,\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise\",\n \"n_jobs=1, optimized_loss = True, use_propensity = True, test_size = 0)\",\n \"uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def\",\n \"== 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1)\",\n \"= t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t\",\n \"optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape == (\",\n \"experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff\",\n \"missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs,\",\n \"np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3\",\n \"True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] >\",\n \"in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3\",\n \"dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs)\",\n \"get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[\",\n \"get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1)\",\n \"activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model\",\n \"= get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1,\",\n \".05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model\",\n \"y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights\",\n \"epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise,\",\n \"[1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 =\",\n \"3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) ==\",\n \"t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1,\",\n \"num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1],\",\n \"= np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,\",\n \"= param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity =\",\n \"assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t\",\n \"1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000])\",\n \"test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t\",\n \"n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid\",\n \"= x, y = y, t = t) assert uplift_model_named.get_erupt_curves()\",\n \"y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in\",\n \"assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 == test_2_values)\",\n \"n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x)\",\n \"[.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1,\",\n \"test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],\",\n \"= param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns\",\n \"= param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs\",\n \"oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] >\",\n \"t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'],\",\n \"= np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid =\",\n \"np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0,\",\n \"y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y,\",\n \"= uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance\",\n \"uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1,\",\n \"activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y,\",\n \"get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],\",\n \"== -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0],\",\n \"in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid =\",\n \"1000 y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1),\",\n \"= True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert\",\n \"= ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean()\",\n \"y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs =\",\n \"x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])]\",\n \"= param_grid, optimized_loss = True, use_propensity = True) varimp_propensity =\",\n \"]).mean() correct_tmts_2 = np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups\",\n \"masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1]))\",\n \"correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups ==\",\n \"masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape\",\n \"MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1 =\",\n \"optimized_loss = True, use_propensity = True, test_size = 0) uplift_model.best_params_net\",\n \"np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean()\",\n \") == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert\",\n \"use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs\",\n \"num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30],\",\n \"[.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t,\",\n \"uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30],\",\n \"copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1),\",\n \"test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs)\",\n \"= param_grid, n_jobs=1, optimized_loss = True, use_propensity = True, test_size\",\n \"= ['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy() y_1\",\n \"tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1)\",\n \"= [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1),\",\n \"np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y,\",\n \"= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss\",\n \"pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1 =\",\n \"= .05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],\",\n \"= y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x\",\n \"= np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments =\",\n \"t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8],\",\n \"= True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0]\",\n \"t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1],\",\n \"varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y,\",\n \"== test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0],\",\n \"def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)\",\n \"test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t\",\n \"= get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values =\",\n \"t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1)\",\n \"uplift_model.get_erupt_curves(x = x, y = y, t = t) def\",\n \"MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss =\",\n \"y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for\",\n \"True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis =\",\n \"uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid =\",\n \"assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000 param_grid\",\n \"== ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape\",\n \"np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) ==\",\n \"n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) <\",\n \"utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape ==\",\n \"def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =\",\n \"def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],\",\n \"1)], axis=1) unique_treatments = np.unique(t, axis = 0) masks =\",\n \"activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1])\",\n \"y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True)\",\n \"varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],\",\n \"tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice\",\n \"param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns =\",\n \"def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =\",\n \"t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid\",\n \"range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def\",\n \"(missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0,\",\n \"[0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 =\",\n \"missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape ==\",\n \"True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE\",\n \"= True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis\",\n \"for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups\",\n \"correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is\",\n \"correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert\",\n \"param_grid = param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer =\",\n \"np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0,\",\n \"0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff\",\n \"assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False assert correct_tmts_no_cutoff>TOLERANCE\",\n \"= np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks,\",\n \"use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1]\",\n \"= MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer\",\n \"param_grid = param_grid, optimized_loss = True, use_propensity = True) varimp_propensity\",\n \"= get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)],\",\n \"np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha\",\n \"1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis\",\n \"in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1\",\n \"0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups =\",\n \"* .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert\",\n \"uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x =\",\n \"epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid =\",\n \"uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =\",\n \"batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift()\",\n \"x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] =\",\n \"batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model = MRUplift()\",\n \"= x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1)\",\n \"((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff\",\n \"axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat,\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x,\",\n \"import prepare_data_optimized_loss import sys import pandas as pd class TestMRUplift(object):\",\n \"MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas\",\n \"MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp\",\n \"rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]]\",\n \"n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity = True)\",\n \"= np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values)\",\n \".5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis = 0)\",\n \"= 1000 y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1,\",\n \"num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1])\",\n \"x_1 = x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for\",\n \"= x, y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self):\",\n \"= 3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]),\",\n \"x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert\",\n \"= MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert\",\n \"[1, .5]]) rmse_tolerance = .05 num_obs = 10000 param_grid =\",\n \"y, t, param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1\",\n \"activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1])\",\n \"y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True,\",\n \"assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert(\",\n \"num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid\",\n \"prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs,\",\n \"assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for\",\n \"= [1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid =\",\n \"oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000 param_grid =\",\n \"= x, y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self):\",\n \"from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift,\",\n \"num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1)\",\n \"axis=1) unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments))\",\n \"unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x,\",\n \"assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000\",\n \"1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1))\",\n \"optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat =\",\n \"numpy as np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data,\",\n \"t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x,\",\n \"= t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y,\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5],\",\n \"y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity\",\n \"np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[\",\n \"get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid\",\n \"np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8],\",\n \"x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1,\",\n \"== (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs =\",\n \"1)[0]] == 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1\",\n \"def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y, x,\",\n \"t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000\",\n \"uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape ==\",\n \".98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8],\",\n \"x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'],\",\n \"True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7,\",\n \"test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 ==\",\n \"y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid\",\n \"dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift()\",\n \"0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights\",\n \"]).mean() correct_tmts_3 = np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups\",\n \"def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1 =\",\n \"t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments\",\n \"2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for x in\",\n \"= get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] =\",\n \"tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1),\",\n \"1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) == 1\",\n \"x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in\",\n \"y, x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1,\",\n \"pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import\",\n \"np import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from\",\n \"for x in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1)\",\n \"test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05\",\n \"correct_tmts_4 = np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==\",\n \"get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1,\",\n \"x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1) param_grid =\",\n \"dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times =\",\n \"= 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False)\",\n \"axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs,\",\n \"assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE\",\n \"= 10000 TOLERANCE = .98 y, x, t, rule_assignment =\",\n \"param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity = True)\",\n \"batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise\",\n \"x, y = y, t = t) assert uplift_model_named.get_erupt_curves() def\",\n \"= np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs =\",\n \"test_2_values) == 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1,\",\n \"uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x =\",\n \"((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE\",\n \"param_grid, n_jobs=1, optimized_loss = True, use_propensity = True, test_size =\",\n \"get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import\",\n \"= uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0\",\n \"= x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x\",\n \"= MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid,\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha =\",\n \"y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True,\",\n \"== 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE\",\n \"= ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1,\",\n \"dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times\",\n \"y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])]\",\n \"= 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks,\",\n \"==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]])\",\n \"weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape\",\n \"= np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]]\",\n \"1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1],\",\n \"uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape ==\",\n \"prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0],\",\n \"10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha\",\n \"param_grid, optimized_loss = True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights\",\n \"epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity =\",\n \"uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1)\",\n \"true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1,\",\n \"np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean()\",\n \"(num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in\",\n \"batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,\",\n \"test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t\",\n \"= True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert\",\n \"= uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x\",\n \"< rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t\",\n \"= np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights,\",\n \"y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs =\",\n \"uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1)\",\n \"optimized_loss = True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights =\",\n \"dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times =\",\n \"y, t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity\",\n \"np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000\",\n \"= True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs *\",\n \"MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape\",\n \"use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat =\",\n \"as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2\",\n \"True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs =\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise,\",\n \"varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def\",\n \"= 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups\",\n \"= np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs =\",\n \"y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y,\",\n \"== 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for x\",\n \"(missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000\",\n \"= [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1,\",\n \"= 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks,\",\n \"tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity\",\n \"assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE = .98\",\n \"t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100],\",\n \"assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False\",\n \"uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2\",\n \"= uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self):\",\n \"test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values\",\n \"TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1\",\n \"= uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def\",\n \"param_grid, n_jobs=1, optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape\",\n \"x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments)\",\n \"import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from\",\n \"= y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs\",\n \"param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0],\",\n \"masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights =\",\n \"t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1\",\n \"True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test =\",\n \"= pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1\",\n \"= True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self):\",\n \"0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for x in\",\n \"x in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns\",\n \"in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2]\",\n \"= get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values =\",\n \"alpha = [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x,\",\n \"optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for\",\n \"3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2)\",\n \"t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x,\",\n \"weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape ==\",\n \"in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4\",\n \"np.unique(t, axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility,\",\n \"num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y =\",\n \"uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t =\",\n \"oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance\",\n \"1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for x in\",\n \"y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift()\",\n \"pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10\",\n \"def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)\",\n \"np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks\",\n \"= param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8],\",\n \"optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1)\",\n \"1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity =\",\n \"in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]]\",\n \"1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) ==\",\n \"def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)\",\n \"== x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE\",\n \"epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model =\",\n \"[.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t,\",\n \"y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == (\",\n \"= 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100],\",\n \"uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs\",\n \".5, num_obs).reshape(-1, 1)], axis=1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'],\",\n \"<gh_stars>10-100 import numpy as np import pytest from mr_uplift.dataset.data_simulation import\",\n \"1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha\",\n \"= y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000\",\n \"0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat, masks, weights\",\n \"= 1) correct_tmts_1 = np.array([x in [0,1] for x in\",\n \"uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =\",\n \"= prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape ==\",\n \"num_obs = 10000 TOLERANCE = .98 y, x, t, rule_assignment\",\n \"get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss\",\n \"x.copy() x_1 = pd.DataFrame(x_1) x_1.columns = ['var_'+str(x) for x in\",\n \"False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat\",\n \"= 1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1)\",\n \"-true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x,\",\n \"assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x\",\n \"import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import\",\n \"MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss\",\n \"0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise,\",\n \"= uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],\",\n \"class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3\",\n \"10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y,\",\n \"for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t,\",\n \"optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise,\",\n \"y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in\",\n \"param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs *\",\n \"== 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0] for x\",\n \"MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer =\",\n \"num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) y_no_noise,\",\n \"param_grid = param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity\",\n \"= [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()\",\n \"from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as pd\",\n \"1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments\",\n \"-999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x,\",\n \"axis=1) assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 ==\",\n \"np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean()\",\n \"x, y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs\",\n \"1) correct_tmts_1 = np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups\",\n \"oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs\",\n \"= np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1)\",\n \"= False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -\",\n \"= t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t\",\n \"dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times\",\n \"[1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid,\",\n \"y = y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs =\",\n \"= [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid =\",\n \"= t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1],\",\n \"test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)],\",\n \"t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks\",\n \"t, param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1 =\",\n \"alpha = [.5], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x,\",\n \"MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss =\",\n \"in range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns =\",\n \"num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model\",\n \"uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1, optimized_loss = True,\",\n \"pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named =\",\n \"True, use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0],\",\n \"for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean()\",\n \"(num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) ==\",\n \"uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1,\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99],\",\n \"correct_tmts_1 = np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==\",\n \"in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x in [0]\",\n \"y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments =\",\n \"get_t_data(np.array([0, 1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1,\",\n \"== -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y,\",\n \"x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert\",\n \"tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re\",\n \"n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7,\",\n \"varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE =\",\n \"== 1) == x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 )\",\n \"uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid\",\n \"= np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]]\",\n \"= [.5], copy_several_times = [2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y,\",\n \"experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]]\",\n \"]).mean() correct_tmts_4 = np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups\",\n \"0], [1, .5]]) rmse_tolerance = .05 num_obs = 10000 param_grid\",\n \"num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2]) uplift_model_propensity\",\n \"x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff\",\n \"uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp =\",\n \"= True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1))\",\n \"= prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs,\",\n \"np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2 == test_2_values) ==\",\n \"def test_model_pred_oos_shapes_single_col_tmt(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)\",\n \"False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2))\",\n \"0 def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0\",\n \"[1, .5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise,\",\n \"t.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss = True, use_propensity\",\n \"assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE =\",\n \"unique_treatments = np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x,\",\n \"[1]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid,\",\n \"dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x,\",\n \"rmse_tolerance def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t =\",\n \"optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1]\",\n \"[0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 =\",\n \"assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self): true_ATE =\",\n \"= 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x\",\n \"range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid,\",\n \"uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss = True,\",\n \"((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs\",\n \"np.array([x in [0] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean()\",\n \"True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1]\",\n \"use_propensity = True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test,\",\n \"['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy() y_1 =\",\n \"= optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1] for\",\n \"= t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0) masks =\",\n \"uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid =\",\n \"optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1)\",\n \"10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2 =\",\n \"uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss =\",\n \"= MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid = param_grid, n_jobs=1, optimized_loss\",\n \"x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid\",\n \"optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2] for\",\n \"epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift()\",\n \"= True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2)) < rmse_tolerance def test_model_pred_oos_shapes(self): num_obs\",\n \"[2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,\",\n \"t = t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],\",\n \"batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x,\",\n \"test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t\",\n \") def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance\",\n \"= get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1,\",\n \"1]), num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1),\",\n \"dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)\",\n \"= MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid,\",\n \"np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility,\",\n \"for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]] ]).mean() correct_tmts_4 = np.array([x\",\n \"true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05 num_obs\",\n \"get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid\",\n \"> 0 def test_varimp(self): num_obs = 10000 param_grid = dict(num_nodes=[8],\",\n \"t, param_grid = param_grid, n_jobs=1, optimized_loss = True, use_propensity =\",\n \"mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift import MRUplift, get_t_data\",\n \"epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()\",\n \"rule_assignment = get_observational_uplift_data_1(num_obs) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20],\",\n \"1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert\",\n \") def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t =\",\n \"t.reshape(-1, 1), n_jobs=1, param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights =\",\n \"epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times = [1]) uplift_model =\",\n \"optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ==\",\n \"1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1\",\n \"y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]):\",\n \"True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1,\",\n \"y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert\",\n \"param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1],\",\n \"10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise,\",\n \"assert uplift_model.get_erupt_curves(x = x, y = y, t = t)\",\n \"utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t, masks ,unique_treatments) assert(utility_weights.shape\",\n \"[.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model\",\n \"n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts()\",\n \"== 1 assert np.mean(test_2 == test_2_values) == 1 def test_model_mean_outputs(self):\",\n \"y_test, x_test, t_test, rule_assignment_test = get_observational_uplift_data_1(num_obs) experiment_groups = np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]]\",\n \"get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as\",\n \"= [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha =\",\n \"import pytest from mr_uplift.dataset.data_simulation import get_no_noise_data, get_simple_uplift_data, get_observational_uplift_data_1 from mr_uplift.mr_uplift\",\n \"np.ones(num_obs_2).reshape(-1, 1)], axis=1) assert np.mean(test_1 == test_1_values) == 1 assert\",\n \"in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2\",\n \"= pd.DataFrame(y_1) y_1.columns = ['var_'+str(x) for x in range(y.shape[1])] uplift_model_named\",\n \"x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 3)[0]] ]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups ==\",\n \"test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance = .05\",\n \"- true_ATE)**2)) < rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0],\",\n \"= True, test_size = 0) uplift_model.best_params_net y_test, x_test, t_test, rule_assignment_test\",\n \"MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid = param_grid, optimized_loss\",\n \"param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True) assert\",\n \"from mr_uplift.mr_uplift import MRUplift, get_t_data from mr_uplift.keras_model_functionality import prepare_data_optimized_loss import\",\n \"= 10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2\",\n \"= x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test,\",\n \"= np.array([x in [0,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 2)[0]]\",\n \"test_get_t_data(self): num_obs_1 = 10 num_obs_2 = 3 test_1 = get_t_data(0,\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model\",\n \"alpha = [.5], copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise =\",\n \"t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=[ 'relu'], num_layers=[1], epochs=[1],\",\n \"num_layers=[2], epochs=[30], batch_size=[100]) y, x, t = get_simple_uplift_data(num_obs) uplift_model =\",\n \"== 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for x\",\n \"pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 = 10 num_obs_2 =\",\n \"1), n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice =\",\n \"num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis = 0) masks\",\n \"0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs = 10000\",\n \"axis = 0) masks = np.ones(num_obs*len(unique_treatments)).reshape(num_obs,len(unique_treatments)) x, utility_weights, missing_utility, missing_y_mat,\",\n \"t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == ( np.unique(t,\",\n \"dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise =\",\n \"= np.array([x in [0,1] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]]\",\n \"3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = False) optim_treatments_cuttoff\",\n \"in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 )\",\n \"= uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True) optim_treatments_cuttoff_cat = optim_treatments_cuttoff.argmax(axis\",\n \"correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat) is False assert\",\n \"uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) x_1 = x.copy()\",\n \"MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape\",\n \"assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self):\",\n \"= 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff =\",\n \"for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x\",\n \"masks ,unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0],\",\n \"= MRUplift() uplift_model_named.fit(x_1, y_1, t, param_grid = param_grid, n_jobs=1) assert\",\n \"param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1], batch_size=[100], alpha =\",\n \"experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff =\",\n \"optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] ]).mean() correct_tmts_3 = np.array([x in [0,2] for\",\n \"batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t, param_grid = param_grid,\",\n \"num_obs_1 = 10 num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1)\",\n \"experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x =\",\n \"== x_test[np.where(experiment_groups == 1)[0],0]).mean() correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0])\",\n \"test_model_propensity(self): num_obs = 10000 TOLERANCE = .98 y, x, t,\",\n \"masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights, missing_utility, missing_y_mat, masks, weights =\",\n \"param_grid = param_grid, n_jobs=1) x_1 = x.copy() x_1 = pd.DataFrame(x_1)\",\n \"t = get_simple_uplift_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1),\",\n \"= 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100])\",\n \"mr_uplift.keras_model_functionality import prepare_data_optimized_loss import sys import pandas as pd class\",\n \"prepare_data_optimized_loss import sys import pandas as pd class TestMRUplift(object): def\",\n \"1), n_jobs=1) oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) -true_ATE)**2))\",\n \"assert uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1])\",\n \",unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1]))\",\n \"dropout=[.1], activation=['relu'], num_layers=[1], epochs=[20], batch_size=[512], alpha = [.9999,.99], copy_several_times =\",\n \"= param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer = True)\",\n \"= np.zeros(num_obs)+2 experiment_groups[np.where(x_test[:,-2]<.5)[0]] = 1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] =\",\n \"use_propensity_score_cutoff = False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff =\",\n \"1000 y, x, t = get_simple_uplift_data(num_obs) t = t.reshape(-1, 1)\",\n \"x, utility_weights, missing_utility, missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape\",\n \"= True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise,\",\n \"optimized_loss = True, use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert\",\n \"]).mean() correct_tmts_experiment_groups_1 = ((optim_treatments_cuttoff_cat[np.where(experiment_groups == 1)[0]] == 1) == x_test[np.where(experiment_groups\",\n \"np.array([.7,-.3,0]).reshape(1,-1)) assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000\",\n \"num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =\",\n \"np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0],\",\n \"unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1]))\",\n \"= uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1,\",\n \"np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE\",\n \"test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0, 1]), num_obs_2) test_1_values\",\n \"param_grid = param_grid) varimp = uplift_model.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) param_grid =\",\n \"'relu'], num_layers=[1], epochs=[1], batch_size=[1000]) uplift_model = MRUplift() uplift_model.fit(x, y, t,\",\n \"np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t,\",\n \"import pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1 =\",\n \"y_1, t, param_grid = param_grid, n_jobs=1) assert uplift_model.predict_ice().shape == (\",\n \"== test_1_values) == 1 assert np.mean(test_2 == test_2_values) == 1\",\n \"t = get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis\",\n \"correct_tmts_no_cutoff = np.mean((optim_treatments_no_cuttoff_cat==1 ) == x_test[:,0]) assert correct_tmts_1>TOLERANCE assert correct_tmts_2>TOLERANCE\",\n \"== (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x,\",\n \"q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1\",\n \"num_obs_2) test_1_values = np.zeros(num_obs_1).reshape(-1, 1) test_2_values = np.concatenate([np.zeros(num_obs_2).reshape(-1, 1), np.ones(num_obs_2).reshape(-1,\",\n \"1), n_jobs=1, param_grid = param_grid, optimized_loss = True) oos_re =\",\n \"n_jobs=1, optimized_loss = True, use_propensity = True) assert uplift_model.predict_ice().shape ==\",\n \"uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y, t =\",\n \"[2]) uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid\",\n \"['var_'+str(x) for x in range(y.shape[1])] uplift_model_named = MRUplift() uplift_model_named.fit(x_1, y_1,\",\n \"True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0]\",\n \"num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs) t =\",\n \"optim_treatments_cuttoff.argmax(axis = 1) optim_treatments_no_cuttoff_cat = optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 =\",\n \"uplift_model = MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid =\",\n \"rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise =\",\n \".05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2],\",\n \"x, y = y, t = t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs\",\n \"y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise,\",\n \"assert correct_tmts_2>TOLERANCE assert correct_tmts_3>TOLERANCE assert correct_tmts_4>TOLERANCE assert correct_tmts_experiment_groups_1>TOLERANCE assert np.array_equal(optim_treatments_cuttoff_cat,optim_treatments_no_cuttoff_cat)\",\n \"x_1.columns = ['var_'+str(x) for x in range(x.shape[1])] y_1 = y.copy()\",\n \"num_layers=[2], epochs=[1], batch_size=[100], alpha = [.5], copy_several_times = [1]) uplift_model\",\n \"num_obs_2 = 3 test_1 = get_t_data(0, num_obs_1) test_2 = get_t_data(np.array([0,\",\n \"= np.array([.7,-.3,0]).reshape(1,-1)) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100],\",\n \"uplift_model.predict_ice().shape == ( np.unique(t, axis=0).shape[0], num_obs * .7, y.shape[1]) assert\",\n \"> 0 assert oos_re_propensity['mean'].iloc[0] > 0 def test_varimp(self): num_obs =\",\n \"def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance =\",\n \"import sys import pandas as pd class TestMRUplift(object): def test_get_t_data(self):\",\n \"num_layers=[2], epochs=[30], batch_size=[100]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =\",\n \"test_1_values) == 1 assert np.mean(test_2 == test_2_values) == 1 def\",\n \".7, y.shape[1]) assert uplift_model.predict_ice(x=x).shape == (np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves()\",\n \"TOLERANCE = .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid\",\n \"= param_grid, n_jobs=1, optimized_loss = True, use_propensity = True) assert\",\n \"oos_ice = uplift_model.predict_ice(response_transformer = True) assert np.sqrt(np.mean((oos_ice.mean(axis=1) - true_ATE)**2)) <\",\n \".5]]) rmse_tolerance = .05 num_obs = 10000 param_grid = dict(num_nodes=[8],\",\n \"t) def test_model_pred_oos_shapes_single_col_tmt_propensity(self): num_obs = 1000 y, x, t =\",\n \"def test_model_pred_oos_shapes(self): num_obs = 1000 y, x, t = get_simple_uplift_data(num_obs)\",\n \"= MRUplift() uplift_model.fit(x, y, t.reshape(-1, 1), n_jobs=1, param_grid = param_grid)\",\n \"t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x,\",\n \"= False) optim_treatments_cuttoff = uplift_model.predict_optimal_treatments(x = x_test, use_propensity_score_cutoff = True)\",\n \"oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1),\",\n \"y = y, t = t) assert uplift_model_named.get_erupt_curves() def test_model_pred_oos_shapes_single_col_tmt(self):\",\n \"= dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5],\",\n \"x, t = get_simple_uplift_data(num_obs) t = np.concatenate([t.reshape(-1, 1), np.random.binomial(1, .5,\",\n \"param_grid = param_grid, optimized_loss = True, use_propensity = True) oos_re_propensity\",\n \"(np.unique(t,axis=0).shape[0], num_obs, y.shape[1]) assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y\",\n \"assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] == -999)).mean() ==1 ) def test_model_optim_mean_outputs(self):\",\n \"= True, use_propensity = True, test_size = 0) uplift_model.best_params_net y_test,\",\n \"assert varimp['permutation_varimp_metric'].iloc[0]>varimp['permutation_varimp_metric'].iloc[1] assert varimp_propensity['permutation_varimp_metric'].iloc[0]>varimp_propensity['permutation_varimp_metric'].iloc[1] def test_model_propensity(self): num_obs = 10000 TOLERANCE\",\n \"== 1 def test_model_mean_outputs(self): true_ATE = np.array([[0, 0], [1, .5]])\",\n \"= .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs)\",\n \"get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1) oos_ice\",\n \"range(x.shape[1])] y_1 = y.copy() y_1 = pd.DataFrame(y_1) y_1.columns = ['var_'+str(x)\",\n \"= t.reshape(-1, 1) param_grid = dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[1],\",\n \"uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0 assert oos_re_propensity['mean'].iloc[0] > 0 def\",\n \"alpha = [.9999,.99], copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x,\",\n \"= .98 y, x, t, rule_assignment = get_observational_uplift_data_1(num_obs) param_grid =\",\n \".5]]) rmse_tolerance = .05 num_obs = 10000 y_no_noise, x_no_noise, tmt_no_noise\",\n \"= y, t = t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000\",\n \"1)], axis=1) assert np.mean(test_1 == test_1_values) == 1 assert np.mean(test_2\",\n \"x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in\",\n \"== (num_obs, y.shape[1])) assert(missing_y_mat.shape == (num_obs, unique_treatments.shape[0], y.shape[1])) for q\",\n \"n_jobs=1, param_grid = param_grid, optimized_loss = False) oos_ice = uplift_model.predict_ice(response_transformer\",\n \"copy_several_times = [1]) uplift_model = MRUplift() uplift_model.fit(x, y[:,0].reshape(-1,1), t, param_grid\",\n \"correct_tmts_2 = np.array([x in [1,2] for x in optim_treatments_cuttoff_cat[np.where(experiment_groups ==\",\n \"in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x in [1,2]\",\n \"-999)).mean() ==1 ) def test_model_optim_mean_outputs(self): true_ATE = np.array([[0, 0], [1,\",\n \"assert uplift_model.get_erupt_curves() assert uplift_model.get_erupt_curves(x = x, y = y, t\",\n \"sys import pandas as pd class TestMRUplift(object): def test_get_t_data(self): num_obs_1\",\n \"activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100], alpha = [.5], copy_several_times = [2])\",\n \"optim_treatments_no_cuttoff.argmax(axis = 1) correct_tmts_1 = np.array([x in [0,1] for x\",\n \"= True, use_propensity = True) assert uplift_model.predict_ice().shape == ( np.unique(t,\",\n \"copy_several_times = [2]) y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model =\",\n \"==1 ) def test_prepare_data_optimized_loss_two_col_tmt(self): num_obs = 1000 y, x, t\",\n \"1 experiment_groups[np.where(x_test[:,-2]<.33)[0]] = 0 experiment_groups[np.where(x_test[:,-1]>.8)[0]] = 3 optim_treatments_no_cuttoff = uplift_model.predict_optimal_treatments(x\",\n \"rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]]) rmse_tolerance\",\n \"= 10000 y_no_noise, x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift()\",\n \"use_propensity = True) oos_re_propensity = uplift_model_propensity.get_random_erupts() assert oos_re['mean'].iloc[0] > 0\",\n \"for x in optim_treatments_cuttoff_cat[np.where(experiment_groups == 0)[0]] ]).mean() correct_tmts_2 = np.array([x\",\n \"True, use_propensity = True) varimp_propensity = uplift_model_propensity.permutation_varimp(objective_weights = np.array([.7,-.3,0]).reshape(1,-1)) assert\",\n \"param_grid, optimized_loss = True) oos_re = uplift_model.get_random_erupts() uplift_model_propensity = MRUplift()\",\n \"== (num_obs, unique_treatments.shape[0], y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0)\",\n \"missing_y_mat, masks, weights = prepare_data_optimized_loss(x,y,t,masks, unique_treatments) assert(utility_weights.shape == (num_obs, y.shape[1]))\",\n \"uplift_model_propensity = MRUplift() uplift_model_propensity.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1, 1), n_jobs=1, param_grid =\",\n \"dict(num_nodes=[8], dropout=[.1], activation=['relu'], num_layers=[2], epochs=[30], batch_size=[100]) y, x, t =\",\n \"y.shape[1])) for q in range(unique_treatments.shape[0]): assert( ((missing_utility[:,q]==0) == (missing_y_mat[:,q,0] ==\",\n \"np.random.binomial(1, .5, num_obs).reshape(-1, 1)], axis=1) unique_treatments = np.unique(t, axis =\",\n \"= np.unique(t, axis = 0) masks = np.ones(num_obs).reshape(num_obs,1) x, utility_weights,\",\n \"get_simple_uplift_data(num_obs) t = t.reshape(len(t),1) unique_treatments = np.unique(t, axis = 0)\",\n \"< rmse_tolerance def test_model_get_random_erupts(self): true_ATE = np.array([[0, 0], [1, .5]])\",\n \"x_no_noise, tmt_no_noise = get_no_noise_data(num_obs) uplift_model = MRUplift() uplift_model.fit(x_no_noise, y_no_noise, tmt_no_noise.reshape(-1,\"\n]","isConversational":false}}},{"rowIdx":681,"cells":{"ngram":{"kind":"list like","value":["y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 =","from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\"","#TODO \"\"\" def test_increment(self): x0 = 0 y0 = increment(x0)","= 0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1)","\"\"\" #TODO \"\"\" def test_increment(self): x0 = 0 y0 =","== 1 self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1)","x1 = 100 y1 = increment(x1) #y1 == 101 self.assertEqual(y1,","example_module import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def","import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def test_increment(self):","ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def test_increment(self): x0 = 0 y0","\"\"\" \"\"\" import unittest from example_module import COLORS, increment class","\"\"\" import unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase):","0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1","class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def test_increment(self): x0 = 0","\"\"\" def test_increment(self): x0 = 0 y0 = increment(x0) #y0","self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1) #y1 ==","1) x1 = 100 y1 = increment(x1) #y1 == 101","x0 = 0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0,","test_increment(self): x0 = 0 y0 = increment(x0) #y0 == 1","def test_increment(self): x0 = 0 y0 = increment(x0) #y0 ==","increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def test_increment(self): x0 =","1 self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1) #y1","= increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 = 100","import unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\"","#y0 == 1 self.assertEqual(y0, 1) x1 = 100 y1 =","= 100 y1 = increment(x1) #y1 == 101 self.assertEqual(y1, 101)","increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 = 100 y1","unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO","COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\" def test_increment(self): x0"],"string":"[\n \"y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 =\",\n \"from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\"\",\n \"#TODO \\\"\\\"\\\" def test_increment(self): x0 = 0 y0 = increment(x0)\",\n \"= 0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1)\",\n \"\\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self): x0 = 0 y0 =\",\n \"== 1 self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1)\",\n \"x1 = 100 y1 = increment(x1) #y1 == 101 self.assertEqual(y1,\",\n \"example_module import COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def\",\n \"import COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self):\",\n \"ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self): x0 = 0 y0\",\n \"\\\"\\\"\\\" \\\"\\\"\\\" import unittest from example_module import COLORS, increment class\",\n \"\\\"\\\"\\\" import unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase):\",\n \"0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1\",\n \"class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self): x0 = 0\",\n \"\\\"\\\"\\\" def test_increment(self): x0 = 0 y0 = increment(x0) #y0\",\n \"self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1) #y1 ==\",\n \"1) x1 = 100 y1 = increment(x1) #y1 == 101\",\n \"x0 = 0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0,\",\n \"test_increment(self): x0 = 0 y0 = increment(x0) #y0 == 1\",\n \"def test_increment(self): x0 = 0 y0 = increment(x0) #y0 ==\",\n \"increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self): x0 =\",\n \"1 self.assertEqual(y0, 1) x1 = 100 y1 = increment(x1) #y1\",\n \"= increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 = 100\",\n \"import unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\"\",\n \"#y0 == 1 self.assertEqual(y0, 1) x1 = 100 y1 =\",\n \"= 100 y1 = increment(x1) #y1 == 101 self.assertEqual(y1, 101)\",\n \"increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 = 100 y1\",\n \"unittest from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO\",\n \"COLORS, increment class ExampleTest(unittest.TestCase): \\\"\\\"\\\" #TODO \\\"\\\"\\\" def test_increment(self): x0\"\n]","isConversational":false}}},{"rowIdx":682,"cells":{"ngram":{"kind":"list like","value":["from django.utils.translation import ugettext as _ from desktop.auth import forms","+ request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200': raise","is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in","exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user',","'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory","oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\",","current_users def first_login_ever(): backends = get_backends() for backend in backends:","this work for additional information # regarding copyright ownership. Cloudera,","if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever","more contributor license agreements. See the NOTICE file # distributed","import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary","response from OAuth provider: %s\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content))","request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to)","# For first login, need to validate user info! first_user_form","backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request, next_page) def","login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie()","profile(request): \"\"\" Dumps JSON for user-profile information. \"\"\" return render(None,","last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be converted email=user.email)","@login_notrequired def oauth_login(request): assert oauth is not None consumer =","2.0 (the # \"License\"); you may not use this file","None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in","based on Twitter as example. @login_notrequired def oauth_login(request): assert oauth","{ 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever':","from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate","\"\"\"Log out the user\"\"\" username = request.user.get_username() request.audit = {","cgi import logging import urllib from datetime import datetime from","of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless","regarding copyright ownership. Cloudera, Inc. licenses this file # to","userprofile.save() msg = 'Successful login for user: %s' % user.username","from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException","a dictionary of the user's IP address and last access","of the user's IP address and last access time\"\"\" current_users","not create home directory.'), exc_info=e) request.error(_('Could not create home directory.'))","object needs to be converted email=user.email) # OAuth is based","request.POST.get('username') } # For first login, need to validate user","'/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\"Invalid response from","from OAuth provider: %s\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url","get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever():","userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist:","not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path =","= authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if","or None first_user = first_user_form and first_user_form.is_valid() if first_user or","backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \"\"\"Log out the","if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username,","resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to","if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path","HttpResponseRedirect from django.utils.translation import ugettext as _ from desktop.auth import","import forms as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util","client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if","distributed with this work for additional information # regarding copyright","= oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer,","consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client =","address and last access time\"\"\" current_users = { } for","the user\"\"\" username = request.user.get_username() request.audit = { 'username': username,","= oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(),","user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if","request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For","access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return","'200': raise Exception(_(\"Invalid response from OAuth provider: %s\") % resp)","get_backends() if backends: for backend in backends: if hasattr(backend, 'logout'):","import User from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect","the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required","try: import oauth2 as oauth except: oauth = None import","additional information # regarding copyright ownership. Cloudera, Inc. licenses this","= AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user","userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login for user:","from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import User","django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import","try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create","= oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content =","'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username } backends","auto_create_home_backends): # Create home directory for first user. try: ensure_home_directory(request.fs,","under the License is distributed on an \"AS IS\" BASIS,","login, get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import","file # distributed with this work for additional information #","True return False def get_backend_names(): return get_backends and [backend.__class__.__name__ for","# Licensed to Cloudera, Inc. under one # or more","the AuthenticationForm. # It provides 'backends' on the User object.","License for the specific language governing permissions and # limitations","Create home directory for first user. try: ensure_home_directory(request.fs, user.username) except","@login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'],","== 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None):","and last access time\"\"\" current_users = { } for session","with id=%d does not exist\" % uid) return current_users def","ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary of","authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import Session from","auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm =","urllib from datetime import datetime from axes.decorators import watch_login import","= session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this resolution","'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] =","bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm","from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from useradmin.views","request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or","useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG =","the License. try: import oauth2 as oauth except: oauth =","if is_first_login_ever or any(backend in backend_names for backend in auto_create_home_backends):","if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True})","import logging import urllib from datetime import datetime from axes.decorators","UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid() if first_user","request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return","to you under the Apache License, Version 2.0 (the #","OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({","uid) return current_users def first_login_ever(): backends = get_backends() for backend","on the User object. user = auth_form.get_user() userprofile = get_profile(user)","userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends =","object. user = auth_form.get_user() userprofile = get_profile(user) login(request, user) if","password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal:","= 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request,","Licensed to Cloudera, Inc. under one # or more contributor","or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm =","information. \"\"\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict(","\"GET\") if resp['status'] != '200': raise Exception(_(\"Invalid response from OAuth","may not use this file except in compliance # with","desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import","software # distributed under the License is distributed on an","{ } for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY)","governing permissions and # limitations under the License. try: import","backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names and (","not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login","if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False def","== 'true': return JsonResponse({'auth': False}) else: first_user_form = None auth_form","first_login_ever(): backends = get_backends() for backend in backends: if hasattr(backend,","License, Version 2.0 (the # \"License\"); you may not use","backend in auto_create_home_backends): # Create home directory for first user.","for user-profile information. \"\"\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user):","is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm =","LOG.debug(\"User with id=%d does not exist\" % uid) return current_users","else: first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and","is based on Twitter as example. @login_notrequired def oauth_login(request): assert","of User objects and a dictionary of the user's IP","first_user = first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever:","def get_current_users(): \"\"\"Return dictionary of User objects and a dictionary","and a dictionary of the user's IP address and last","is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get())","else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed","# datetime object needs to be converted email=user.email) # OAuth","datetime from axes.decorators import watch_login import django.contrib.auth.views from django.core import","using the AuthenticationForm. # It provides 'backends' on the User","'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, {","user\"\"\" username = request.user.get_username() request.audit = { 'username': username, 'operation':","backend_names for backend in auto_create_home_backends): # Create home directory for","watch_login import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import","current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\"User with id=%d","OAuth provider: %s\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url =","See the NOTICE file # distributed with this work for","backend.is_first_login_ever(): return True return False def get_backend_names(): return get_backends and","user-profile information. \"\"\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return","to validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or","%s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal","import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn,","desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from","ugettext as _ from desktop.auth import forms as auth_forms from","response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\" Dumps","django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\" Dumps JSON for user-profile information.","+ '/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\"Invalid response","Apache License, Version 2.0 (the # \"License\"); you may not","import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__)","get_current_users(): \"\"\"Return dictionary of User objects and a dictionary of","backends: for backend in backends: if hasattr(backend, 'logout'): response =","from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako'","try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed,","OF ANY KIND, either express or implied. # See the","WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.","request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend","response = backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request,","ownership. Cloudera, Inc. licenses this file # to you under","ANY KIND, either express or implied. # See the License","See the License for the specific language governing permissions and","oauth = None import cgi import logging import urllib from","return django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\" Dumps JSON for user-profile","at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable","for the specific language governing permissions and # limitations under","secret_key changed, this resolution won't work. uid = None if","file except in compliance # with the License. You may","to in writing, software # distributed under the License is","IP address and last access time\"\"\" current_users = { }","content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() +","# See the License for the specific language governing permissions","next_page) if response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request):","import datetime from axes.decorators import watch_login import django.contrib.auth.views from django.core","dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names","for user: %s' % user.username request.audit['operationText'] = msg access_warn(request, msg)","or agreed to in writing, software # distributed under the","you may not use this file except in compliance #","required by applicable law or agreed to in writing, software","def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret'])","% request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal or","work. uid = None if uid is not None: try:","try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except","BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either","last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\"User with id=%d does not","'true': return JsonResponse({'auth': False}) else: first_user_form = None auth_form =","use this file except in compliance # with the License.","return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs","if resp['status'] != '200': raise Exception(_(\"Invalid response from OAuth provider:","or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else: return","contributor license agreements. See the NOTICE file # distributed with","changed, this resolution won't work. uid = None if uid","user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/') return","first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form =","%s' % user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal","import JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf import","_profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object","= get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend',","for backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to","hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from useradmin.views import","= False msg = 'Failed login for user: %s' %","agreed to in writing, software # distributed under the License","[backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False):","or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST',","%s\") % resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request,","} for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except","provider: %s\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\"","exist\" % uid) return current_users def first_login_ever(): backends = get_backends()","is_active_directory }) def dt_logout(request, next_page=None): \"\"\"Log out the user\"\"\" username","token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content","in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If","distributed under the License is distributed on an \"AS IS\"","create home directory.'), exc_info=e) request.error(_('Could not create home directory.')) if","'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \"\"\"Log","{ }) except User.DoesNotExist: LOG.debug(\"User with id=%d does not exist\"","with this work for additional information # regarding copyright ownership.","except: oauth = None import cgi import logging import urllib","= oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://'","first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method ==","Exception(_(\"Invalid response from OAuth provider: %s\") % resp) access_token =","User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\"User with","is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer)","express or implied. # See the License for the specific","% user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal or","}) except User.DoesNotExist: LOG.debug(\"User with id=%d does not exist\" %","import watch_login import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions","email=user.email) # OAuth is based on Twitter as example. @login_notrequired","from OAuth provider: %s\") % resp) access_token = dict(cgi.parse_qsl(content)) user","from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from django.utils.translation","user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user","desktop.auth import forms as auth_forms from desktop.lib.django_util import render from","import WebHdfsException from useradmin.models import get_profile from useradmin.views import ensure_home_directory,","raise Exception(_(\"Invalid response from OAuth provider: %s\") % resp) request.session['request_token']","example. @login_notrequired def oauth_login(request): assert oauth is not None consumer","= get_backends() if backends: for backend in backends: if hasattr(backend,","writing, software # distributed under the License is distributed on","return current_users def first_login_ever(): backends = get_backends() for backend in","directory.'), exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile): return","consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content =","the License. You may obtain a copy of the License","under the License. try: import oauth2 as oauth except: oauth","import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login,","AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user =","DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>')","\"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if","OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile","\"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer","forms as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import","in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return","auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends","DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from","under the Apache License, Version 2.0 (the # \"License\"); you","as oauth except: oauth = None import cgi import logging","if uid is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj]","AuthenticationForm. # It provides 'backends' on the User object. user","request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged","CONDITIONS OF ANY KIND, either express or implied. # See","require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary of User","last access time\"\"\" current_users = { } for session in","return False def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend","is distributed on an \"AS IS\" BASIS, # WITHOUT WARRANTIES","need to validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST)","hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response: return response","resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host()","from axes.decorators import watch_login import django.contrib.auth.views from django.core import urlresolvers","= 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) )","kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg","django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import User from","body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status']","for user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg)","e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not create","get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import Session","datetime.now() userprofile.save() msg = 'Successful login for user: %s' %","validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None","request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form =","or more contributor license agreements. See the NOTICE file #","first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be converted","'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' %","'Failed login for user: %s' % request.POST.get('username') request.audit['operationText'] = msg","first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names and","from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users():","dt_logout(request, next_page=None): \"\"\"Log out the user\"\"\" username = request.user.get_username() request.audit","_ from desktop.auth import forms as auth_forms from desktop.lib.django_util import","= datetime.now() userprofile.save() msg = 'Successful login for user: %s'","auth_form.is_valid(): # Must login by using the AuthenticationForm. # It","JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg","the NOTICE file # distributed with this work for additional","'backends' on the User object. user = auth_form.get_user() userprofile =","_profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), #","AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit = {","login by using the AuthenticationForm. # It provides 'backends' on","except (IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e)","OR CONDITIONS OF ANY KIND, either express or implied. #","not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp,","LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import","this resolution won't work. uid = None if uid is","= get_backend_names() is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get())","the License is distributed on an \"AS IS\" BASIS, #","objects and a dictionary of the user's IP address and","if backends: for backend in backends: if hasattr(backend, 'logout'): response","return JsonResponse({'auth': False}) else: first_user_form = None auth_form = AuthenticationForm()","user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'),","render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from","# If secret_key changed, this resolution won't work. uid =","is_first_login_ever or any(backend in backend_names for backend in auto_create_home_backends): #","HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed login for","False}) else: first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get()","session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this resolution won't","'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if","user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako'","from useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG","backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True","!= '200': raise Exception(_(\"Invalid response from OAuth provider: %s\") %","out the user\"\"\" username = request.user.get_username() request.audit = { 'username':","client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' }))","login for user: %s' % user.username request.audit['operationText'] = msg access_warn(request,","from datetime import datetime from axes.decorators import watch_login import django.contrib.auth.views","request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return","'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \"\"\"Log out","be converted email=user.email) # OAuth is based on Twitter as","'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username }","\"\"\" Dumps JSON for user-profile information. \"\"\" return render(None, request,","})) if resp['status'] != '200': raise Exception(_(\"Invalid response from OAuth","'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200':","law or agreed to in writing, software # distributed under","def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in get_backends()]","directory for first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException),","get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend',","urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends,","from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth","oauth_login(request): assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get())","access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to =","request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false')","= backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request, next_page)","# limitations under the License. try: import oauth2 as oauth","= first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names","ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path =","def profile(request): \"\"\" Dumps JSON for user-profile information. \"\"\" return","client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] != '200': raise Exception(_(\"Invalid response from","create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login","# regarding copyright ownership. Cloudera, Inc. licenses this file #","import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation","home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login =","is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username,","request.error(_('Could not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username':","SuspiciousOperation: # If secret_key changed, this resolution won't work. uid","home directory for first user. try: ensure_home_directory(request.fs, user.username) except (IOError,","False msg = 'Failed login for user: %s' % request.POST.get('username')","= ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names","may obtain a copy of the License at # #","user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal',","auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names':","'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first login, need","username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username","request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory =","user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not","url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def","= auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm","IS\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,","'/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend'","on Twitter as example. @login_notrequired def oauth_login(request): assert oauth is","raise Exception(_(\"Invalid response from OAuth provider: %s\") % resp) access_token","request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal',","else: request.audit['allowed'] = False msg = 'Failed login for user:","'logout'): response = backend.logout(request, next_page) if response: return response return","dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired","WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or","= auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit","Session from django.http import HttpResponseRedirect from django.utils.translation import ugettext as","AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm","user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if","User.DoesNotExist: LOG.debug(\"User with id=%d does not exist\" % uid) return","for additional information # regarding copyright ownership. Cloudera, Inc. licenses","{ 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s'","resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token'])","# # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law","if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid():","JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP,","on an \"AS IS\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS","first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form","user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to)","Inc. under one # or more contributor license agreements. See","# Must login by using the AuthenticationForm. # It provides","first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if","= None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin'","def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever()","first login, need to validate user info! first_user_form = is_first_login_ever","= auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie()","in compliance # with the License. You may obtain a","# to you under the Apache License, Version 2.0 (the","redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names()","in backend_names for backend in auto_create_home_backends): # Create home directory","JsonResponse({'auth': False}) else: first_user_form = None auth_form = AuthenticationForm() if","renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path,","hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False def get_backend_names():","from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util","django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from","# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express","dictionary of the user's IP address and last access time\"\"\"","render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next':","if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action':","= 'Failed login for user: %s' % request.POST.get('username') request.audit['operationText'] =","auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using","Exception(_(\"Invalid response from OAuth provider: %s\") % resp) request.session['request_token'] =","if auth_form.is_valid(): # Must login by using the AuthenticationForm. #","'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names,","this file # to you under the Apache License, Version","'Successful login for user: %s' % user.username request.audit['operationText'] = msg","home directory.'), exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile):","= request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory","does not exist\" % uid) return current_users def first_login_ever(): backends","% uid) return current_users def first_login_ever(): backends = get_backends() for","response from OAuth provider: %s\") % resp) access_token = dict(cgi.parse_qsl(content))","import ugettext as _ from desktop.auth import forms as auth_forms",") if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else:","needs to be converted email=user.email) # OAuth is based on","for backend in auto_create_home_backends): # Create home directory for first","{ 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first login,","get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def","%s\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" %","return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now()","@login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever","request.audit['allowed'] = False msg = 'Failed login for user: %s'","dictionary of User objects and a dictionary of the user's","= AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using the","auth_forms.AuthenticationForm if request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN',","http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed","OAuth is based on Twitter as example. @login_notrequired def oauth_login(request):","= \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request):","Twitter as example. @login_notrequired def oauth_login(request): assert oauth is not","oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(),","renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form':","auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method","in auto_create_home_backends): # Create home directory for first user. try:","WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not","'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever,","backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response:","% username } backends = get_backends() if backends: for backend","get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/')","from django.contrib.auth.models import User from django.contrib.sessions.models import Session from django.http","or implied. # See the License for the specific language","for backend in backends: if hasattr(backend, 'logout'): response = backend.logout(request,","require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity =","else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method ==","information # regarding copyright ownership. Cloudera, Inc. licenses this file","request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request,","the user's IP address and last access time\"\"\" current_users =","and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm =","# \"License\"); you may not use this file except in","KIND, either express or implied. # See the License for","specific language governing permissions and # limitations under the License.","import Session from django.http import HttpResponseRedirect from django.utils.translation import ugettext","current_users = { } for session in Session.objects.all(): try: uid","UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST':","= 'Successful login for user: %s' % user.username request.audit['operationText'] =","\"License\"); you may not use this file except in compliance","from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else:","python # Licensed to Cloudera, Inc. under one # or","oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token)","License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by","user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path","None first_user = first_user_form and first_user_form.is_valid() if first_user or not","django.http import HttpResponseRedirect from django.utils.translation import ugettext as _ from","msg = 'Successful login for user: %s' % user.username request.audit['operationText']","'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method","# distributed with this work for additional information # regarding","auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from","ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create home","def first_login_ever(): backends = get_backends() for backend in backends: if","converted email=user.email) # OAuth is based on Twitter as example.","@watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever =","first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could","permissions and # limitations under the License. try: import oauth2","and UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid() if","user: %s' % user.username request.audit['operationText'] = msg access_warn(request, msg) if","= oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status']","%s' % username } backends = get_backends() if backends: for","= is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form and","= { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user:","request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to,","License. try: import oauth2 as oauth except: oauth = None","Inc. licenses this file # to you under the Apache","msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true':","import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models","request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\")","def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime","in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory:","return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False","is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid()","= 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form","= auth_forms.AuthenticationForm if request.method == 'POST': request.audit = { 'operation':","desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED","request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\"Invalid","# # Unless required by applicable law or agreed to","Version 2.0 (the # \"License\"); you may not use this","except User.DoesNotExist: LOG.debug(\"User with id=%d does not exist\" % uid)","= request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText':","'Logged out user: %s' % username } backends = get_backends()","client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback':","backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False","one # or more contributor license agreements. See the NOTICE","obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0","Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key","% resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(),","SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import","or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must","last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import","directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False","login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend']","return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form,","% resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user)","OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp,","except in compliance # with the License. You may obtain","= None if uid is not None: try: userobj =","import cgi import logging import urllib from datetime import datetime","and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST)","\"\"\"Return dictionary of User objects and a dictionary of the","return get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login","from django.http import HttpResponseRedirect from django.utils.translation import ugettext as _","= client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] != '200': raise Exception(_(\"Invalid response","implied. # See the License for the specific language governing","oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] !=","NOTICE file # distributed with this work for additional information","limitations under the License. try: import oauth2 as oauth except:","False def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in","For first login, need to validate user info! first_user_form =","Dumps JSON for user-profile information. \"\"\" return render(None, request, _profile_dict(request.user))","'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } #","this file except in compliance # with the License. You","if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False})","return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if","from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'),","id=%d does not exist\" % uid) return current_users def first_login_ever():","return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token","UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm","from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map from","not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username}))","not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user)","\"\"\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username,","datetime object needs to be converted email=user.email) # OAuth is","resp['status'] != '200': raise Exception(_(\"Invalid response from OAuth provider: %s\")","as example. @login_notrequired def oauth_login(request): assert oauth is not None","license agreements. See the NOTICE file # distributed with this","user's IP address and last access time\"\"\" current_users = {","by applicable law or agreed to in writing, software #","resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] != '200': raise","= User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\"User","first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): #","useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \"\"\"Return","content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] != '200': raise Exception(_(\"Invalid","True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg =","#!/usr/bin/env python # Licensed to Cloudera, Inc. under one #","and backend.is_first_login_ever(): return True return False def get_backend_names(): return get_backends","to be converted email=user.email) # OAuth is based on Twitter","if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response: return","User from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from","backends = get_backends() if backends: for backend in backends: if","oauth except: oauth = None import cgi import logging import","HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save()","provides 'backends' on the User object. user = auth_form.get_user() userprofile","'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or","request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return","from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access","and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def dt_login(request,","OAuth provider: %s\") % resp) access_token = dict(cgi.parse_qsl(content)) user =","access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions","not use this file except in compliance # with the","request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out","user = auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked():","= logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary of User objects and","logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary of User objects and a","'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for backend in","an \"AS IS\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF","WebHdfsException from useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password","for first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e:","return True return False def get_backend_names(): return get_backends and [backend.__class__.__name__","Unless required by applicable law or agreed to in writing,","django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from django.utils.translation import","auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit =","auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST:","Cloudera, Inc. licenses this file # to you under the","HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal:","None if uid is not None: try: userobj = User.objects.get(pk=uid)","the specific language governing permissions and # limitations under the","Must login by using the AuthenticationForm. # It provides 'backends'","# or more contributor license agreements. See the NOTICE file","time\"\"\" current_users = { } for session in Session.objects.all(): try:","if response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\"","and # limitations under the License. try: import oauth2 as","= msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') ==","% (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer =","applicable law or agreed to in writing, software # distributed","in backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if","userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful","'username': request.POST.get('username') } # For first login, need to validate","not exist\" % uid) return current_users def first_login_ever(): backends =","None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content","datetime import datetime from axes.decorators import watch_login import django.contrib.auth.views from","next_page) def profile(request): \"\"\" Dumps JSON for user-profile information. \"\"\"","uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this","== 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') }","= dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next',","in writing, software # distributed under the License is distributed","import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models","['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for","(the # \"License\"); you may not use this file except","= dict(cgi.parse_qsl(content)) url = \"%s?oauth_token=%s\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url)","HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token =","desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import","backend in backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page)","} backends = get_backends() if backends: for backend in backends:","= client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/'","won't work. uid = None if uid is not None:","from desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP, OAUTH,","= { } for session in Session.objects.all(): try: uid =","def dt_logout(request, next_page=None): \"\"\"Log out the user\"\"\" username = request.user.get_username()","as _ from desktop.auth import forms as auth_forms from desktop.lib.django_util","username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be","resolution won't work. uid = None if uid is not","is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by","msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth':","for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation:","next_page=None): \"\"\"Log out the user\"\"\" username = request.user.get_username() request.audit =","LOG = logging.getLogger(__name__) def get_current_users(): \"\"\"Return dictionary of User objects","uid = None if uid is not None: try: userobj","JSON for user-profile information. \"\"\" return render(None, request, _profile_dict(request.user)) def","render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name,","(OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(),","License is distributed on an \"AS IS\" BASIS, # WITHOUT","not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, {","} # For first login, need to validate user info!","oauth2 as oauth except: oauth = None import cgi import","oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client","import access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from","None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { })","backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm","user: %s' % username } backends = get_backends() if backends:","LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not create home","get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired","info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user =","False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login for","copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # #","return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed login","'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \"\"\"Log out the user\"\"\"","with the License. You may obtain a copy of the","'USER_LOGIN', 'username': request.POST.get('username') } # For first login, need to","login, need to validate user info! first_user_form = is_first_login_ever and","login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map","import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse","from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else:","first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not","AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using the AuthenticationForm.","except SuspiciousOperation: # If secret_key changed, this resolution won't work.","urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors':","token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \"GET\") if resp['status'] != '200':","dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/')","django.contrib.auth.models import User from django.contrib.sessions.models import Session from django.http import","auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in","the License for the specific language governing permissions and #","'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def","out user: %s' % username } backends = get_backends() if","dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to","'is_first_login_ever') and backend.is_first_login_ever(): return True return False def get_backend_names(): return","msg = 'Failed login for user: %s' % request.POST.get('username') request.audit['operationText']","by using the AuthenticationForm. # It provides 'backends' on the","file # to you under the Apache License, Version 2.0","either express or implied. # See the License for the","login for user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request,","# with the License. You may obtain a copy of","redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory':","# http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or","'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] !=","the User object. user = auth_form.get_user() userprofile = get_profile(user) login(request,","import oauth2 as oauth except: oauth = None import cgi","assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client","from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names =","in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username)","If secret_key changed, this resolution won't work. uid = None","logging import urllib from datetime import datetime from axes.decorators import","copyright ownership. Cloudera, Inc. licenses this file # to you","under one # or more contributor license agreements. See the","username = request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT',","or any(backend in backend_names for backend in auto_create_home_backends): # Create","a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 #","Cloudera, Inc. under one # or more contributor license agreements.","import login, get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models","request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get())","axes.decorators import watch_login import django.contrib.auth.views from django.core import urlresolvers from","if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity","for backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return","None import cgi import logging import urllib from datetime import","# Create home directory for first user. try: ensure_home_directory(request.fs, user.username)","to Cloudera, Inc. under one # or more contributor license","the Apache License, Version 2.0 (the # \"License\"); you may","= first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form","user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg =","is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory })","def oauth_login(request): assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(),","oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client =","request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login),","desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf","authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not","import urllib from datetime import datetime from axes.decorators import watch_login","you under the Apache License, Version 2.0 (the # \"License\");","from desktop.auth import forms as auth_forms from desktop.lib.django_util import render","'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs,","if request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username':","session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: #","language governing permissions and # limitations under the License. try:","return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name,","# It provides 'backends' on the User object. user =","'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for backend","and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request,","agreements. See the NOTICE file # distributed with this work","distributed on an \"AS IS\" BASIS, # WITHOUT WARRANTIES OR","response return django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\" Dumps JSON for","licenses this file # to you under the Apache License,","= False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login","get_backend_names() is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or","return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \"\"\" Dumps JSON","backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to =","# distributed under the License is distributed on an \"AS","'false') == 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else:","# OAuth is based on Twitter as example. @login_notrequired def","work for additional information # regarding copyright ownership. Cloudera, Inc.","# Unless required by applicable law or agreed to in","uid is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] =","= last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\"User with id=%d does","= { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first","get_profile from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def","bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm","backends = get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever')","'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form = None","}) def dt_logout(request, next_page=None): \"\"\"Log out the user\"\"\" username =","access time\"\"\" current_users = { } for session in Session.objects.all():","\"AS IS\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY","provider: %s\") % resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token)","or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form","= authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/') return HttpResponseRedirect(redirect_to)","oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \"POST\", body=urllib.urlencode({ 'oauth_callback': 'http://' +","last_login=str(user.last_login), # datetime object needs to be converted email=user.email) #","It provides 'backends' on the User object. user = auth_form.get_user()","License. You may obtain a copy of the License at","User objects and a dictionary of the user's IP address","request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username')","any(backend in backend_names for backend in auto_create_home_backends): # Create home","You may obtain a copy of the License at #","django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from","User object. user = auth_form.get_user() userprofile = get_profile(user) login(request, user)","import HttpResponseRedirect from django.utils.translation import ugettext as _ from desktop.auth","as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired","(IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could","( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm","username } backends = get_backends() if backends: for backend in","== 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed']","'operationText': 'Logged out user: %s' % username } backends =","compliance # with the License. You may obtain a copy","= auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if","in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next',","if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm","= None import cgi import logging import urllib from datetime","= get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever') and","django.utils.translation import ugettext as _ from desktop.auth import forms as"],"string":"[\n \"from django.utils.translation import ugettext as _ from desktop.auth import forms\",\n \"+ request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200': raise\",\n \"is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in\",\n \"exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user',\",\n \"'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory\",\n \"oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\",\",\n \"current_users def first_login_ever(): backends = get_backends() for backend in backends:\",\n \"this work for additional information # regarding copyright ownership. Cloudera,\",\n \"if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever\",\n \"more contributor license agreements. See the NOTICE file # distributed\",\n \"import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary\",\n \"response from OAuth provider: %s\\\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content))\",\n \"request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to)\",\n \"# For first login, need to validate user info! first_user_form\",\n \"backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request, next_page) def\",\n \"login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie()\",\n \"profile(request): \\\"\\\"\\\" Dumps JSON for user-profile information. \\\"\\\"\\\" return render(None,\",\n \"last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be converted email=user.email)\",\n \"@login_notrequired def oauth_login(request): assert oauth is not None consumer =\",\n \"2.0 (the # \\\"License\\\"); you may not use this file\",\n \"None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in\",\n \"based on Twitter as example. @login_notrequired def oauth_login(request): assert oauth\",\n \"{ 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever':\",\n \"from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate\",\n \"\\\"\\\"\\\"Log out the user\\\"\\\"\\\" username = request.user.get_username() request.audit = {\",\n \"cgi import logging import urllib from datetime import datetime from\",\n \"of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless\",\n \"regarding copyright ownership. Cloudera, Inc. licenses this file # to\",\n \"userprofile.save() msg = 'Successful login for user: %s' % user.username\",\n \"from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException\",\n \"a dictionary of the user's IP address and last access\",\n \"of the user's IP address and last access time\\\"\\\"\\\" current_users\",\n \"not create home directory.'), exc_info=e) request.error(_('Could not create home directory.'))\",\n \"object needs to be converted email=user.email) # OAuth is based\",\n \"request.POST.get('username') } # For first login, need to validate user\",\n \"'/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\\\"Invalid response from\",\n \"from OAuth provider: %s\\\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url\",\n \"get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever():\",\n \"userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist:\",\n \"not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path =\",\n \"= authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if\",\n \"or None first_user = first_user_form and first_user_form.is_valid() if first_user or\",\n \"backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the\",\n \"if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username,\",\n \"resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to\",\n \"if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path\",\n \"HttpResponseRedirect from django.utils.translation import ugettext as _ from desktop.auth import\",\n \"import forms as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util\",\n \"client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if\",\n \"distributed with this work for additional information # regarding copyright\",\n \"= oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer,\",\n \"consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client =\",\n \"address and last access time\\\"\\\"\\\" current_users = { } for\",\n \"the user\\\"\\\"\\\" username = request.user.get_username() request.audit = { 'username': username,\",\n \"= oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(),\",\n \"user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if\",\n \"request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For\",\n \"access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return\",\n \"'200': raise Exception(_(\\\"Invalid response from OAuth provider: %s\\\") % resp)\",\n \"get_backends() if backends: for backend in backends: if hasattr(backend, 'logout'):\",\n \"import User from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect\",\n \"the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required\",\n \"try: import oauth2 as oauth except: oauth = None import\",\n \"additional information # regarding copyright ownership. Cloudera, Inc. licenses this\",\n \"= AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user\",\n \"userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login for user:\",\n \"from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import User\",\n \"django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import\",\n \"try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create\",\n \"= oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content =\",\n \"'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username } backends\",\n \"auto_create_home_backends): # Create home directory for first user. try: ensure_home_directory(request.fs,\",\n \"under the License is distributed on an \\\"AS IS\\\" BASIS,\",\n \"login, get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import\",\n \"file # distributed with this work for additional information #\",\n \"True return False def get_backend_names(): return get_backends and [backend.__class__.__name__ for\",\n \"# Licensed to Cloudera, Inc. under one # or more\",\n \"the AuthenticationForm. # It provides 'backends' on the User object.\",\n \"License for the specific language governing permissions and # limitations\",\n \"Create home directory for first user. try: ensure_home_directory(request.fs, user.username) except\",\n \"@login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'],\",\n \"== 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None):\",\n \"and last access time\\\"\\\"\\\" current_users = { } for session\",\n \"with id=%d does not exist\\\" % uid) return current_users def\",\n \"ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary of\",\n \"authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import Session from\",\n \"auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm =\",\n \"urllib from datetime import datetime from axes.decorators import watch_login import\",\n \"= session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this resolution\",\n \"'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] =\",\n \"bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm\",\n \"from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from useradmin.views\",\n \"request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or\",\n \"useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG =\",\n \"the License. try: import oauth2 as oauth except: oauth =\",\n \"if is_first_login_ever or any(backend in backend_names for backend in auto_create_home_backends):\",\n \"if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True})\",\n \"import logging import urllib from datetime import datetime from axes.decorators\",\n \"UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid() if first_user\",\n \"request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return\",\n \"to you under the Apache License, Version 2.0 (the #\",\n \"OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({\",\n \"uid) return current_users def first_login_ever(): backends = get_backends() for backend\",\n \"on the User object. user = auth_form.get_user() userprofile = get_profile(user)\",\n \"userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends =\",\n \"object. user = auth_form.get_user() userprofile = get_profile(user) login(request, user) if\",\n \"password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal:\",\n \"= 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request,\",\n \"Licensed to Cloudera, Inc. under one # or more contributor\",\n \"or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm =\",\n \"information. \\\"\\\"\\\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict(\",\n \"\\\"GET\\\") if resp['status'] != '200': raise Exception(_(\\\"Invalid response from OAuth\",\n \"may not use this file except in compliance # with\",\n \"desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import\",\n \"software # distributed under the License is distributed on an\",\n \"{ } for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY)\",\n \"governing permissions and # limitations under the License. try: import\",\n \"backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names and (\",\n \"not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login\",\n \"if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False def\",\n \"== 'true': return JsonResponse({'auth': False}) else: first_user_form = None auth_form\",\n \"first_login_ever(): backends = get_backends() for backend in backends: if hasattr(backend,\",\n \"License, Version 2.0 (the # \\\"License\\\"); you may not use\",\n \"backend in auto_create_home_backends): # Create home directory for first user.\",\n \"for user-profile information. \\\"\\\"\\\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user):\",\n \"is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm =\",\n \"LOG.debug(\\\"User with id=%d does not exist\\\" % uid) return current_users\",\n \"else: first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and\",\n \"is based on Twitter as example. @login_notrequired def oauth_login(request): assert\",\n \"of User objects and a dictionary of the user's IP\",\n \"first_user = first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever:\",\n \"def get_current_users(): \\\"\\\"\\\"Return dictionary of User objects and a dictionary\",\n \"and a dictionary of the user's IP address and last\",\n \"is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get())\",\n \"else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed\",\n \"# datetime object needs to be converted email=user.email) # OAuth\",\n \"datetime from axes.decorators import watch_login import django.contrib.auth.views from django.core import\",\n \"using the AuthenticationForm. # It provides 'backends' on the User\",\n \"'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, {\",\n \"user\\\"\\\"\\\" username = request.user.get_username() request.audit = { 'username': username, 'operation':\",\n \"backend_names for backend in auto_create_home_backends): # Create home directory for\",\n \"watch_login import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import\",\n \"current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\\\"User with id=%d\",\n \"OAuth provider: %s\\\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url =\",\n \"See the NOTICE file # distributed with this work for\",\n \"backend.is_first_login_ever(): return True return False def get_backend_names(): return get_backends and\",\n \"user-profile information. \\\"\\\"\\\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return\",\n \"to validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or\",\n \"%s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal\",\n \"import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn,\",\n \"desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from\",\n \"ugettext as _ from desktop.auth import forms as auth_forms from\",\n \"response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\" Dumps\",\n \"django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\" Dumps JSON for user-profile information.\",\n \"+ '/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\\\"Invalid response\",\n \"Apache License, Version 2.0 (the # \\\"License\\\"); you may not\",\n \"import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__)\",\n \"get_current_users(): \\\"\\\"\\\"Return dictionary of User objects and a dictionary of\",\n \"backends: for backend in backends: if hasattr(backend, 'logout'): response =\",\n \"from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako'\",\n \"try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed,\",\n \"OF ANY KIND, either express or implied. # See the\",\n \"WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\",\n \"request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend\",\n \"response = backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request,\",\n \"ownership. Cloudera, Inc. licenses this file # to you under\",\n \"ANY KIND, either express or implied. # See the License\",\n \"See the License for the specific language governing permissions and\",\n \"oauth = None import cgi import logging import urllib from\",\n \"return django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\" Dumps JSON for user-profile\",\n \"at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable\",\n \"for the specific language governing permissions and # limitations under\",\n \"secret_key changed, this resolution won't work. uid = None if\",\n \"file except in compliance # with the License. You may\",\n \"to in writing, software # distributed under the License is\",\n \"IP address and last access time\\\"\\\"\\\" current_users = { }\",\n \"content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() +\",\n \"# See the License for the specific language governing permissions\",\n \"next_page) if response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request):\",\n \"import datetime from axes.decorators import watch_login import django.contrib.auth.views from django.core\",\n \"dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names\",\n \"for user: %s' % user.username request.audit['operationText'] = msg access_warn(request, msg)\",\n \"or agreed to in writing, software # distributed under the\",\n \"you may not use this file except in compliance #\",\n \"required by applicable law or agreed to in writing, software\",\n \"def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret'])\",\n \"% request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal or\",\n \"work. uid = None if uid is not None: try:\",\n \"try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except\",\n \"BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either\",\n \"last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\\\"User with id=%d does not\",\n \"'true': return JsonResponse({'auth': False}) else: first_user_form = None auth_form =\",\n \"use this file except in compliance # with the License.\",\n \"return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs\",\n \"if resp['status'] != '200': raise Exception(_(\\\"Invalid response from OAuth provider:\",\n \"or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else: return\",\n \"contributor license agreements. See the NOTICE file # distributed with\",\n \"changed, this resolution won't work. uid = None if uid\",\n \"user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/') return\",\n \"first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form =\",\n \"%s' % user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal\",\n \"import JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf import\",\n \"_profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object\",\n \"= get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend',\",\n \"for backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to\",\n \"hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from useradmin.views import\",\n \"= False msg = 'Failed login for user: %s' %\",\n \"agreed to in writing, software # distributed under the License\",\n \"[backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False):\",\n \"or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST',\",\n \"%s\\\") % resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request,\",\n \"} for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except\",\n \"provider: %s\\\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\"\",\n \"exist\\\" % uid) return current_users def first_login_ever(): backends = get_backends()\",\n \"is_active_directory }) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\" username\",\n \"token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content\",\n \"in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If\",\n \"distributed under the License is distributed on an \\\"AS IS\\\"\",\n \"create home directory.'), exc_info=e) request.error(_('Could not create home directory.')) if\",\n \"'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log\",\n \"{ }) except User.DoesNotExist: LOG.debug(\\\"User with id=%d does not exist\\\"\",\n \"with this work for additional information # regarding copyright ownership.\",\n \"except: oauth = None import cgi import logging import urllib\",\n \"= oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://'\",\n \"first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method ==\",\n \"Exception(_(\\\"Invalid response from OAuth provider: %s\\\") % resp) access_token =\",\n \"User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\\\"User with\",\n \"is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer)\",\n \"express or implied. # See the License for the specific\",\n \"% user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal or\",\n \"}) except User.DoesNotExist: LOG.debug(\\\"User with id=%d does not exist\\\" %\",\n \"import watch_login import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions\",\n \"email=user.email) # OAuth is based on Twitter as example. @login_notrequired\",\n \"from OAuth provider: %s\\\") % resp) access_token = dict(cgi.parse_qsl(content)) user\",\n \"from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from django.utils.translation\",\n \"user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user\",\n \"desktop.auth import forms as auth_forms from desktop.lib.django_util import render from\",\n \"import WebHdfsException from useradmin.models import get_profile from useradmin.views import ensure_home_directory,\",\n \"raise Exception(_(\\\"Invalid response from OAuth provider: %s\\\") % resp) request.session['request_token']\",\n \"example. @login_notrequired def oauth_login(request): assert oauth is not None consumer\",\n \"= get_backends() if backends: for backend in backends: if hasattr(backend,\",\n \"writing, software # distributed under the License is distributed on\",\n \"return current_users def first_login_ever(): backends = get_backends() for backend in\",\n \"directory.'), exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile): return\",\n \"consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content =\",\n \"the License. You may obtain a copy of the License\",\n \"under the License. try: import oauth2 as oauth except: oauth\",\n \"import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login,\",\n \"AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user =\",\n \"DEMO_ENABLED.get() and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>')\",\n \"\\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if\",\n \"OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile\",\n \"\\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer\",\n \"forms as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import\",\n \"in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return\",\n \"auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends\",\n \"DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import get_profile from\",\n \"under the Apache License, Version 2.0 (the # \\\"License\\\"); you\",\n \"as oauth except: oauth = None import cgi import logging\",\n \"if uid is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj]\",\n \"AuthenticationForm. # It provides 'backends' on the User object. user\",\n \"request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged\",\n \"CONDITIONS OF ANY KIND, either express or implied. # See\",\n \"require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary of User\",\n \"last access time\\\"\\\"\\\" current_users = { } for session in\",\n \"return False def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend\",\n \"is distributed on an \\\"AS IS\\\" BASIS, # WITHOUT WARRANTIES\",\n \"need to validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST)\",\n \"hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response: return response\",\n \"resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host()\",\n \"from axes.decorators import watch_login import django.contrib.auth.views from django.core import urlresolvers\",\n \"= 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) )\",\n \"kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg\",\n \"django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import User from\",\n \"body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status']\",\n \"for user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg)\",\n \"e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not create\",\n \"get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models import Session\",\n \"datetime.now() userprofile.save() msg = 'Successful login for user: %s' %\",\n \"validate user info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None\",\n \"request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form =\",\n \"or more contributor license agreements. See the NOTICE file #\",\n \"first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be converted\",\n \"'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' %\",\n \"'Failed login for user: %s' % request.POST.get('username') request.audit['operationText'] = msg\",\n \"first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names and\",\n \"from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users():\",\n \"dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\" username = request.user.get_username() request.audit\",\n \"_ from desktop.auth import forms as auth_forms from desktop.lib.django_util import\",\n \"= datetime.now() userprofile.save() msg = 'Successful login for user: %s'\",\n \"auth_form.is_valid(): # Must login by using the AuthenticationForm. # It\",\n \"JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg\",\n \"the NOTICE file # distributed with this work for additional\",\n \"'backends' on the User object. user = auth_form.get_user() userprofile =\",\n \"_profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), #\",\n \"AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit = {\",\n \"login by using the AuthenticationForm. # It provides 'backends' on\",\n \"except (IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e)\",\n \"OR CONDITIONS OF ANY KIND, either express or implied. #\",\n \"not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp,\",\n \"LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models import\",\n \"this resolution won't work. uid = None if uid is\",\n \"= get_backend_names() is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get())\",\n \"the License is distributed on an \\\"AS IS\\\" BASIS, #\",\n \"objects and a dictionary of the user's IP address and\",\n \"if backends: for backend in backends: if hasattr(backend, 'logout'): response\",\n \"return JsonResponse({'auth': False}) else: first_user_form = None auth_form = AuthenticationForm()\",\n \"user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'),\",\n \"render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from\",\n \"# If secret_key changed, this resolution won't work. uid =\",\n \"is_first_login_ever or any(backend in backend_names for backend in auto_create_home_backends): #\",\n \"HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed login for\",\n \"False}) else: first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get()\",\n \"session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this resolution won't\",\n \"'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if\",\n \"user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako'\",\n \"from useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password LOG\",\n \"backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True\",\n \"!= '200': raise Exception(_(\\\"Invalid response from OAuth provider: %s\\\") %\",\n \"out the user\\\"\\\"\\\" username = request.user.get_username() request.audit = { 'username':\",\n \"client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' }))\",\n \"login for user: %s' % user.username request.audit['operationText'] = msg access_warn(request,\",\n \"from datetime import datetime from axes.decorators import watch_login import django.contrib.auth.views\",\n \"request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return\",\n \"'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out\",\n \"be converted email=user.email) # OAuth is based on Twitter as\",\n \"'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username }\",\n \"\\\"\\\"\\\" Dumps JSON for user-profile information. \\\"\\\"\\\" return render(None, request,\",\n \"})) if resp['status'] != '200': raise Exception(_(\\\"Invalid response from OAuth\",\n \"'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200':\",\n \"law or agreed to in writing, software # distributed under\",\n \"def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in get_backends()]\",\n \"directory for first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException),\",\n \"get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend',\",\n \"urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends,\",\n \"from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from django.contrib.auth\",\n \"oauth_login(request): assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get())\",\n \"access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to =\",\n \"request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false')\",\n \"= backend.logout(request, next_page) if response: return response return django.contrib.auth.views.logout(request, next_page)\",\n \"# limitations under the License. try: import oauth2 as oauth\",\n \"= first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in backend_names\",\n \"ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path =\",\n \"def profile(request): \\\"\\\"\\\" Dumps JSON for user-profile information. \\\"\\\"\\\" return\",\n \"client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] != '200': raise Exception(_(\\\"Invalid response from\",\n \"create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login\",\n \"# regarding copyright ownership. Cloudera, Inc. licenses this file #\",\n \"import django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation\",\n \"home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login =\",\n \"is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username,\",\n \"request.error(_('Could not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username':\",\n \"SuspiciousOperation: # If secret_key changed, this resolution won't work. uid\",\n \"home directory for first user. try: ensure_home_directory(request.fs, user.username) except (IOError,\",\n \"False msg = 'Failed login for user: %s' % request.POST.get('username')\",\n \"= ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names\",\n \"may obtain a copy of the License at # #\",\n \"user.username request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal',\",\n \"auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names':\",\n \"'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first login, need\",\n \"username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s' % username\",\n \"request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory =\",\n \"user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not\",\n \"url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def\",\n \"= auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm\",\n \"IS\\\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,\",\n \"'/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend'\",\n \"on Twitter as example. @login_notrequired def oauth_login(request): assert oauth is\",\n \"raise Exception(_(\\\"Invalid response from OAuth provider: %s\\\") % resp) access_token\",\n \"request.POST.get('username') request.audit['operationText'] = msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal',\",\n \"else: request.audit['allowed'] = False msg = 'Failed login for user:\",\n \"'logout'): response = backend.logout(request, next_page) if response: return response return\",\n \"dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired\",\n \"WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or\",\n \"= auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit\",\n \"Session from django.http import HttpResponseRedirect from django.utils.translation import ugettext as\",\n \"AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm\",\n \"user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if\",\n \"User.DoesNotExist: LOG.debug(\\\"User with id=%d does not exist\\\" % uid) return\",\n \"for additional information # regarding copyright ownership. Cloudera, Inc. licenses\",\n \"{ 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user: %s'\",\n \"resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token'])\",\n \"# # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law\",\n \"if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid():\",\n \"JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP,\",\n \"on an \\\"AS IS\\\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS\",\n \"first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form\",\n \"user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to)\",\n \"Inc. under one # or more contributor license agreements. See\",\n \"# Must login by using the AuthenticationForm. # It provides\",\n \"first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if\",\n \"= None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin'\",\n \"def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever()\",\n \"first login, need to validate user info! first_user_form = is_first_login_ever\",\n \"= auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie()\",\n \"in compliance # with the License. You may obtain a\",\n \"# to you under the Apache License, Version 2.0 (the\",\n \"redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names()\",\n \"in backend_names for backend in auto_create_home_backends): # Create home directory\",\n \"JsonResponse({'auth': False}) else: first_user_form = None auth_form = AuthenticationForm() if\",\n \"renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path,\",\n \"hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False def get_backend_names():\",\n \"from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util\",\n \"django.core.exceptions import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from\",\n \"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express\",\n \"dictionary of the user's IP address and last access time\\\"\\\"\\\"\",\n \"render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next':\",\n \"if from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action':\",\n \"= 'Failed login for user: %s' % request.POST.get('username') request.audit['operationText'] =\",\n \"auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using\",\n \"Exception(_(\\\"Invalid response from OAuth provider: %s\\\") % resp) request.session['request_token'] =\",\n \"if auth_form.is_valid(): # Must login by using the AuthenticationForm. #\",\n \"'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names,\",\n \"this file # to you under the Apache License, Version\",\n \"'Successful login for user: %s' % user.username request.audit['operationText'] = msg\",\n \"home directory.'), exc_info=e) request.error(_('Could not create home directory.')) if require_change_password(userprofile):\",\n \"= request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory\",\n \"does not exist\\\" % uid) return current_users def first_login_ever(): backends\",\n \"% uid) return current_users def first_login_ever(): backends = get_backends() for\",\n \"response from OAuth provider: %s\\\") % resp) access_token = dict(cgi.parse_qsl(content))\",\n \"import ugettext as _ from desktop.auth import forms as auth_forms\",\n \") if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else:\",\n \"needs to be converted email=user.email) # OAuth is based on\",\n \"for backend in auto_create_home_backends): # Create home directory for first\",\n \"{ 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first login,\",\n \"get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def\",\n \"%s\\\") % resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" %\",\n \"return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now()\",\n \"@login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever\",\n \"request.audit['allowed'] = False msg = 'Failed login for user: %s'\",\n \"dictionary of User objects and a dictionary of the user's\",\n \"= AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using the\",\n \"auth_forms.AuthenticationForm if request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN',\",\n \"http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed\",\n \"OAuth is based on Twitter as example. @login_notrequired def oauth_login(request):\",\n \"= \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request):\",\n \"Twitter as example. @login_notrequired def oauth_login(request): assert oauth is not\",\n \"oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(),\",\n \"renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form':\",\n \"auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method\",\n \"in auto_create_home_backends): # Create home directory for first user. try:\",\n \"WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not\",\n \"'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever,\",\n \"backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response:\",\n \"% username } backends = get_backends() if backends: for backend\",\n \"get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/')\",\n \"from django.contrib.auth.models import User from django.contrib.sessions.models import Session from django.http\",\n \"or implied. # See the License for the specific language\",\n \"for backend in backends: if hasattr(backend, 'logout'): response = backend.logout(request,\",\n \"require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity =\",\n \"else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method ==\",\n \"information # regarding copyright ownership. Cloudera, Inc. licenses this file\",\n \"request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def dt_logout(request,\",\n \"the user's IP address and last access time\\\"\\\"\\\" current_users =\",\n \"and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm =\",\n \"# \\\"License\\\"); you may not use this file except in\",\n \"KIND, either express or implied. # See the License for\",\n \"specific language governing permissions and # limitations under the License.\",\n \"import Session from django.http import HttpResponseRedirect from django.utils.translation import ugettext\",\n \"current_users = { } for session in Session.objects.all(): try: uid\",\n \"UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST':\",\n \"= 'Successful login for user: %s' % user.username request.audit['operationText'] =\",\n \"\\\"License\\\"); you may not use this file except in compliance\",\n \"from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': True}) else:\",\n \"python # Licensed to Cloudera, Inc. under one # or\",\n \"oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token)\",\n \"License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by\",\n \"user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path\",\n \"None first_user = first_user_form and first_user_form.is_valid() if first_user or not\",\n \"django.http import HttpResponseRedirect from django.utils.translation import ugettext as _ from\",\n \"msg = 'Successful login for user: %s' % user.username request.audit['operationText']\",\n \"'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method\",\n \"# distributed with this work for additional information # regarding\",\n \"auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired from\",\n \"ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could not create home\",\n \"def first_login_ever(): backends = get_backends() for backend in backends: if\",\n \"converted email=user.email) # OAuth is based on Twitter as example.\",\n \"@watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever =\",\n \"first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e: LOG.error(_('Could\",\n \"permissions and # limitations under the License. try: import oauth2\",\n \"and UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid() if\",\n \"user: %s' % user.username request.audit['operationText'] = msg access_warn(request, msg) if\",\n \"= oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status']\",\n \"%s' % username } backends = get_backends() if backends: for\",\n \"= is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form and\",\n \"= { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out user:\",\n \"request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to,\",\n \"License. try: import oauth2 as oauth except: oauth = None\",\n \"Inc. licenses this file # to you under the Apache\",\n \"msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true':\",\n \"import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import WebHdfsException from useradmin.models\",\n \"request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\")\",\n \"def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime\",\n \"in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory:\",\n \"return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False\",\n \"is_first_login_ever and UserCreationForm(data=request.POST) or None first_user = first_user_form and first_user_form.is_valid()\",\n \"= 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form\",\n \"= auth_forms.AuthenticationForm if request.method == 'POST': request.audit = { 'operation':\",\n \"desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED\",\n \"request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200': raise Exception(_(\\\"Invalid\",\n \"# # Unless required by applicable law or agreed to\",\n \"Version 2.0 (the # \\\"License\\\"); you may not use this\",\n \"except User.DoesNotExist: LOG.debug(\\\"User with id=%d does not exist\\\" % uid)\",\n \"= request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText':\",\n \"'Logged out user: %s' % username } backends = get_backends()\",\n \"client = oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback':\",\n \"backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return True return False\",\n \"one # or more contributor license agreements. See the NOTICE\",\n \"obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0\",\n \"Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key\",\n \"% resp) request.session['request_token'] = dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(),\",\n \"SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models import\",\n \"or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must\",\n \"last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions import\",\n \"directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False\",\n \"login(request, user) if request.session.test_cookie_worked(): request.session.delete_test_cookie() auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend']\",\n \"return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form,\",\n \"% resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user)\",\n \"OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client = oauth.Client(consumer, token) resp,\",\n \"except in compliance # with the License. You may obtain\",\n \"= None if uid is not None: try: userobj =\",\n \"import cgi import logging import urllib from datetime import datetime\",\n \"and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST)\",\n \"\\\"\\\"\\\"Return dictionary of User objects and a dictionary of the\",\n \"return get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login\",\n \"from django.http import HttpResponseRedirect from django.utils.translation import ugettext as _\",\n \"= client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] != '200': raise Exception(_(\\\"Invalid response\",\n \"implied. # See the License for the specific language governing\",\n \"oauth.Client(consumer, token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] !=\",\n \"NOTICE file # distributed with this work for additional information\",\n \"limitations under the License. try: import oauth2 as oauth except:\",\n \"False def get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in\",\n \"For first login, need to validate user info! first_user_form =\",\n \"Dumps JSON for user-profile information. \\\"\\\"\\\" return render(None, request, _profile_dict(request.user))\",\n \"'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } #\",\n \"this file except in compliance # with the License. You\",\n \"if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False})\",\n \"return HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if\",\n \"from_modal: renderable_path = 'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'),\",\n \"id=%d does not exist\\\" % uid) return current_users def first_login_ever():\",\n \"return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token\",\n \"UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm\",\n \"from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map from\",\n \"not create home directory.')) if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username}))\",\n \"not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user)\",\n \"\\\"\\\"\\\" return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username,\",\n \"datetime object needs to be converted email=user.email) # OAuth is\",\n \"resp['status'] != '200': raise Exception(_(\\\"Invalid response from OAuth provider: %s\\\")\",\n \"as example. @login_notrequired def oauth_login(request): assert oauth is not None\",\n \"license agreements. See the NOTICE file # distributed with this\",\n \"user's IP address and last access time\\\"\\\"\\\" current_users = {\",\n \"by applicable law or agreed to in writing, software #\",\n \"resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] != '200': raise\",\n \"= User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\\\"User\",\n \"first_user or not is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): #\",\n \"useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return\",\n \"content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] != '200': raise Exception(_(\\\"Invalid\",\n \"True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg =\",\n \"#!/usr/bin/env python # Licensed to Cloudera, Inc. under one #\",\n \"and backend.is_first_login_ever(): return True return False def get_backend_names(): return get_backends\",\n \"to be converted email=user.email) # OAuth is based on Twitter\",\n \"if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if response: return\",\n \"User from django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from\",\n \"backends = get_backends() if backends: for backend in backends: if\",\n \"oauth except: oauth = None import cgi import logging import\",\n \"HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save()\",\n \"provides 'backends' on the User object. user = auth_form.get_user() userprofile\",\n \"'login_modal.mako' return render(renderable_path, request, { 'action': urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or\",\n \"request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal: renderable_path = 'login_modal.mako' return\",\n \"from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access\",\n \"and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired @watch_login def dt_login(request,\",\n \"OAuth provider: %s\\\") % resp) access_token = dict(cgi.parse_qsl(content)) user =\",\n \"access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from hadoop.fs.exceptions\",\n \"not use this file except in compliance # with the\",\n \"request.audit = { 'username': username, 'operation': 'USER_LOGOUT', 'operationText': 'Logged out\",\n \"user = auth_form.get_user() userprofile = get_profile(user) login(request, user) if request.session.test_cookie_worked():\",\n \"= logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary of User objects and\",\n \"logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary of User objects and a\",\n \"'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for backend in\",\n \"an \\\"AS IS\\\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF\",\n \"WebHdfsException from useradmin.models import get_profile from useradmin.views import ensure_home_directory, require_change_password\",\n \"for first user. try: ensure_home_directory(request.fs, user.username) except (IOError, WebHdfsException), e:\",\n \"return True return False def get_backend_names(): return get_backends and [backend.__class__.__name__\",\n \"Unless required by applicable law or agreed to in writing,\",\n \"django.contrib.sessions.models import Session from django.http import HttpResponseRedirect from django.utils.translation import\",\n \"auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if request.method == 'POST': request.audit =\",\n \"auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not 'admin' in request.REQUEST:\",\n \"Cloudera, Inc. licenses this file # to you under the\",\n \"HttpResponseRedirect(redirect_to) if not from_modal: request.session.set_test_cookie() renderable_path = 'login.mako' if from_modal:\",\n \"None if uid is not None: try: userobj = User.objects.get(pk=uid)\",\n \"the specific language governing permissions and # limitations under the\",\n \"Must login by using the AuthenticationForm. # It provides 'backends'\",\n \"# or more contributor license agreements. See the NOTICE file\",\n \"time\\\"\\\"\\\" current_users = { } for session in Session.objects.all(): try:\",\n \"if response: return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\"\",\n \"and # limitations under the License. try: import oauth2 as\",\n \"= msg access_warn(request, msg) if from_modal or request.REQUEST.get('fromModal', 'false') ==\",\n \"% (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer =\",\n \"applicable law or agreed to in writing, software # distributed\",\n \"in backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page) if\",\n \"userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful\",\n \"'username': request.POST.get('username') } # For first login, need to validate\",\n \"not exist\\\" % uid) return current_users def first_login_ever(): backends =\",\n \"None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client = oauth.Client(consumer) resp, content\",\n \"datetime import datetime from axes.decorators import watch_login import django.contrib.auth.views from\",\n \"next_page) def profile(request): \\\"\\\"\\\" Dumps JSON for user-profile information. \\\"\\\"\\\"\",\n \"uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: # If secret_key changed, this\",\n \"== 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') }\",\n \"= dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next',\",\n \"in writing, software # distributed under the License is distributed\",\n \"import SuspiciousOperation from django.contrib.auth import login, get_backends, authenticate from django.contrib.auth.models\",\n \"['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for\",\n \"(the # \\\"License\\\"); you may not use this file except\",\n \"= dict(cgi.parse_qsl(content)) url = \\\"%s?oauth_token=%s\\\" % (OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url)\",\n \"HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token =\",\n \"desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import\",\n \"backend in backends: if hasattr(backend, 'logout'): response = backend.logout(request, next_page)\",\n \"} backends = get_backends() if backends: for backend in backends:\",\n \"= client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/'\",\n \"won't work. uid = None if uid is not None:\",\n \"from desktop.log.access import access_warn, last_access_map from desktop.conf import LDAP, OAUTH,\",\n \"= { } for session in Session.objects.all(): try: uid =\",\n \"def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\" username = request.user.get_username()\",\n \"as _ from desktop.auth import forms as auth_forms from desktop.lib.django_util\",\n \"username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to be\",\n \"resolution won't work. uid = None if uid is not\",\n \"is_first_login_ever: auth_form = AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by\",\n \"msg) if from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth':\",\n \"for session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation:\",\n \"next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\" username = request.user.get_username() request.audit =\",\n \"LOG = logging.getLogger(__name__) def get_current_users(): \\\"\\\"\\\"Return dictionary of User objects\",\n \"uid = None if uid is not None: try: userobj\",\n \"JSON for user-profile information. \\\"\\\"\\\" return render(None, request, _profile_dict(request.user)) def\",\n \"render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name,\",\n \"(OAUTH.AUTHENTICATE_URL.get(), request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(),\",\n \"License is distributed on an \\\"AS IS\\\" BASIS, # WITHOUT\",\n \"not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, {\",\n \"} # For first login, need to validate user info!\",\n \"oauth2 as oauth except: oauth = None import cgi import\",\n \"oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) token = oauth.Token(request.session['request_token']['oauth_token'], request.session['request_token']['oauth_token_secret']) client\",\n \"import access_warn, last_access_map from desktop.conf import LDAP, OAUTH, DEMO_ENABLED from\",\n \"None: try: userobj = User.objects.get(pk=uid) current_users[userobj] = last_access_map.get(userobj.username, { })\",\n \"backend_names and ( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm\",\n \"user: %s' % username } backends = get_backends() if backends:\",\n \"LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could not create home\",\n \"get_backend_names(): return get_backends and [backend.__class__.__name__ for backend in get_backends()] @login_notrequired\",\n \"info! first_user_form = is_first_login_ever and UserCreationForm(data=request.POST) or None first_user =\",\n \"False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login for\",\n \"copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # #\",\n \"return HttpResponseRedirect(redirect_to) else: request.audit['allowed'] = False msg = 'Failed login\",\n \"'active_directory': is_active_directory }) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\"\",\n \"with the License. You may obtain a copy of the\",\n \"'USER_LOGIN', 'username': request.POST.get('username') } # For first login, need to\",\n \"login, need to validate user info! first_user_form = is_first_login_ever and\",\n \"login_notrequired from desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map\",\n \"import render from desktop.lib.django_util import login_notrequired from desktop.lib.django_util import JsonResponse\",\n \"from_modal or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else:\",\n \"first_user_form = None auth_form = AuthenticationForm() if DEMO_ENABLED.get() and not\",\n \"AuthenticationForm(data=request.POST) if auth_form.is_valid(): # Must login by using the AuthenticationForm.\",\n \"except SuspiciousOperation: # If secret_key changed, this resolution won't work.\",\n \"urlresolvers.reverse('desktop.auth.views.dt_login'), 'form': first_user_form or auth_form, 'next': redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors':\",\n \"token) resp, content = client.request(OAUTH.ACCESS_TOKEN_URL.get(), \\\"GET\\\") if resp['status'] != '200':\",\n \"dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/')\",\n \"django.contrib.auth.models import User from django.contrib.sessions.models import Session from django.http import\",\n \"auto_create_home_backends = ['AllowAllBackend', 'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in\",\n \"the License for the specific language governing permissions and #\",\n \"'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory }) def\",\n \"out user: %s' % username } backends = get_backends() if\",\n \"dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login), # datetime object needs to\",\n \"'is_first_login_ever') and backend.is_first_login_ever(): return True return False def get_backend_names(): return\",\n \"msg = 'Failed login for user: %s' % request.POST.get('username') request.audit['operationText']\",\n \"by using the AuthenticationForm. # It provides 'backends' on the\",\n \"file # to you under the Apache License, Version 2.0\",\n \"either express or implied. # See the License for the\",\n \"login for user: %s' % request.POST.get('username') request.audit['operationText'] = msg access_warn(request,\",\n \"# with the License. You may obtain a copy of\",\n \"redirect_to, 'first_login_ever': is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory':\",\n \"# http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or\",\n \"'oauth_callback': 'http://' + request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] !=\",\n \"the User object. user = auth_form.get_user() userprofile = get_profile(user) login(request,\",\n \"import oauth2 as oauth except: oauth = None import cgi\",\n \"assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client\",\n \"from_modal=False): redirect_to = request.REQUEST.get('next', '/') is_first_login_ever = first_login_ever() backend_names =\",\n \"in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username)\",\n \"If secret_key changed, this resolution won't work. uid = None\",\n \"logging import urllib from datetime import datetime from axes.decorators import\",\n \"copyright ownership. Cloudera, Inc. licenses this file # to you\",\n \"under one # or more contributor license agreements. See the\",\n \"username = request.user.get_username() request.audit = { 'username': username, 'operation': 'USER_LOGOUT',\",\n \"or any(backend in backend_names for backend in auto_create_home_backends): # Create\",\n \"a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 #\",\n \"Cloudera, Inc. under one # or more contributor license agreements.\",\n \"import login, get_backends, authenticate from django.contrib.auth.models import User from django.contrib.sessions.models\",\n \"request.session['request_token']['oauth_token']) return HttpResponseRedirect(url) @login_notrequired def oauth_authenticated(request): consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get())\",\n \"axes.decorators import watch_login import django.contrib.auth.views from django.core import urlresolvers from\",\n \"if require_change_password(userprofile): return HttpResponseRedirect(urlresolvers.reverse('useradmin.views.edit_user', kwargs={'username': user.username})) userprofile.first_login = False userprofile.last_activity\",\n \"for backend in backends: if hasattr(backend, 'is_first_login_ever') and backend.is_first_login_ever(): return\",\n \"None import cgi import logging import urllib from datetime import\",\n \"# Create home directory for first user. try: ensure_home_directory(request.fs, user.username)\",\n \"to Cloudera, Inc. under one # or more contributor license\",\n \"the Apache License, Version 2.0 (the # \\\"License\\\"); you may\",\n \"= first_user_form and first_user_form.is_valid() if first_user or not is_first_login_ever: auth_form\",\n \"user.username})) userprofile.first_login = False userprofile.last_activity = datetime.now() userprofile.save() msg =\",\n \"is_first_login_ever, 'login_errors': request.method == 'POST', 'backend_names': backend_names, 'active_directory': is_active_directory })\",\n \"def oauth_login(request): assert oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(),\",\n \"oauth is not None consumer = oauth.Consumer(OAUTH.CONSUMER_KEY.get(), OAUTH.CONSUMER_SECRET.get()) client =\",\n \"request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name, last_name=user.last_name, last_login=str(user.last_login),\",\n \"desktop.lib.django_util import JsonResponse from desktop.log.access import access_warn, last_access_map from desktop.conf\",\n \"authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs, user.username) return HttpResponseRedirect(redirect_to) if not\",\n \"import urllib from datetime import datetime from axes.decorators import watch_login\",\n \"you under the Apache License, Version 2.0 (the # \\\"License\\\");\",\n \"from desktop.auth import forms as auth_forms from desktop.lib.django_util import render\",\n \"'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request, user) ensure_home_directory(request.fs,\",\n \"if request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username':\",\n \"session in Session.objects.all(): try: uid = session.get_decoded().get(django.contrib.auth.SESSION_KEY) except SuspiciousOperation: #\",\n \"language governing permissions and # limitations under the License. try:\",\n \"return render(None, request, _profile_dict(request.user)) def _profile_dict(user): return dict( username=user.username, first_name=user.first_name,\",\n \"# It provides 'backends' on the User object. user =\",\n \"'LdapBackend', 'SpnegoDjangoBackend'] if is_first_login_ever or any(backend in backend_names for backend\",\n \"and not 'admin' in request.REQUEST: user = authenticate(username=request.user.username, password='<PASSWORD>') login(request,\",\n \"agreements. See the NOTICE file # distributed with this work\",\n \"distributed on an \\\"AS IS\\\" BASIS, # WITHOUT WARRANTIES OR\",\n \"response return django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\" Dumps JSON for\",\n \"licenses this file # to you under the Apache License,\",\n \"= False userprofile.last_activity = datetime.now() userprofile.save() msg = 'Successful login\",\n \"get_backend_names() is_active_directory = 'LdapBackend' in backend_names and ( bool(LDAP.NT_DOMAIN.get()) or\",\n \"return response return django.contrib.auth.views.logout(request, next_page) def profile(request): \\\"\\\"\\\" Dumps JSON\",\n \"backend in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to =\",\n \"# distributed under the License is distributed on an \\\"AS\",\n \"'false') == 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else:\",\n \"# OAuth is based on Twitter as example. @login_notrequired def\",\n \"work for additional information # regarding copyright ownership. Cloudera, Inc.\",\n \"# Unless required by applicable law or agreed to in\",\n \"uid is not None: try: userobj = User.objects.get(pk=uid) current_users[userobj] =\",\n \"= last_access_map.get(userobj.username, { }) except User.DoesNotExist: LOG.debug(\\\"User with id=%d does\",\n \"= { 'operation': 'USER_LOGIN', 'username': request.POST.get('username') } # For first\",\n \"get_profile from useradmin.views import ensure_home_directory, require_change_password LOG = logging.getLogger(__name__) def\",\n \"bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm\",\n \"backends = get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever')\",\n \"'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form = None\",\n \"}) def dt_logout(request, next_page=None): \\\"\\\"\\\"Log out the user\\\"\\\"\\\" username =\",\n \"access time\\\"\\\"\\\" current_users = { } for session in Session.objects.all():\",\n \"\\\"AS IS\\\" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY\",\n \"provider: %s\\\") % resp) access_token = dict(cgi.parse_qsl(content)) user = authenticate(access_token=access_token)\",\n \"or request.REQUEST.get('fromModal', 'false') == 'true': return JsonResponse({'auth': False}) else: first_user_form\",\n \"= authenticate(access_token=access_token) login(request, user) redirect_to = request.REQUEST.get('next', '/') return HttpResponseRedirect(redirect_to)\",\n \"oauth.Client(consumer) resp, content = client.request(OAUTH.REQUEST_TOKEN_URL.get(), \\\"POST\\\", body=urllib.urlencode({ 'oauth_callback': 'http://' +\",\n \"last_login=str(user.last_login), # datetime object needs to be converted email=user.email) #\",\n \"It provides 'backends' on the User object. user = auth_form.get_user()\",\n \"License. You may obtain a copy of the License at\",\n \"User objects and a dictionary of the user's IP address\",\n \"request.method == 'POST': request.audit = { 'operation': 'USER_LOGIN', 'username': request.POST.get('username')\",\n \"any(backend in backend_names for backend in auto_create_home_backends): # Create home\",\n \"You may obtain a copy of the License at #\",\n \"django.contrib.auth.views from django.core import urlresolvers from django.core.exceptions import SuspiciousOperation from\",\n \"User object. user = auth_form.get_user() userprofile = get_profile(user) login(request, user)\",\n \"import HttpResponseRedirect from django.utils.translation import ugettext as _ from desktop.auth\",\n \"as auth_forms from desktop.lib.django_util import render from desktop.lib.django_util import login_notrequired\",\n \"(IOError, WebHdfsException), e: LOG.error(_('Could not create home directory.'), exc_info=e) request.error(_('Could\",\n \"( bool(LDAP.NT_DOMAIN.get()) or bool(LDAP.LDAP_SERVERS.get()) ) if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm\",\n \"username } backends = get_backends() if backends: for backend in\",\n \"== 'true': return JsonResponse({'auth': True}) else: return HttpResponseRedirect(redirect_to) else: request.audit['allowed']\",\n \"'operationText': 'Logged out user: %s' % username } backends =\",\n \"compliance # with the License. You may obtain a copy\",\n \"= auth_forms.LdapAuthenticationForm else: UserCreationForm = auth_forms.UserCreationForm AuthenticationForm = auth_forms.AuthenticationForm if\",\n \"in get_backends()] @login_notrequired @watch_login def dt_login(request, from_modal=False): redirect_to = request.REQUEST.get('next',\",\n \"if is_active_directory: UserCreationForm = auth_forms.LdapUserCreationForm AuthenticationForm = auth_forms.LdapAuthenticationForm else: UserCreationForm\",\n \"= None import cgi import logging import urllib from datetime\",\n \"= get_backends() for backend in backends: if hasattr(backend, 'is_first_login_ever') and\",\n \"django.utils.translation import ugettext as _ from desktop.auth import forms as\"\n]","isConversational":false}}},{"rowIdx":683,"cells":{"ngram":{"kind":"list like","value":["= name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list","== other.get_follow_list() \\ and self.intention == other.get_intention() and self.lane ==","get_name(self): return self.name def set_name(self, name): self.name = name def","= intention def get_lane(self): return self.lane def set_lane(self, lane): self.lane","follow_list def get_intention(self): return self.intention def set_intention(self, intention): self.intention =","intention, lane): self.name = name self.follow_list = follow_list self.intention =","def set_intention(self, intention): self.intention = intention def get_lane(self): return self.lane","self.intention def set_intention(self, intention): self.intention = intention def get_lane(self): return","self.follow_list = follow_list self.intention = intention self.lane = lane def","get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def","intention): self.intention = intention def get_lane(self): return self.lane def set_lane(self,","__eq__(self, other): if isinstance(other, Node): if self.name == other.get_name() and","True return False def get_name(self): return self.name def set_name(self, name):","return False def get_name(self): return self.name def set_name(self, name): self.name","__init__(self, name, follow_list, intention, lane): self.name = name self.follow_list =","Node(object): def __init__(self, name, follow_list, intention, lane): self.name = name","lane def __eq__(self, other): if isinstance(other, Node): if self.name ==","== other.get_lane(): return True return False def get_name(self): return self.name","return self.name def set_name(self, name): self.name = name def get_follow_list(self):","self.intention = intention def get_lane(self): return self.lane def set_lane(self, lane):","== other.get_intention() and self.lane == other.get_lane(): return True return False","and self.follow_list == other.get_follow_list() \\ and self.intention == other.get_intention() and","intention self.lane = lane def __eq__(self, other): if isinstance(other, Node):","self.follow_list == other.get_follow_list() \\ and self.intention == other.get_intention() and self.lane","class Node(object): def __init__(self, name, follow_list, intention, lane): self.name =","lane): self.name = name self.follow_list = follow_list self.intention = intention","def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list","self.name = name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list):","= follow_list def get_intention(self): return self.intention def set_intention(self, intention): self.intention","and self.lane == other.get_lane(): return True return False def get_name(self):","return True return False def get_name(self): return self.name def set_name(self,","self.lane == other.get_lane(): return True return False def get_name(self): return","follow_list, intention, lane): self.name = name self.follow_list = follow_list self.intention","= lane def __eq__(self, other): if isinstance(other, Node): if self.name","== other.get_name() and self.follow_list == other.get_follow_list() \\ and self.intention ==","set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return self.intention def","def get_lane(self): return self.lane def set_lane(self, lane): self.lane = lane","self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return","self.name == other.get_name() and self.follow_list == other.get_follow_list() \\ and self.intention","def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return self.intention","self.intention == other.get_intention() and self.lane == other.get_lane(): return True return","def __eq__(self, other): if isinstance(other, Node): if self.name == other.get_name()","name self.follow_list = follow_list self.intention = intention self.lane = lane","self.name = name self.follow_list = follow_list self.intention = intention self.lane","other.get_follow_list() \\ and self.intention == other.get_intention() and self.lane == other.get_lane():","def set_name(self, name): self.name = name def get_follow_list(self): return self.follow_list","set_name(self, name): self.name = name def get_follow_list(self): return self.follow_list def","other): if isinstance(other, Node): if self.name == other.get_name() and self.follow_list","def __init__(self, name, follow_list, intention, lane): self.name = name self.follow_list","and self.intention == other.get_intention() and self.lane == other.get_lane(): return True","intention def get_lane(self): return self.lane def set_lane(self, lane): self.lane =","get_intention(self): return self.intention def set_intention(self, intention): self.intention = intention def","self.name def set_name(self, name): self.name = name def get_follow_list(self): return","return self.intention def set_intention(self, intention): self.intention = intention def get_lane(self):","if isinstance(other, Node): if self.name == other.get_name() and self.follow_list ==","False def get_name(self): return self.name def set_name(self, name): self.name =","isinstance(other, Node): if self.name == other.get_name() and self.follow_list == other.get_follow_list()","return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self):","def get_intention(self): return self.intention def set_intention(self, intention): self.intention = intention","self.follow_list = follow_list def get_intention(self): return self.intention def set_intention(self, intention):","= name self.follow_list = follow_list self.intention = intention self.lane =","other.get_intention() and self.lane == other.get_lane(): return True return False def","self.lane = lane def __eq__(self, other): if isinstance(other, Node): if","Node): if self.name == other.get_name() and self.follow_list == other.get_follow_list() \\","def get_name(self): return self.name def set_name(self, name): self.name = name","if self.name == other.get_name() and self.follow_list == other.get_follow_list() \\ and","follow_list self.intention = intention self.lane = lane def __eq__(self, other):","= intention self.lane = lane def __eq__(self, other): if isinstance(other,","set_intention(self, intention): self.intention = intention def get_lane(self): return self.lane def","other.get_name() and self.follow_list == other.get_follow_list() \\ and self.intention == other.get_intention()","\\ and self.intention == other.get_intention() and self.lane == other.get_lane(): return","self.intention = intention self.lane = lane def __eq__(self, other): if","name, follow_list, intention, lane): self.name = name self.follow_list = follow_list","name): self.name = name def get_follow_list(self): return self.follow_list def set_follow_list(self,","follow_list): self.follow_list = follow_list def get_intention(self): return self.intention def set_intention(self,","other.get_lane(): return True return False def get_name(self): return self.name def","= follow_list self.intention = intention self.lane = lane def __eq__(self,","name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list ="],"string":"[\n \"= name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list\",\n \"== other.get_follow_list() \\\\ and self.intention == other.get_intention() and self.lane ==\",\n \"get_name(self): return self.name def set_name(self, name): self.name = name def\",\n \"= intention def get_lane(self): return self.lane def set_lane(self, lane): self.lane\",\n \"follow_list def get_intention(self): return self.intention def set_intention(self, intention): self.intention =\",\n \"intention, lane): self.name = name self.follow_list = follow_list self.intention =\",\n \"def set_intention(self, intention): self.intention = intention def get_lane(self): return self.lane\",\n \"self.intention def set_intention(self, intention): self.intention = intention def get_lane(self): return\",\n \"self.follow_list = follow_list self.intention = intention self.lane = lane def\",\n \"get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def\",\n \"intention): self.intention = intention def get_lane(self): return self.lane def set_lane(self,\",\n \"__eq__(self, other): if isinstance(other, Node): if self.name == other.get_name() and\",\n \"True return False def get_name(self): return self.name def set_name(self, name):\",\n \"return False def get_name(self): return self.name def set_name(self, name): self.name\",\n \"__init__(self, name, follow_list, intention, lane): self.name = name self.follow_list =\",\n \"Node(object): def __init__(self, name, follow_list, intention, lane): self.name = name\",\n \"lane def __eq__(self, other): if isinstance(other, Node): if self.name ==\",\n \"== other.get_lane(): return True return False def get_name(self): return self.name\",\n \"return self.name def set_name(self, name): self.name = name def get_follow_list(self):\",\n \"self.intention = intention def get_lane(self): return self.lane def set_lane(self, lane):\",\n \"== other.get_intention() and self.lane == other.get_lane(): return True return False\",\n \"and self.follow_list == other.get_follow_list() \\\\ and self.intention == other.get_intention() and\",\n \"intention self.lane = lane def __eq__(self, other): if isinstance(other, Node):\",\n \"self.follow_list == other.get_follow_list() \\\\ and self.intention == other.get_intention() and self.lane\",\n \"class Node(object): def __init__(self, name, follow_list, intention, lane): self.name =\",\n \"lane): self.name = name self.follow_list = follow_list self.intention = intention\",\n \"def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list\",\n \"self.name = name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list):\",\n \"= follow_list def get_intention(self): return self.intention def set_intention(self, intention): self.intention\",\n \"and self.lane == other.get_lane(): return True return False def get_name(self):\",\n \"return True return False def get_name(self): return self.name def set_name(self,\",\n \"self.lane == other.get_lane(): return True return False def get_name(self): return\",\n \"follow_list, intention, lane): self.name = name self.follow_list = follow_list self.intention\",\n \"= lane def __eq__(self, other): if isinstance(other, Node): if self.name\",\n \"== other.get_name() and self.follow_list == other.get_follow_list() \\\\ and self.intention ==\",\n \"set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return self.intention def\",\n \"def get_lane(self): return self.lane def set_lane(self, lane): self.lane = lane\",\n \"self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return\",\n \"self.name == other.get_name() and self.follow_list == other.get_follow_list() \\\\ and self.intention\",\n \"def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self): return self.intention\",\n \"self.intention == other.get_intention() and self.lane == other.get_lane(): return True return\",\n \"def __eq__(self, other): if isinstance(other, Node): if self.name == other.get_name()\",\n \"name self.follow_list = follow_list self.intention = intention self.lane = lane\",\n \"self.name = name self.follow_list = follow_list self.intention = intention self.lane\",\n \"other.get_follow_list() \\\\ and self.intention == other.get_intention() and self.lane == other.get_lane():\",\n \"def set_name(self, name): self.name = name def get_follow_list(self): return self.follow_list\",\n \"set_name(self, name): self.name = name def get_follow_list(self): return self.follow_list def\",\n \"other): if isinstance(other, Node): if self.name == other.get_name() and self.follow_list\",\n \"def __init__(self, name, follow_list, intention, lane): self.name = name self.follow_list\",\n \"and self.intention == other.get_intention() and self.lane == other.get_lane(): return True\",\n \"intention def get_lane(self): return self.lane def set_lane(self, lane): self.lane =\",\n \"get_intention(self): return self.intention def set_intention(self, intention): self.intention = intention def\",\n \"self.name def set_name(self, name): self.name = name def get_follow_list(self): return\",\n \"return self.intention def set_intention(self, intention): self.intention = intention def get_lane(self):\",\n \"if isinstance(other, Node): if self.name == other.get_name() and self.follow_list ==\",\n \"False def get_name(self): return self.name def set_name(self, name): self.name =\",\n \"isinstance(other, Node): if self.name == other.get_name() and self.follow_list == other.get_follow_list()\",\n \"return self.follow_list def set_follow_list(self, follow_list): self.follow_list = follow_list def get_intention(self):\",\n \"def get_intention(self): return self.intention def set_intention(self, intention): self.intention = intention\",\n \"self.follow_list = follow_list def get_intention(self): return self.intention def set_intention(self, intention):\",\n \"= name self.follow_list = follow_list self.intention = intention self.lane =\",\n \"other.get_intention() and self.lane == other.get_lane(): return True return False def\",\n \"self.lane = lane def __eq__(self, other): if isinstance(other, Node): if\",\n \"Node): if self.name == other.get_name() and self.follow_list == other.get_follow_list() \\\\\",\n \"def get_name(self): return self.name def set_name(self, name): self.name = name\",\n \"if self.name == other.get_name() and self.follow_list == other.get_follow_list() \\\\ and\",\n \"follow_list self.intention = intention self.lane = lane def __eq__(self, other):\",\n \"= intention self.lane = lane def __eq__(self, other): if isinstance(other,\",\n \"set_intention(self, intention): self.intention = intention def get_lane(self): return self.lane def\",\n \"other.get_name() and self.follow_list == other.get_follow_list() \\\\ and self.intention == other.get_intention()\",\n \"\\\\ and self.intention == other.get_intention() and self.lane == other.get_lane(): return\",\n \"self.intention = intention self.lane = lane def __eq__(self, other): if\",\n \"name, follow_list, intention, lane): self.name = name self.follow_list = follow_list\",\n \"name): self.name = name def get_follow_list(self): return self.follow_list def set_follow_list(self,\",\n \"follow_list): self.follow_list = follow_list def get_intention(self): return self.intention def set_intention(self,\",\n \"other.get_lane(): return True return False def get_name(self): return self.name def\",\n \"= follow_list self.intention = intention self.lane = lane def __eq__(self,\",\n \"name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list =\"\n]","isConversational":false}}},{"rowIdx":684,"cells":{"ngram":{"kind":"list like","value":["re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if","headers = int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1]","path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query)","from __future__ import division from __future__ import print_function from __future__","args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc,","absolute_import from __future__ import division from __future__ import print_function from","# fallback to regex to extract from match = FROM_REGEX.search(sql)","<reponame>tim-werner/gsheets-db-api<filename>gsheetsdb/url.py from __future__ import absolute_import from __future__ import division from","in qs: headers = int(qs['headers'][-1]) if 'gid' in qs: gid","if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if headers","qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args","gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None,","0: args['headers'] = headers if sheet is not None: args['sheet']","sheet else: args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse(","parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs","= parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq'))","= qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1] if","in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):]","parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path","gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')]","extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex","= '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs:","args = OrderedDict() if headers > 0: args['headers'] = headers","qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if","qs = parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1])","qs: gid = qs['gid'][-1] if 'sheet' in qs: sheet =","= qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict()","args['headers'] = headers if sheet is not None: args['sheet'] =","(parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql): try: return","path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in","parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path","import re from six.moves.urllib import parse FROM_REGEX = re.compile(' from","'gid' in qs: gid = qs['gid'][-1] if 'sheet' in qs:","'headers' in qs: headers = int(qs['headers'][-1]) if 'gid' in qs:","get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path","qs: headers = int(qs['headers'][-1]) if 'gid' in qs: gid =","six.moves.urllib import parse FROM_REGEX = re.compile(' from (\"http.*?\")', re.IGNORECASE) def","= gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path,","headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path =","in qs: gid = qs['gid'][-1] if 'sheet' in qs: sheet","sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else:","__future__ import print_function from __future__ import unicode_literals from collections import","__future__ import division from __future__ import print_function from __future__ import","import OrderedDict from moz_sql_parser import parse as parse_sql import pyparsing","parse_sql import pyparsing import re from six.moves.urllib import parse FROM_REGEX","return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql):","parts.fragment[len('gid='):] args = OrderedDict() if headers > 0: args['headers'] =","OrderedDict() if headers > 0: args['headers'] = headers if sheet","to extract from match = FROM_REGEX.search(sql) if match: return match.group(1).strip('\"')","import print_function from __future__ import unicode_literals from collections import OrderedDict","as parse_sql import pyparsing import re from six.moves.urllib import parse","'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid =","None: args['sheet'] = sheet else: args['gid'] = gid params =","if 'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid","> 0: args['headers'] = headers if sheet is not None:","from collections import OrderedDict from moz_sql_parser import parse as parse_sql","= headers if sheet is not None: args['sheet'] = sheet","parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def","(\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url)","try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to","pyparsing import re from six.moves.urllib import parse FROM_REGEX = re.compile('","from __future__ import print_function from __future__ import unicode_literals from collections","FROM_REGEX = re.compile(' from (\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0, gid=0,","headers if sheet is not None: args['sheet'] = sheet else:","from __future__ import absolute_import from __future__ import division from __future__","None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback","parse as parse_sql import pyparsing import re from six.moves.urllib import","sheet is not None: args['sheet'] = sheet else: args['gid'] =","else: args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme,","if headers > 0: args['headers'] = headers if sheet is","qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='):","from __future__ import unicode_literals from collections import OrderedDict from moz_sql_parser","from (\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts =","import unicode_literals from collections import OrderedDict from moz_sql_parser import parse","return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to extract","except pyparsing.ParseException: # fallback to regex to extract from match","headers > 0: args['headers'] = headers if sheet is not","def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'):","is not None: args['sheet'] = sheet else: args['gid'] = gid","params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: #","= OrderedDict() if headers > 0: args['headers'] = headers if","args['sheet'] = sheet else: args['gid'] = gid params = parse.urlencode(args)","= sheet else: args['gid'] = gid params = parse.urlencode(args) return","re.compile(' from (\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts","import pyparsing import re from six.moves.urllib import parse FROM_REGEX =","parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers'","unicode_literals from collections import OrderedDict from moz_sql_parser import parse as","= parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path =","sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args =","= int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1] if","import division from __future__ import print_function from __future__ import unicode_literals","if sheet is not None: args['sheet'] = sheet else: args['gid']","__future__ import absolute_import from __future__ import division from __future__ import","gid = parts.fragment[len('gid='):] args = OrderedDict() if headers > 0:","collections import OrderedDict from moz_sql_parser import parse as parse_sql import","gid = qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1]","division from __future__ import print_function from __future__ import unicode_literals from","not None: args['sheet'] = sheet else: args['gid'] = gid params","path = parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'),","parts.netloc, path, None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from']","parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to extract from","parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if headers >","params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params,","parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1]) if 'gid'","regex to extract from match = FROM_REGEX.search(sql) if match: return","parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql): try:","int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1] if 'sheet'","= re.compile(' from (\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None):","= parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1]) if","= parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if","re from six.moves.urllib import parse FROM_REGEX = re.compile(' from (\"http.*?\")',","from six.moves.urllib import parse FROM_REGEX = re.compile(' from (\"http.*?\")', re.IGNORECASE)","print_function from __future__ import unicode_literals from collections import OrderedDict from","pyparsing.ParseException: # fallback to regex to extract from match =","def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to","if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path path","if 'gid' in qs: gid = qs['gid'][-1] if 'sheet' in","= parts.fragment[len('gid='):] args = OrderedDict() if headers > 0: args['headers']","'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs: headers =","None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException:","OrderedDict from moz_sql_parser import parse as parse_sql import pyparsing import","'/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs: headers","fallback to regex to extract from match = FROM_REGEX.search(sql) if","import parse FROM_REGEX = re.compile(' from (\"http.*?\")', re.IGNORECASE) def get_url(url,","from moz_sql_parser import parse as parse_sql import pyparsing import re","if 'headers' in qs: headers = int(qs['headers'][-1]) if 'gid' in","= parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None))","else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs =","moz_sql_parser import parse as parse_sql import pyparsing import re from","parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path path =","path, None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except","import absolute_import from __future__ import division from __future__ import print_function","import parse as parse_sql import pyparsing import re from six.moves.urllib","__future__ import unicode_literals from collections import OrderedDict from moz_sql_parser import","parse FROM_REGEX = re.compile(' from (\"http.*?\")', re.IGNORECASE) def get_url(url, headers=0,","to regex to extract from match = FROM_REGEX.search(sql) if match:"],"string":"[\n \"re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if\",\n \"headers = int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1]\",\n \"path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query)\",\n \"from __future__ import division from __future__ import print_function from __future__\",\n \"args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc,\",\n \"absolute_import from __future__ import division from __future__ import print_function from\",\n \"# fallback to regex to extract from match = FROM_REGEX.search(sql)\",\n \"<reponame>tim-werner/gsheets-db-api<filename>gsheetsdb/url.py from __future__ import absolute_import from __future__ import division from\",\n \"in qs: headers = int(qs['headers'][-1]) if 'gid' in qs: gid\",\n \"if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if headers\",\n \"qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args\",\n \"gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None,\",\n \"0: args['headers'] = headers if sheet is not None: args['sheet']\",\n \"sheet else: args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse(\",\n \"parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs\",\n \"= parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq'))\",\n \"= qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1] if\",\n \"in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):]\",\n \"parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path\",\n \"gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')]\",\n \"extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex\",\n \"= '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs:\",\n \"args = OrderedDict() if headers > 0: args['headers'] = headers\",\n \"qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if\",\n \"qs = parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1])\",\n \"qs: gid = qs['gid'][-1] if 'sheet' in qs: sheet =\",\n \"= qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict()\",\n \"args['headers'] = headers if sheet is not None: args['sheet'] =\",\n \"(parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql): try: return\",\n \"path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in\",\n \"parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path\",\n \"import re from six.moves.urllib import parse FROM_REGEX = re.compile(' from\",\n \"'gid' in qs: gid = qs['gid'][-1] if 'sheet' in qs:\",\n \"'headers' in qs: headers = int(qs['headers'][-1]) if 'gid' in qs:\",\n \"get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path\",\n \"qs: headers = int(qs['headers'][-1]) if 'gid' in qs: gid =\",\n \"six.moves.urllib import parse FROM_REGEX = re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def\",\n \"= gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path,\",\n \"headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path =\",\n \"in qs: gid = qs['gid'][-1] if 'sheet' in qs: sheet\",\n \"sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else:\",\n \"__future__ import print_function from __future__ import unicode_literals from collections import\",\n \"__future__ import division from __future__ import print_function from __future__ import\",\n \"import OrderedDict from moz_sql_parser import parse as parse_sql import pyparsing\",\n \"parse_sql import pyparsing import re from six.moves.urllib import parse FROM_REGEX\",\n \"return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql):\",\n \"parts.fragment[len('gid='):] args = OrderedDict() if headers > 0: args['headers'] =\",\n \"OrderedDict() if headers > 0: args['headers'] = headers if sheet\",\n \"to extract from match = FROM_REGEX.search(sql) if match: return match.group(1).strip('\\\"')\",\n \"import print_function from __future__ import unicode_literals from collections import OrderedDict\",\n \"as parse_sql import pyparsing import re from six.moves.urllib import parse\",\n \"'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid =\",\n \"None: args['sheet'] = sheet else: args['gid'] = gid params =\",\n \"if 'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid\",\n \"> 0: args['headers'] = headers if sheet is not None:\",\n \"from collections import OrderedDict from moz_sql_parser import parse as parse_sql\",\n \"= headers if sheet is not None: args['sheet'] = sheet\",\n \"parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def\",\n \"(\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url)\",\n \"try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to\",\n \"pyparsing import re from six.moves.urllib import parse FROM_REGEX = re.compile('\",\n \"from __future__ import print_function from __future__ import unicode_literals from collections\",\n \"FROM_REGEX = re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0, gid=0,\",\n \"headers if sheet is not None: args['sheet'] = sheet else:\",\n \"from __future__ import absolute_import from __future__ import division from __future__\",\n \"None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback\",\n \"parse as parse_sql import pyparsing import re from six.moves.urllib import\",\n \"sheet is not None: args['sheet'] = sheet else: args['gid'] =\",\n \"else: args['gid'] = gid params = parse.urlencode(args) return parse.urlunparse( (parts.scheme,\",\n \"if headers > 0: args['headers'] = headers if sheet is\",\n \"qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='):\",\n \"from __future__ import unicode_literals from collections import OrderedDict from moz_sql_parser\",\n \"from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts =\",\n \"import unicode_literals from collections import OrderedDict from moz_sql_parser import parse\",\n \"return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to extract\",\n \"except pyparsing.ParseException: # fallback to regex to extract from match\",\n \"headers > 0: args['headers'] = headers if sheet is not\",\n \"def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if parts.path.endswith('/edit'):\",\n \"is not None: args['sheet'] = sheet else: args['gid'] = gid\",\n \"params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: #\",\n \"= OrderedDict() if headers > 0: args['headers'] = headers if\",\n \"args['sheet'] = sheet else: args['gid'] = gid params = parse.urlencode(args)\",\n \"= sheet else: args['gid'] = gid params = parse.urlencode(args) return\",\n \"re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts\",\n \"import pyparsing import re from six.moves.urllib import parse FROM_REGEX =\",\n \"parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers'\",\n \"unicode_literals from collections import OrderedDict from moz_sql_parser import parse as\",\n \"= parse.urlparse(url) if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path =\",\n \"sheet = qs['sheet'][-1] if parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args =\",\n \"= int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1] if\",\n \"import division from __future__ import print_function from __future__ import unicode_literals\",\n \"if sheet is not None: args['sheet'] = sheet else: args['gid']\",\n \"__future__ import absolute_import from __future__ import division from __future__ import\",\n \"gid = parts.fragment[len('gid='):] args = OrderedDict() if headers > 0:\",\n \"collections import OrderedDict from moz_sql_parser import parse as parse_sql import\",\n \"gid = qs['gid'][-1] if 'sheet' in qs: sheet = qs['sheet'][-1]\",\n \"division from __future__ import print_function from __future__ import unicode_literals from\",\n \"not None: args['sheet'] = sheet else: args['gid'] = gid params\",\n \"path = parts.path[:-len('/edit')] else: path = parts.path path = '/'.join((path.rstrip('/'),\",\n \"parts.netloc, path, None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from']\",\n \"parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to regex to extract from\",\n \"parts.fragment.startswith('gid='): gid = parts.fragment[len('gid='):] args = OrderedDict() if headers >\",\n \"params = parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params,\",\n \"parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1]) if 'gid'\",\n \"regex to extract from match = FROM_REGEX.search(sql) if match: return\",\n \"parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None)) def extract_url(sql): try:\",\n \"int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1] if 'sheet'\",\n \"= re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None):\",\n \"= parse.parse_qs(parts.query) if 'headers' in qs: headers = int(qs['headers'][-1]) if\",\n \"= parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if\",\n \"re from six.moves.urllib import parse FROM_REGEX = re.compile(' from (\\\"http.*?\\\")',\",\n \"from six.moves.urllib import parse FROM_REGEX = re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE)\",\n \"print_function from __future__ import unicode_literals from collections import OrderedDict from\",\n \"pyparsing.ParseException: # fallback to regex to extract from match =\",\n \"def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException: # fallback to\",\n \"if parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path path\",\n \"if 'gid' in qs: gid = qs['gid'][-1] if 'sheet' in\",\n \"= parts.fragment[len('gid='):] args = OrderedDict() if headers > 0: args['headers']\",\n \"'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs: headers =\",\n \"None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except pyparsing.ParseException:\",\n \"OrderedDict from moz_sql_parser import parse as parse_sql import pyparsing import\",\n \"'/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query) if 'headers' in qs: headers\",\n \"fallback to regex to extract from match = FROM_REGEX.search(sql) if\",\n \"import parse FROM_REGEX = re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url,\",\n \"from moz_sql_parser import parse as parse_sql import pyparsing import re\",\n \"if 'headers' in qs: headers = int(qs['headers'][-1]) if 'gid' in\",\n \"= parse.urlencode(args) return parse.urlunparse( (parts.scheme, parts.netloc, path, None, params, None))\",\n \"else: path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs =\",\n \"moz_sql_parser import parse as parse_sql import pyparsing import re from\",\n \"parts.path.endswith('/edit'): path = parts.path[:-len('/edit')] else: path = parts.path path =\",\n \"path, None, params, None)) def extract_url(sql): try: return parse_sql(sql)['from'] except\",\n \"import absolute_import from __future__ import division from __future__ import print_function\",\n \"import parse as parse_sql import pyparsing import re from six.moves.urllib\",\n \"__future__ import unicode_literals from collections import OrderedDict from moz_sql_parser import\",\n \"parse FROM_REGEX = re.compile(' from (\\\"http.*?\\\")', re.IGNORECASE) def get_url(url, headers=0,\",\n \"to regex to extract from match = FROM_REGEX.search(sql) if match:\"\n]","isConversational":false}}},{"rowIdx":685,"cells":{"ngram":{"kind":"list like","value":["len(boxes) == 0 and len(labels) == 0: boxes.append([0, 0, 2,","self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path =","= Image.open(img_path).convert(\"RGB\") w, h = img.size if \"segments_info\" in ann_info:","# Instead of finding boxes, just take the one from","self.ann_file = ann_file self.transforms = transforms self.return_masks = return_masks def","box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b = [xs,","= img_info['width'] return height, width def build(image_set, args): root =","/ ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks,","build(image_set, args): root = Path(args.data_path) assert ( root.exists() ), f\"provided","= masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels # Instead of finding boxes,","4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"])","w, h = img.size if \"segments_info\" in ann_info: masks =","target[\"boxes\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h),","area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd,","# for name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name]","make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w, h =","= { \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"),","[] for ann in segments: if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"])","{} target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info else ann_info[\"id\"]])","iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return","iscrowd target['area'] = area # if \"segments_info\" in ann_info: #","2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and","path {root} does not exist\" mode = \"panoptic\" PATHS =","ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \"r\") as f: self.coco","logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx):","ys, w, h = x.unbind(-1) b = [xs, ys, (xs",") target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\" in","# dtype=torch.int64, # ) target = {} target['image_id'] = torch.tensor([ann_info['image_id']","\"segments_info\" in ann_info: # for name in ['iscrowd', 'area']: #","= \"panoptic\" PATHS = { \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\":","0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes =","= torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area","iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) == 0: boxes.append([0,","self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) # sanity check if","target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is","if \"segments_info\" in ann_info: # for name in ['iscrowd', 'area']:","zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder","= img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms =","return_masks=True): with open(ann_file, \"r\") as f: self.coco = json.load(f) #","[ann[\"category_id\"] for ann in ann_info[\"segments_info\"]], # dtype=torch.int64, # ) target","for name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for","+ w), (ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments):","img_info = self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return","return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height =","img_folder_path = root / img_folder ann_folder_path = root / ann_folder","def build(image_set, args): root = Path(args.data_path) assert ( root.exists() ),","= Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder) / ann_info[\"file_name\"]","finding boxes, just take the one from json info available","= Path(self.ann_folder) / ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w, h =","idx): img_info = self.coco['images'][idx] height = img_info['height'] width = img_info['width']","= torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes,","as np import torch from PIL import Image from panopticapi.utils","ann_info['segments_info']]) if self.transforms is not None: img, target = self.transforms(img,","len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height']","iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd","boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes,","img, target except Exception as e: logging.error(ann_info) raise e def","'images' field so that they are aligned with 'annotations' #","from PIL import Image from panopticapi.utils import rgb2id # from","[] iscrowd = [] area = [] for ann in","+ h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = []","in ann_info: # for name in ['iscrowd', 'area']: # target[name]","target except Exception as e: logging.error(ann_info) raise e def __len__(self):","img.size if \"segments_info\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks","= torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\"category_id\"] for","ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids =","if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0, 2,","does not exist\" mode = \"panoptic\" PATHS = { \"train\":","# labels = torch.tensor( # [ann[\"category_id\"] for ann in ann_info[\"segments_info\"]],","= [xs, ys, (xs + w), (ys + h)] return","[] labels = [] iscrowd = [] area = []","if self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area =","= sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) # sanity check if \"annotations\"","0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64)","# i.e., in alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x:","in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann in","in segments: if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0,","transforms=None, return_masks=True): with open(ann_file, \"r\") as f: self.coco = json.load(f)","for img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4]","ann_folder ann_file = root / ann_file dataset = ConstructionPanoptic( img_folder_path,",") img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder)","one from json info available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"]","in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids","x.unbind(-1) b = [xs, ys, (xs + w), (ys +","= root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file,","len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0)","ann_file self.transforms = transforms self.return_masks = return_masks def __getitem__(self, idx):","in ann_info['segments_info']]) if self.transforms is not None: img, target =","PATHS = { \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\",","ann_folder_path = root / ann_folder ann_file = root / ann_file","return_masks def __getitem__(self, idx): try: ann_info = ( self.coco[\"annotations\"][idx] if","img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms = transforms","ann in segments: if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else:",".construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w,","masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels","( self.coco[\"annotations\"][idx] if \"annotations\" in self.coco else self.coco[\"images\"][idx] ) img_path","area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if","ann_path = Path(self.ann_folder) / ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w, h","self.coco: for img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] ==","boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd =","for ann in ann_info[\"segments_info\"]], # dtype=torch.int64, # ) target =","box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self,","return boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder,","img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root / img_folder","root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set),","masks = rgb2id(masks) ids = np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]])","= img_info['height'] width = img_info['width'] return height, width def build(image_set,","Panoptic path {root} does not exist\" mode = \"panoptic\" PATHS","def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b =","masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\"category_id\"]","img, target = self.transforms(img, target) return img, target except Exception","iscrowd = [] area = [] for ann in segments:","in ann_info else ann_info[\"id\"]]) if self.return_masks: target['masks'] = masks boxes,","exist\" mode = \"panoptic\" PATHS = { \"train\": (\"images\", f\"{mode}\",","= ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, ) return dataset","is not None: img, target = self.transforms(img, target) return img,","Exception as e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images'])","All Rights Reserved import json from pathlib import Path import","get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height'] width =","= torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd","self.coco else self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\")","None: img, target = self.transforms(img, target) return img, target except","e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info =","(xs + w), (ys + h)] return torch.stack(b, dim=-1) def","return img, target except Exception as e: logging.error(ann_info) raise e","width def build(image_set, args): root = Path(args.data_path) assert ( root.exists()","ann in ann_info[\"segments_info\"]]) masks = masks == ids[:, None, None]","2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels =","torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes =","# sanity check if \"annotations\" in self.coco: for img, ann","from util.box_ops import masks_to_boxes from .construction import make_construction_transforms import logging","iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels # Instead of","img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder) /","in alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) #","name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann","take the one from json info available # target[\"boxes\"] =","available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size'] =","= labels # Instead of finding boxes, just take the","dtype=torch.uint8) # labels = torch.tensor( # [ann[\"category_id\"] for ann in","= rgb2id(masks) ids = np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]]) masks","height = img_info['height'] width = img_info['width'] return height, width def","Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved","area # if \"segments_info\" in ann_info: # for name in","return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels =","torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info else ann_info[\"id\"]]) if self.return_masks: target['masks']","torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area # if","import json from pathlib import Path import numpy as np","idx): try: ann_info = ( self.coco[\"annotations\"][idx] if \"annotations\" in self.coco","x[\"id\"]) # sanity check if \"annotations\" in self.coco: for img,","import Path import numpy as np import torch from PIL","img = Image.open(img_path).convert(\"RGB\") w, h = img.size if \"segments_info\" in","0 and len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1)","labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels,","ann_info[\"id\"]]) if self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area","w), (ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes","= root / img_folder ann_folder_path = root / ann_folder ann_file","if len(boxes) == 0 and len(labels) == 0: boxes.append([0, 0,","target['labels'] = labels # Instead of finding boxes, just take","import torch from PIL import Image from panopticapi.utils import rgb2id","ann_file, transforms=None, return_masks=True): with open(ann_file, \"r\") as f: self.coco =","from json info available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] =","from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys,","masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] =","== 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes","ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder =","masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels # Instead of finding boxes, just","assert ( root.exists() ), f\"provided Panoptic path {root} does not","self.transforms is not None: img, target = self.transforms(img, target) return","in self.coco: for img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4]","masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\"id\"]","x: x[\"id\"]) # sanity check if \"annotations\" in self.coco: for","with open(ann_file, \"r\") as f: self.coco = json.load(f) # sort","ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w, h = img.size if \"segments_info\"","so that they are aligned with 'annotations' # i.e., in","boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) ==","root = Path(args.data_path) assert ( root.exists() ), f\"provided Panoptic path","import masks_to_boxes from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x):","= torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info else ann_info[\"id\"]]) if self.return_masks:","np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]]) masks = masks == ids[:,","import numpy as np import torch from PIL import Image","dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes)","\"image_id\" in ann_info else ann_info[\"id\"]]) if self.return_masks: target['masks'] = masks","info available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size']","self.transforms(img, target) return img, target except Exception as e: logging.error(ann_info)","ids = np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]]) masks = masks","the one from json info available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"])","# from util.box_ops import masks_to_boxes from .construction import make_construction_transforms import","2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels)","'area']: # target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if","key=lambda x: x[\"id\"]) # sanity check if \"annotations\" in self.coco:","\"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file = PATHS[image_set]","np import torch from PIL import Image from panopticapi.utils import","= x.unbind(-1) b = [xs, ys, (xs + w), (ys","check if \"annotations\" in self.coco: for img, ann in zip(self.coco[\"images\"],","json info available # target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes","labels = torch.tensor( # [ann[\"category_id\"] for ann in ann_info[\"segments_info\"]], #","that they are aligned with 'annotations' # i.e., in alphabetical","self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return height, width","dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, ) return","for ann in segments: if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area'])","field so that they are aligned with 'annotations' # i.e.,","as e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def","util.box_ops import masks_to_boxes from .construction import make_construction_transforms import logging def","= box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic: def","dtype=torch.int64, # ) target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if","def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height'] width","# Copyright (c) Facebook, Inc. and its affiliates. All Rights","labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels # Instead","Instead of finding boxes, just take the one from json","= boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)])","panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes from .construction","xs, ys, w, h = x.unbind(-1) b = [xs, ys,","in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder = img_folder","if \"image_id\" in ann_info else ann_info[\"id\"]]) if self.return_masks: target['masks'] =","= torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd)","masks == ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) #","self.coco = json.load(f) # sort 'images' field so that they","target) return img, target except Exception as e: logging.error(ann_info) raise","ann_file = PATHS[image_set] img_folder_path = root / img_folder ann_folder_path =","\"panoptic\" PATHS = { \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\",","ann in ann_info['segments_info']]) if self.transforms is not None: img, target","f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file","sanity check if \"annotations\" in self.coco: for img, ann in","def masks_to_boxes(segments): boxes = [] labels = [] iscrowd =","ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks = return_masks","= self.transforms(img, target) return img, target except Exception as e:","ann_info[\"segments_info\"]]) masks = masks == ids[:, None, None] masks =","labels # Instead of finding boxes, just take the one","target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)])","boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd'])","target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info","else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes)","labels = [] iscrowd = [] area = [] for","boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder,","boxes, just take the one from json info available #","= return_masks def __getitem__(self, idx): try: ann_info = ( self.coco[\"annotations\"][idx]","else self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path","ann_info else ann_info[\"id\"]]) if self.return_masks: target['masks'] = masks boxes, labels,","= ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks =","# ) target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\"","area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels # Instead of finding","json.load(f) # sort 'images' field so that they are aligned","class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with","boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] = labels #","0, 2, 2]) area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0","img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file","args): root = Path(args.data_path) assert ( root.exists() ), f\"provided Panoptic","from pathlib import Path import numpy as np import torch","they are aligned with 'annotations' # i.e., in alphabetical order","boxes = [] labels = [] iscrowd = [] area","torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels = []","# sort 'images' field so that they are aligned with","sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) # sanity check if \"annotations\" in","order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) # sanity check","f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root","json from pathlib import Path import numpy as np import","import logging def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1)","raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info","if \"segments_info\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks =","img_folder ann_folder_path = root / ann_folder ann_file = root /","= ( self.coco[\"annotations\"][idx] if \"annotations\" in self.coco else self.coco[\"images\"][idx] )","numpy as np import torch from PIL import Image from","as f: self.coco = json.load(f) # sort 'images' field so","{root} does not exist\" mode = \"panoptic\" PATHS = {","None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor(","= {} target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info else","/ ann_folder ann_file = root / ann_file dataset = ConstructionPanoptic(","# target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms","target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area","root / img_folder ann_folder_path = root / ann_folder ann_file =","root.exists() ), f\"provided Panoptic path {root} does not exist\" mode","import rgb2id # from util.box_ops import masks_to_boxes from .construction import","int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] =","of finding boxes, just take the one from json info","PIL import Image from panopticapi.utils import rgb2id # from util.box_ops","len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2])","self.coco[\"annotations\"][idx] if \"annotations\" in self.coco else self.coco[\"images\"][idx] ) img_path =","# if \"segments_info\" in ann_info: # for name in ['iscrowd',","= json.load(f) # sort 'images' field so that they are","labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area)","affiliates. All Rights Reserved import json from pathlib import Path","h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels","\"annotations\" in self.coco: for img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert","Image from panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes","= masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels'] =","int(w)]) target['iscrowd'] = iscrowd target['area'] = area # if \"segments_info\"","def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx]","= np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]]) masks = masks ==","if self.transforms is not None: img, target = self.transforms(img, target)","ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, )","open(ann_file, \"r\") as f: self.coco = json.load(f) # sort 'images'","ann in ann_info[\"segments_info\"]], # dtype=torch.int64, # ) target = {}","masks = masks == ids[:, None, None] masks = torch.as_tensor(masks,","= torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is not","np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\"id\"] for ann","torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\"category_id\"] for ann","None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( #","= [] for ann in segments: if len(ann[\"bbox\"]) == 4:","ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file = ann_file","\".jpg\") ann_path = Path(self.ann_folder) / ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w,","torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class","[] area = [] for ann in segments: if len(ann[\"bbox\"])","else ann_info[\"id\"]]) if self.return_masks: target['masks'] = masks boxes, labels, iscrowd,","sort 'images' field so that they are aligned with 'annotations'","self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms","= Path(args.data_path) assert ( root.exists() ), f\"provided Panoptic path {root}","\"segments_info\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks)","= [] iscrowd = [] area = [] for ann","PATHS[image_set] img_folder_path = root / img_folder ann_folder_path = root /","# [ann[\"category_id\"] for ann in ann_info[\"segments_info\"]], # dtype=torch.int64, # )","torch from PIL import Image from panopticapi.utils import rgb2id #","dim=-1) def masks_to_boxes(segments): boxes = [] labels = [] iscrowd","2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels","'annotations' # i.e., in alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda","= img.size if \"segments_info\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32)","== 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4)","\"annotations\" in self.coco else self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) /","boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic:","in self.coco else self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\",","Facebook, Inc. and its affiliates. All Rights Reserved import json","assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder","if \"annotations\" in self.coco: for img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]):","== 0 and len(labels) == 0: boxes.append([0, 0, 2, 2])","= area # if \"segments_info\" in ann_info: # for name","__len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height","e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self,","masks_to_boxes(segments): boxes = [] labels = [] iscrowd = []","Path(self.ann_folder) / ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w, h = img.size","alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x: x[\"id\"]) # sanity","img, ann in zip(self.coco[\"images\"], self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder","area = [] for ann in segments: if len(ann[\"bbox\"]) ==","torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels,","dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area =","= self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return height,","w, h = x.unbind(-1) b = [xs, ys, (xs +","self.ann_folder = ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks","= [] area = [] for ann in segments: if","from panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes from","logging def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b","masks_to_boxes from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs,","and its affiliates. All Rights Reserved import json from pathlib","def __getitem__(self, idx): try: ann_info = ( self.coco[\"annotations\"][idx] if \"annotations\"","__init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \"r\") as","(\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path","{ \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), }","with 'annotations' # i.e., in alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"],","Image.open(img_path).convert(\"RGB\") w, h = img.size if \"segments_info\" in ann_info: masks","labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file,","= iscrowd target['area'] = area # if \"segments_info\" in ann_info:","), f\"provided Panoptic path {root} does not exist\" mode =","i.e., in alphabetical order self.coco[\"images\"] = sorted(self.coco[\"images\"], key=lambda x: x[\"id\"])","= transforms self.return_masks = return_masks def __getitem__(self, idx): try: ann_info","Path(args.data_path) assert ( root.exists() ), f\"provided Panoptic path {root} does","self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"])","ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file,","(ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes =","= torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area #","= masks == ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8)","f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file =","[xs, ys, (xs + w), (ys + h)] return torch.stack(b,","== ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels","f: self.coco = json.load(f) # sort 'images' field so that","(c) Facebook, Inc. and its affiliates. All Rights Reserved import","torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is not None:","} img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root /","torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area","boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd']","img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \"r\") as f:","in ann_info[\"segments_info\"]]) masks = masks == ids[:, None, None] masks","ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels =","__getitem__(self, idx): try: ann_info = ( self.coco[\"annotations\"][idx] if \"annotations\" in","transforms self.return_masks = return_masks def __getitem__(self, idx): try: ann_info =","= torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes","self.return_masks = return_masks def __getitem__(self, idx): try: ann_info = (","img_info['height'] width = img_info['width'] return height, width def build(image_set, args):","/ img_folder ann_folder_path = root / ann_folder ann_file = root","def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \"r\")","pathlib import Path import numpy as np import torch from","ys, (xs + w), (ys + h)] return torch.stack(b, dim=-1)","/ ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder) / ann_info[\"file_name\"] img =","just take the one from json info available # target[\"boxes\"]","target['area'] = area # if \"segments_info\" in ann_info: # for","try: ann_info = ( self.coco[\"annotations\"][idx] if \"annotations\" in self.coco else","= torch.tensor( # [ann[\"category_id\"] for ann in ann_info[\"segments_info\"]], # dtype=torch.int64,","( root.exists() ), f\"provided Panoptic path {root} does not exist\"","area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True):","torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area']","= ann_file self.transforms = transforms self.return_masks = return_masks def __getitem__(self,","= masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size']","ann_file = root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path,","ann_info: # for name in ['iscrowd', 'area']: # target[name] =","aligned with 'annotations' # i.e., in alphabetical order self.coco[\"images\"] =","Reserved import json from pathlib import Path import numpy as","mode = \"panoptic\" PATHS = { \"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"),","width = img_info['width'] return height, width def build(image_set, args): root","except Exception as e: logging.error(ann_info) raise e def __len__(self): return","labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) == 0:","if \"annotations\" in self.coco else self.coco[\"images\"][idx] ) img_path = Path(self.img_folder)","self.coco[\"annotations\"]): assert img[\"file_name\"][:-4] == ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder =","f\"provided Panoptic path {root} does not exist\" mode = \"panoptic\"","return height, width def build(image_set, args): root = Path(args.data_path) assert","ann_folder, ann_file = PATHS[image_set] img_folder_path = root / img_folder ann_folder_path","Path import numpy as np import torch from PIL import","for ann in ann_info[\"segments_info\"]]) masks = masks == ids[:, None,","f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path =","for ann in ann_info['segments_info']]) if self.transforms is not None: img,","ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder) / ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\")","Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path = Path(self.ann_folder) / ann_info[\"file_name\"] img","# target[\"boxes\"] = masks_to_boxes(ann_info[\"segments_info\"]) target[\"boxes\"] = boxes target['size'] = torch.as_tensor([int(h),","area.append(4) labels.append(ann[\"category_id\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) ==","target['image_id'] = torch.tensor([ann_info['image_id'] if \"image_id\" in ann_info else ann_info[\"id\"]]) if","rgb2id # from util.box_ops import masks_to_boxes from .construction import make_construction_transforms","h = img.size if \"segments_info\" in ann_info: masks = np.asarray(Image.open(ann_path),","in ann_info[\"segments_info\"]], # dtype=torch.int64, # ) target = {} target['image_id']","Inc. and its affiliates. All Rights Reserved import json from","not None: img, target = self.transforms(img, target) return img, target","not exist\" mode = \"panoptic\" PATHS = { \"train\": (\"images\",","area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64)","/ ann_info[\"file_name\"] img = Image.open(img_path).convert(\"RGB\") w, h = img.size if","img_info['width'] return height, width def build(image_set, args): root = Path(args.data_path)","b = [xs, ys, (xs + w), (ys + h)]","= root / ann_folder ann_file = root / ann_file dataset","iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None,","height, width def build(image_set, args): root = Path(args.data_path) assert (","root / ann_folder ann_file = root / ann_file dataset =","ann_info[\"segments_info\"]], # dtype=torch.int64, # ) target = {} target['image_id'] =","Rights Reserved import json from pathlib import Path import numpy","target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] =","= PATHS[image_set] img_folder_path = root / img_folder ann_folder_path = root","import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w, h","target['masks'] = masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\"segments_info\"]) target['labels']","(\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder, ann_folder,","\"r\") as f: self.coco = json.load(f) # sort 'images' field","= np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\"id\"] for","segments: if len(ann[\"bbox\"]) == 4: boxes.append(ann[\"bbox\"]) area.append(ann['area']) else: boxes.append([0, 0,","self.transforms = transforms self.return_masks = return_masks def __getitem__(self, idx): try:","['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']])","import Image from panopticapi.utils import rgb2id # from util.box_ops import","rgb2id(masks) ids = np.array([ann[\"id\"] for ann in ann_info[\"segments_info\"]]) masks =","ann_info = ( self.coco[\"annotations\"][idx] if \"annotations\" in self.coco else self.coco[\"images\"][idx]","h = x.unbind(-1) b = [xs, ys, (xs + w),","= [] labels = [] iscrowd = [] area =","its affiliates. All Rights Reserved import json from pathlib import","and len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4)","== ann[\"file_name\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file =","are aligned with 'annotations' # i.e., in alphabetical order self.coco[\"images\"]","target['iscrowd'] = iscrowd target['area'] = area # if \"segments_info\" in","dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\"id\"] for ann in","\"train\": (\"images\", f\"{mode}\", f\"{mode}.json\"), \"val\": (\"images\", f\"val_{mode}\", f\"val_{mode}.json\"), } img_folder,","target = self.transforms(img, target) return img, target except Exception as","torch.tensor( # [ann[\"category_id\"] for ann in ann_info[\"segments_info\"]], # dtype=torch.int64, #"],"string":"[\n \"len(boxes) == 0 and len(labels) == 0: boxes.append([0, 0, 2,\",\n \"self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path =\",\n \"= Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size if \\\"segments_info\\\" in ann_info:\",\n \"# Instead of finding boxes, just take the one from\",\n \"self.ann_file = ann_file self.transforms = transforms self.return_masks = return_masks def\",\n \"box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b = [xs,\",\n \"= img_info['width'] return height, width def build(image_set, args): root =\",\n \"/ ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks,\",\n \"build(image_set, args): root = Path(args.data_path) assert ( root.exists() ), f\\\"provided\",\n \"= masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels # Instead of finding boxes,\",\n \"4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"])\",\n \"w, h = img.size if \\\"segments_info\\\" in ann_info: masks =\",\n \"target[\\\"boxes\\\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h),\",\n \"area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd,\",\n \"# for name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name]\",\n \"make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w, h =\",\n \"= { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"),\",\n \"[] for ann in segments: if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"])\",\n \"{} target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]])\",\n \"iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return\",\n \"iscrowd target['area'] = area # if \\\"segments_info\\\" in ann_info: #\",\n \"2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and\",\n \"path {root} does not exist\\\" mode = \\\"panoptic\\\" PATHS =\",\n \"ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \\\"r\\\") as f: self.coco\",\n \"logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx):\",\n \"ys, w, h = x.unbind(-1) b = [xs, ys, (xs\",\n \") target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in\",\n \"# dtype=torch.int64, # ) target = {} target['image_id'] = torch.tensor([ann_info['image_id']\",\n \"\\\"segments_info\\\" in ann_info: # for name in ['iscrowd', 'area']: #\",\n \"= \\\"panoptic\\\" PATHS = { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\":\",\n \"0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes =\",\n \"= torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area\",\n \"iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) == 0: boxes.append([0,\",\n \"self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) # sanity check if\",\n \"target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is\",\n \"if \\\"segments_info\\\" in ann_info: # for name in ['iscrowd', 'area']:\",\n \"zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder\",\n \"= img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms =\",\n \"return_masks=True): with open(ann_file, \\\"r\\\") as f: self.coco = json.load(f) #\",\n \"[ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # ) target\",\n \"for name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for\",\n \"+ w), (ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments):\",\n \"img_info = self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return\",\n \"return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height =\",\n \"img_folder_path = root / img_folder ann_folder_path = root / ann_folder\",\n \"def build(image_set, args): root = Path(args.data_path) assert ( root.exists() ),\",\n \"= Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"]\",\n \"finding boxes, just take the one from json info available\",\n \"= Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w, h =\",\n \"idx): img_info = self.coco['images'][idx] height = img_info['height'] width = img_info['width']\",\n \"= torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes,\",\n \"as np import torch from PIL import Image from panopticapi.utils\",\n \"ann_info['segments_info']]) if self.transforms is not None: img, target = self.transforms(img,\",\n \"len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height']\",\n \"iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd\",\n \"boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes,\",\n \"img, target except Exception as e: logging.error(ann_info) raise e def\",\n \"'images' field so that they are aligned with 'annotations' #\",\n \"from PIL import Image from panopticapi.utils import rgb2id # from\",\n \"[] iscrowd = [] area = [] for ann in\",\n \"+ h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = []\",\n \"in ann_info: # for name in ['iscrowd', 'area']: # target[name]\",\n \"target except Exception as e: logging.error(ann_info) raise e def __len__(self):\",\n \"img.size if \\\"segments_info\\\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks\",\n \"= torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\\\"category_id\\\"] for\",\n \"ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids =\",\n \"if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0, 2,\",\n \"does not exist\\\" mode = \\\"panoptic\\\" PATHS = { \\\"train\\\":\",\n \"# labels = torch.tensor( # [ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]],\",\n \"= [xs, ys, (xs + w), (ys + h)] return\",\n \"[] labels = [] iscrowd = [] area = []\",\n \"if self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area =\",\n \"= sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) # sanity check if \\\"annotations\\\"\",\n \"0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64)\",\n \"# i.e., in alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x:\",\n \"in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann in\",\n \"in segments: if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0,\",\n \"transforms=None, return_masks=True): with open(ann_file, \\\"r\\\") as f: self.coco = json.load(f)\",\n \"for img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4]\",\n \"ann_folder ann_file = root / ann_file dataset = ConstructionPanoptic( img_folder_path,\",\n \") img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder)\",\n \"one from json info available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"]\",\n \"in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids\",\n \"x.unbind(-1) b = [xs, ys, (xs + w), (ys +\",\n \"= root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file,\",\n \"len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0)\",\n \"ann_file self.transforms = transforms self.return_masks = return_masks def __getitem__(self, idx):\",\n \"in ann_info['segments_info']]) if self.transforms is not None: img, target =\",\n \"PATHS = { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\",\",\n \"ann_folder_path = root / ann_folder ann_file = root / ann_file\",\n \"return_masks def __getitem__(self, idx): try: ann_info = ( self.coco[\\\"annotations\\\"][idx] if\",\n \"img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms = transforms\",\n \"ann in segments: if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else:\",\n \".construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w,\",\n \"masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels\",\n \"( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx] ) img_path\",\n \"area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if\",\n \"ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w, h\",\n \"self.coco: for img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] ==\",\n \"boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd =\",\n \"for ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # ) target =\",\n \"box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self,\",\n \"return boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder,\",\n \"img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root / img_folder\",\n \"root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set),\",\n \"masks = rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]])\",\n \"= img_info['height'] width = img_info['width'] return height, width def build(image_set,\",\n \"Panoptic path {root} does not exist\\\" mode = \\\"panoptic\\\" PATHS\",\n \"def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b =\",\n \"masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\\\"category_id\\\"]\",\n \"img, target = self.transforms(img, target) return img, target except Exception\",\n \"iscrowd = [] area = [] for ann in segments:\",\n \"in ann_info else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] = masks boxes,\",\n \"exist\\\" mode = \\\"panoptic\\\" PATHS = { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\",\",\n \"= ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, ) return dataset\",\n \"is not None: img, target = self.transforms(img, target) return img,\",\n \"Exception as e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images'])\",\n \"All Rights Reserved import json from pathlib import Path import\",\n \"get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height'] width =\",\n \"= torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd\",\n \"self.coco else self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\")\",\n \"None: img, target = self.transforms(img, target) return img, target except\",\n \"e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info =\",\n \"(xs + w), (ys + h)] return torch.stack(b, dim=-1) def\",\n \"return img, target except Exception as e: logging.error(ann_info) raise e\",\n \"width def build(image_set, args): root = Path(args.data_path) assert ( root.exists()\",\n \"ann in ann_info[\\\"segments_info\\\"]]) masks = masks == ids[:, None, None]\",\n \"2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels =\",\n \"torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes =\",\n \"# sanity check if \\\"annotations\\\" in self.coco: for img, ann\",\n \"from util.box_ops import masks_to_boxes from .construction import make_construction_transforms import logging\",\n \"iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels # Instead of\",\n \"img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder) /\",\n \"in alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) #\",\n \"name in ['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann\",\n \"take the one from json info available # target[\\\"boxes\\\"] =\",\n \"available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size'] =\",\n \"= labels # Instead of finding boxes, just take the\",\n \"dtype=torch.uint8) # labels = torch.tensor( # [ann[\\\"category_id\\\"] for ann in\",\n \"= rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]]) masks\",\n \"height = img_info['height'] width = img_info['width'] return height, width def\",\n \"Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved\",\n \"area # if \\\"segments_info\\\" in ann_info: # for name in\",\n \"return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels =\",\n \"torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks']\",\n \"torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area # if\",\n \"import json from pathlib import Path import numpy as np\",\n \"idx): try: ann_info = ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco\",\n \"x[\\\"id\\\"]) # sanity check if \\\"annotations\\\" in self.coco: for img,\",\n \"import Path import numpy as np import torch from PIL\",\n \"img = Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size if \\\"segments_info\\\" in\",\n \"0 and len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1)\",\n \"labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels,\",\n \"ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area\",\n \"w), (ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes\",\n \"= root / img_folder ann_folder_path = root / ann_folder ann_file\",\n \"if len(boxes) == 0 and len(labels) == 0: boxes.append([0, 0,\",\n \"target['labels'] = labels # Instead of finding boxes, just take\",\n \"import torch from PIL import Image from panopticapi.utils import rgb2id\",\n \"ann_file, transforms=None, return_masks=True): with open(ann_file, \\\"r\\\") as f: self.coco =\",\n \"from json info available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] =\",\n \"from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys,\",\n \"masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] =\",\n \"== 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes\",\n \"ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder =\",\n \"masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels # Instead of finding boxes, just\",\n \"assert ( root.exists() ), f\\\"provided Panoptic path {root} does not\",\n \"self.transforms is not None: img, target = self.transforms(img, target) return\",\n \"in self.coco: for img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4]\",\n \"masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\\\"id\\\"]\",\n \"x: x[\\\"id\\\"]) # sanity check if \\\"annotations\\\" in self.coco: for\",\n \"with open(ann_file, \\\"r\\\") as f: self.coco = json.load(f) # sort\",\n \"ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size if \\\"segments_info\\\"\",\n \"so that they are aligned with 'annotations' # i.e., in\",\n \"boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) ==\",\n \"root = Path(args.data_path) assert ( root.exists() ), f\\\"provided Panoptic path\",\n \"import masks_to_boxes from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x):\",\n \"= torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]]) if self.return_masks:\",\n \"np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]]) masks = masks == ids[:,\",\n \"import numpy as np import torch from PIL import Image\",\n \"dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes)\",\n \"\\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] = masks\",\n \"info available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size']\",\n \"self.transforms(img, target) return img, target except Exception as e: logging.error(ann_info)\",\n \"ids = np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]]) masks = masks\",\n \"the one from json info available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"])\",\n \"# from util.box_ops import masks_to_boxes from .construction import make_construction_transforms import\",\n \"2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels)\",\n \"'area']: # target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if\",\n \"key=lambda x: x[\\\"id\\\"]) # sanity check if \\\"annotations\\\" in self.coco:\",\n \"\\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file = PATHS[image_set]\",\n \"np import torch from PIL import Image from panopticapi.utils import\",\n \"= x.unbind(-1) b = [xs, ys, (xs + w), (ys\",\n \"check if \\\"annotations\\\" in self.coco: for img, ann in zip(self.coco[\\\"images\\\"],\",\n \"json info available # target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes\",\n \"labels = torch.tensor( # [ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]], #\",\n \"that they are aligned with 'annotations' # i.e., in alphabetical\",\n \"self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return height, width\",\n \"dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, ) return\",\n \"for ann in segments: if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area'])\",\n \"field so that they are aligned with 'annotations' # i.e.,\",\n \"as e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def\",\n \"util.box_ops import masks_to_boxes from .construction import make_construction_transforms import logging def\",\n \"= box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic: def\",\n \"dtype=torch.int64, # ) target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if\",\n \"def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height = img_info['height'] width\",\n \"# Copyright (c) Facebook, Inc. and its affiliates. All Rights\",\n \"labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels # Instead\",\n \"Instead of finding boxes, just take the one from json\",\n \"= boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)])\",\n \"panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes from .construction\",\n \"xs, ys, w, h = x.unbind(-1) b = [xs, ys,\",\n \"in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder\",\n \"if \\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] =\",\n \"= torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd)\",\n \"masks == ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) #\",\n \"self.coco = json.load(f) # sort 'images' field so that they\",\n \"target) return img, target except Exception as e: logging.error(ann_info) raise\",\n \"ann_file = PATHS[image_set] img_folder_path = root / img_folder ann_folder_path =\",\n \"\\\"panoptic\\\" PATHS = { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\",\",\n \"ann in ann_info['segments_info']]) if self.transforms is not None: img, target\",\n \"f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file\",\n \"sanity check if \\\"annotations\\\" in self.coco: for img, ann in\",\n \"def masks_to_boxes(segments): boxes = [] labels = [] iscrowd =\",\n \"ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks = return_masks\",\n \"= self.transforms(img, target) return img, target except Exception as e:\",\n \"ann_info[\\\"segments_info\\\"]]) masks = masks == ids[:, None, None] masks =\",\n \"labels # Instead of finding boxes, just take the one\",\n \"target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)])\",\n \"boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd'])\",\n \"target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info\",\n \"else: boxes.append([0, 0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes)\",\n \"labels = [] iscrowd = [] area = [] for\",\n \"boxes, labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder,\",\n \"boxes, just take the one from json info available #\",\n \"= return_masks def __getitem__(self, idx): try: ann_info = ( self.coco[\\\"annotations\\\"][idx]\",\n \"else self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path\",\n \"ann_info else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] = masks boxes, labels,\",\n \"= ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks =\",\n \"# ) target = {} target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\"\",\n \"area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels # Instead of finding\",\n \"json.load(f) # sort 'images' field so that they are aligned\",\n \"class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with\",\n \"boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] = labels #\",\n \"0, 2, 2]) area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0\",\n \"img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file\",\n \"args): root = Path(args.data_path) assert ( root.exists() ), f\\\"provided Panoptic\",\n \"from pathlib import Path import numpy as np import torch\",\n \"they are aligned with 'annotations' # i.e., in alphabetical order\",\n \"boxes = [] labels = [] iscrowd = [] area\",\n \"torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels = []\",\n \"# sort 'images' field so that they are aligned with\",\n \"sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) # sanity check if \\\"annotations\\\" in\",\n \"order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) # sanity check\",\n \"f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root\",\n \"json from pathlib import Path import numpy as np import\",\n \"import logging def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1)\",\n \"raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info\",\n \"if \\\"segments_info\\\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks =\",\n \"img_folder ann_folder_path = root / ann_folder ann_file = root /\",\n \"= ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx] )\",\n \"numpy as np import torch from PIL import Image from\",\n \"as f: self.coco = json.load(f) # sort 'images' field so\",\n \"{root} does not exist\\\" mode = \\\"panoptic\\\" PATHS = {\",\n \"None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor(\",\n \"= {} target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info else\",\n \"/ ann_folder ann_file = root / ann_file dataset = ConstructionPanoptic(\",\n \"# target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms\",\n \"target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area\",\n \"root / img_folder ann_folder_path = root / ann_folder ann_file =\",\n \"root.exists() ), f\\\"provided Panoptic path {root} does not exist\\\" mode\",\n \"import rgb2id # from util.box_ops import masks_to_boxes from .construction import\",\n \"int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] =\",\n \"of finding boxes, just take the one from json info\",\n \"PIL import Image from panopticapi.utils import rgb2id # from util.box_ops\",\n \"len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2])\",\n \"self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx] ) img_path =\",\n \"# if \\\"segments_info\\\" in ann_info: # for name in ['iscrowd',\",\n \"= json.load(f) # sort 'images' field so that they are\",\n \"labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area)\",\n \"affiliates. All Rights Reserved import json from pathlib import Path\",\n \"h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes = [] labels\",\n \"\\\"annotations\\\" in self.coco: for img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert\",\n \"Image from panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes\",\n \"= masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels'] =\",\n \"int(w)]) target['iscrowd'] = iscrowd target['area'] = area # if \\\"segments_info\\\"\",\n \"def __len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx]\",\n \"= np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]]) masks = masks ==\",\n \"if self.transforms is not None: img, target = self.transforms(img, target)\",\n \"ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path, ann_file, transforms=make_construction_transforms(image_set), return_masks=args.masks, )\",\n \"open(ann_file, \\\"r\\\") as f: self.coco = json.load(f) # sort 'images'\",\n \"ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # ) target = {}\",\n \"masks = masks == ids[:, None, None] masks = torch.as_tensor(masks,\",\n \"= torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is not\",\n \"np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for ann\",\n \"torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( # [ann[\\\"category_id\\\"] for ann\",\n \"None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels = torch.tensor( #\",\n \"= [] for ann in segments: if len(ann[\\\"bbox\\\"]) == 4:\",\n \"ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file = ann_file\",\n \"\\\".jpg\\\") ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w,\",\n \"torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class\",\n \"[] area = [] for ann in segments: if len(ann[\\\"bbox\\\"])\",\n \"else ann_info[\\\"id\\\"]]) if self.return_masks: target['masks'] = masks boxes, labels, iscrowd,\",\n \"sort 'images' field so that they are aligned with 'annotations'\",\n \"self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file = ann_file self.transforms\",\n \"= Path(args.data_path) assert ( root.exists() ), f\\\"provided Panoptic path {root}\",\n \"\\\"segments_info\\\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks)\",\n \"= [] iscrowd = [] area = [] for ann\",\n \"PATHS[image_set] img_folder_path = root / img_folder ann_folder_path = root /\",\n \"# [ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # )\",\n \"torch from PIL import Image from panopticapi.utils import rgb2id #\",\n \"dim=-1) def masks_to_boxes(segments): boxes = [] labels = [] iscrowd\",\n \"2, 2]) labels.append(1) area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels\",\n \"'annotations' # i.e., in alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda\",\n \"= img.size if \\\"segments_info\\\" in ann_info: masks = np.asarray(Image.open(ann_path), dtype=np.uint32)\",\n \"== 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0, 2, 2]) area.append(4)\",\n \"\\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder) /\",\n \"boxes = box_xywh_to_xyxy(boxes) return boxes, labels, iscrowd, area class ConstructionPanoptic:\",\n \"in self.coco else self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\",\",\n \"Facebook, Inc. and its affiliates. All Rights Reserved import json\",\n \"assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder\",\n \"if \\\"annotations\\\" in self.coco: for img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]):\",\n \"== 0 and len(labels) == 0: boxes.append([0, 0, 2, 2])\",\n \"= area # if \\\"segments_info\\\" in ann_info: # for name\",\n \"__len__(self): return len(self.coco['images']) def get_height_and_width(self, idx): img_info = self.coco['images'][idx] height\",\n \"e: logging.error(ann_info) raise e def __len__(self): return len(self.coco['images']) def get_height_and_width(self,\",\n \"masks_to_boxes(segments): boxes = [] labels = [] iscrowd = []\",\n \"Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size\",\n \"alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"]) # sanity\",\n \"img, ann in zip(self.coco[\\\"images\\\"], self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder\",\n \"area = [] for ann in segments: if len(ann[\\\"bbox\\\"]) ==\",\n \"torch.tensor(iscrowd) area = torch.tensor(area) boxes = box_xywh_to_xyxy(boxes) return boxes, labels,\",\n \"dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area =\",\n \"= self.coco['images'][idx] height = img_info['height'] width = img_info['width'] return height,\",\n \"w, h = x.unbind(-1) b = [xs, ys, (xs +\",\n \"self.ann_folder = ann_folder self.ann_file = ann_file self.transforms = transforms self.return_masks\",\n \"= [] area = [] for ann in segments: if\",\n \"from panopticapi.utils import rgb2id # from util.box_ops import masks_to_boxes from\",\n \"logging def box_xywh_to_xyxy(x): xs, ys, w, h = x.unbind(-1) b\",\n \"masks_to_boxes from .construction import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs,\",\n \"and its affiliates. All Rights Reserved import json from pathlib\",\n \"def __getitem__(self, idx): try: ann_info = ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\"\",\n \"__init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \\\"r\\\") as\",\n \"(\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path\",\n \"{ \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), }\",\n \"with 'annotations' # i.e., in alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"],\",\n \"Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size if \\\"segments_info\\\" in ann_info: masks\",\n \"labels, iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file,\",\n \"= iscrowd target['area'] = area # if \\\"segments_info\\\" in ann_info:\",\n \"), f\\\"provided Panoptic path {root} does not exist\\\" mode =\",\n \"i.e., in alphabetical order self.coco[\\\"images\\\"] = sorted(self.coco[\\\"images\\\"], key=lambda x: x[\\\"id\\\"])\",\n \"= transforms self.return_masks = return_masks def __getitem__(self, idx): try: ann_info\",\n \"Path(args.data_path) assert ( root.exists() ), f\\\"provided Panoptic path {root} does\",\n \"self.return_masks: target['masks'] = masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"])\",\n \"ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file,\",\n \"(ys + h)] return torch.stack(b, dim=-1) def masks_to_boxes(segments): boxes =\",\n \"= torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area'] = area #\",\n \"= masks == ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8)\",\n \"f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file =\",\n \"[xs, ys, (xs + w), (ys + h)] return torch.stack(b,\",\n \"== ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels\",\n \"f: self.coco = json.load(f) # sort 'images' field so that\",\n \"(c) Facebook, Inc. and its affiliates. All Rights Reserved import\",\n \"torch.tensor([ann[name] for ann in ann_info['segments_info']]) if self.transforms is not None:\",\n \"} img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path = root /\",\n \"torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area\",\n \"boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd']\",\n \"img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \\\"r\\\") as f:\",\n \"in ann_info[\\\"segments_info\\\"]]) masks = masks == ids[:, None, None] masks\",\n \"ids[:, None, None] masks = torch.as_tensor(masks, dtype=torch.uint8) # labels =\",\n \"__getitem__(self, idx): try: ann_info = ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in\",\n \"transforms self.return_masks = return_masks def __getitem__(self, idx): try: ann_info =\",\n \"= torch.tensor(labels, dtype=torch.int64) iscrowd = torch.tensor(iscrowd) area = torch.tensor(area) boxes\",\n \"self.return_masks = return_masks def __getitem__(self, idx): try: ann_info = (\",\n \"img_info['height'] width = img_info['width'] return height, width def build(image_set, args):\",\n \"/ img_folder ann_folder_path = root / ann_folder ann_file = root\",\n \"def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True): with open(ann_file, \\\"r\\\")\",\n \"pathlib import Path import numpy as np import torch from\",\n \"ys, (xs + w), (ys + h)] return torch.stack(b, dim=-1)\",\n \"/ ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img =\",\n \"just take the one from json info available # target[\\\"boxes\\\"]\",\n \"target['area'] = area # if \\\"segments_info\\\" in ann_info: # for\",\n \"try: ann_info = ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco else\",\n \"= torch.tensor( # [ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64,\",\n \"( root.exists() ), f\\\"provided Panoptic path {root} does not exist\\\"\",\n \"area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None, return_masks=True):\",\n \"torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] = iscrowd target['area']\",\n \"= ann_file self.transforms = transforms self.return_masks = return_masks def __getitem__(self,\",\n \"= masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size']\",\n \"ann_file = root / ann_file dataset = ConstructionPanoptic( img_folder_path, ann_folder_path,\",\n \"ann_info: # for name in ['iscrowd', 'area']: # target[name] =\",\n \"aligned with 'annotations' # i.e., in alphabetical order self.coco[\\\"images\\\"] =\",\n \"Reserved import json from pathlib import Path import numpy as\",\n \"mode = \\\"panoptic\\\" PATHS = { \\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"),\",\n \"width = img_info['width'] return height, width def build(image_set, args): root\",\n \"except Exception as e: logging.error(ann_info) raise e def __len__(self): return\",\n \"labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) == 0:\",\n \"if \\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx] ) img_path = Path(self.img_folder)\",\n \"self.coco[\\\"annotations\\\"]): assert img[\\\"file_name\\\"][:-4] == ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder =\",\n \"f\\\"provided Panoptic path {root} does not exist\\\" mode = \\\"panoptic\\\"\",\n \"return height, width def build(image_set, args): root = Path(args.data_path) assert\",\n \"ann_folder, ann_file = PATHS[image_set] img_folder_path = root / img_folder ann_folder_path\",\n \"Path import numpy as np import torch from PIL import\",\n \"for ann in ann_info[\\\"segments_info\\\"]]) masks = masks == ids[:, None,\",\n \"f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder, ann_file = PATHS[image_set] img_folder_path =\",\n \"for ann in ann_info['segments_info']]) if self.transforms is not None: img,\",\n \"ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\")\",\n \"Path(self.img_folder) / ann_info[\\\"file_name\\\"].replace(\\\".png\\\", \\\".jpg\\\") ann_path = Path(self.ann_folder) / ann_info[\\\"file_name\\\"] img\",\n \"# target[\\\"boxes\\\"] = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target[\\\"boxes\\\"] = boxes target['size'] = torch.as_tensor([int(h),\",\n \"area.append(4) labels.append(ann[\\\"category_id\\\"]) iscrowd.append(ann['iscrowd']) if len(boxes) == 0 and len(labels) ==\",\n \"target['image_id'] = torch.tensor([ann_info['image_id'] if \\\"image_id\\\" in ann_info else ann_info[\\\"id\\\"]]) if\",\n \"rgb2id # from util.box_ops import masks_to_boxes from .construction import make_construction_transforms\",\n \"h = img.size if \\\"segments_info\\\" in ann_info: masks = np.asarray(Image.open(ann_path),\",\n \"in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # ) target = {} target['image_id']\",\n \"Inc. and its affiliates. All Rights Reserved import json from\",\n \"not None: img, target = self.transforms(img, target) return img, target\",\n \"not exist\\\" mode = \\\"panoptic\\\" PATHS = { \\\"train\\\": (\\\"images\\\",\",\n \"area.append(4) iscrowd.append(0) boxes = torch.tensor(boxes, dtype=torch.int64) labels = torch.tensor(labels, dtype=torch.int64)\",\n \"/ ann_info[\\\"file_name\\\"] img = Image.open(img_path).convert(\\\"RGB\\\") w, h = img.size if\",\n \"img_info['width'] return height, width def build(image_set, args): root = Path(args.data_path)\",\n \"b = [xs, ys, (xs + w), (ys + h)]\",\n \"= root / ann_folder ann_file = root / ann_file dataset\",\n \"iscrowd, area class ConstructionPanoptic: def __init__(self, img_folder, ann_folder, ann_file, transforms=None,\",\n \"height, width def build(image_set, args): root = Path(args.data_path) assert (\",\n \"root / ann_folder ann_file = root / ann_file dataset =\",\n \"ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, # ) target = {} target['image_id'] =\",\n \"Rights Reserved import json from pathlib import Path import numpy\",\n \"target['size'] = torch.as_tensor([int(h), int(w)]) target['orig_size'] = torch.as_tensor([int(h), int(w)]) target['iscrowd'] =\",\n \"= PATHS[image_set] img_folder_path = root / img_folder ann_folder_path = root\",\n \"import make_construction_transforms import logging def box_xywh_to_xyxy(x): xs, ys, w, h\",\n \"target['masks'] = masks boxes, labels, iscrowd, area = masks_to_boxes(ann_info[\\\"segments_info\\\"]) target['labels']\",\n \"(\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder, ann_folder,\",\n \"\\\"r\\\") as f: self.coco = json.load(f) # sort 'images' field\",\n \"= np.asarray(Image.open(ann_path), dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for\",\n \"segments: if len(ann[\\\"bbox\\\"]) == 4: boxes.append(ann[\\\"bbox\\\"]) area.append(ann['area']) else: boxes.append([0, 0,\",\n \"self.transforms = transforms self.return_masks = return_masks def __getitem__(self, idx): try:\",\n \"['iscrowd', 'area']: # target[name] = torch.tensor([ann[name] for ann in ann_info['segments_info']])\",\n \"import Image from panopticapi.utils import rgb2id # from util.box_ops import\",\n \"rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for ann in ann_info[\\\"segments_info\\\"]]) masks =\",\n \"ann_info = ( self.coco[\\\"annotations\\\"][idx] if \\\"annotations\\\" in self.coco else self.coco[\\\"images\\\"][idx]\",\n \"h = x.unbind(-1) b = [xs, ys, (xs + w),\",\n \"= [] labels = [] iscrowd = [] area =\",\n \"its affiliates. All Rights Reserved import json from pathlib import\",\n \"and len(labels) == 0: boxes.append([0, 0, 2, 2]) labels.append(1) area.append(4)\",\n \"== ann[\\\"file_name\\\"][:-4] self.img_folder = img_folder self.ann_folder = ann_folder self.ann_file =\",\n \"are aligned with 'annotations' # i.e., in alphabetical order self.coco[\\\"images\\\"]\",\n \"target['iscrowd'] = iscrowd target['area'] = area # if \\\"segments_info\\\" in\",\n \"dtype=np.uint32) masks = rgb2id(masks) ids = np.array([ann[\\\"id\\\"] for ann in\",\n \"\\\"train\\\": (\\\"images\\\", f\\\"{mode}\\\", f\\\"{mode}.json\\\"), \\\"val\\\": (\\\"images\\\", f\\\"val_{mode}\\\", f\\\"val_{mode}.json\\\"), } img_folder,\",\n \"target = self.transforms(img, target) return img, target except Exception as\",\n \"torch.tensor( # [ann[\\\"category_id\\\"] for ann in ann_info[\\\"segments_info\\\"]], # dtype=torch.int64, #\"\n]","isConversational":false}}},{"rowIdx":686,"cells":{"ngram":{"kind":"list like","value":["middle = left + (right - left) // 2 if","equals to nth-item \"\"\" if len(array) <= index: return index","search recursively here if left is None and right is","0, len(array) - 1 if len(array) == 0: return None","O(n) because you are returning the function continuously until index","return middle elif item > array[middle]: left = middle +","1, right) else: return binary_search_recursive(array, item, left, middle - 1)","item): \"\"\"return the index of item in sorted array or","TIME AND SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array, item): \"\"\"return the","+ 1 else: right = middle - 1 return None","def binary_search_recursive(array, item, left=None, right=None): \"\"\"Time Complexity: O(log*n) Space Complexity:","Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack space\"\"\" #","FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array, item):","the function continuously until index equals to nth-item \"\"\" if","returning the function continuously until index equals to nth-item \"\"\"","are returning the function continuously until index equals to nth-item","> array[middle]: left = middle + 1 else: right =","Space Complexity: 0(log*n) recursion call stack space\"\"\" # TODO: implement","0(log*n) recursion call stack space\"\"\" # TODO: implement binary search","until item is found for index, value in enumerate(array): #O(n)","return index # found O(1) return None # not found","left, right = 0, len(array) - 1 middle = left","== 0: return None while left <= right: middle =","Complexity: 0(log*n) recursion call stack space\"\"\" # TODO: implement binary","binary_search(array, item): \"\"\"return the index of item in sorted array","#!python \"\"\" ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \"\"\"","over all array values until item is found for index,","are n amount of items\"\"\" # loop over all array","while left <= right: middle = left + (right -","return index if array[index] == item: return index else: return","O(1) def linear_search_recursive(array, item, index=0): \"\"\"Time complexity: O(n) because you","continuously until index equals to nth-item \"\"\" if len(array) <=","None # not found O(1) def linear_search_recursive(array, item, index=0): \"\"\"Time","# return binary_search_recursive(array, item) def binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n)","# TODO: implement binary search recursively here if left is","(right - left) // 2 if item == array[middle]: return","item == value: #O(1) return index # found O(1) return","index=0): \"\"\"Time complexity: O(n) because you are returning the function","COMPLEXITY!!!!! \"\"\" def linear_search(array, item): \"\"\"return the first index of","O(n) because you iterate through n amount of items in","array[middle]: return middle elif item > array[middle]: left = middle","first index of item in array or None if item","if item is not found\"\"\" return binary_search_iterative(array, item) # return","\"\"\"Time complexity: O(n) because you are returning the function continuously","else: return linear_search_recursive(array, item, index + 1) def binary_search(array, item):","are constantly dividing the length of array by 2 until","array[middle]: left = middle + 1 else: right = middle","1 return None def binary_search_recursive(array, item, left=None, right=None): \"\"\"Time Complexity:","space\"\"\" # TODO: implement binary search recursively here if left","found\"\"\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array,","right = 0, len(array) - 1 if len(array) == 0:","index equals to nth-item \"\"\" if len(array) <= index: return","left) // 2 if item == array[middle]: return middle elif","array[middle]: return binary_search_recursive(array, item, middle + 1, right) else: return","until index equals to nth-item \"\"\" if len(array) <= index:","\"\"\" ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \"\"\" def","function continuously until index equals to nth-item \"\"\" if len(array)","binary_search_recursive(array, item, middle + 1, right) else: return binary_search_recursive(array, item,","left + (right - left) // 2 if item ==","item) def linear_search_iterative(array, item): \"\"\"Time complexity: O(n) because you iterate","array[middle] == item: return middle elif item > array[middle]: return","#O(1) return index # found O(1) return None # not","recursively here if left is None and right is None:","item > array[middle]: return binary_search_recursive(array, item, middle + 1, right)","dividing the length of array by 2 until array length","constantly dividing the length of array by 2 until array","middle elif item > array[middle]: left = middle + 1","array[index] == item: return index else: return linear_search_recursive(array, item, index","\"\"\"Time Complexity: O(log*n) because you are constantly dividing the length","(right - left) // 2 if left > right: return","return index else: return linear_search_recursive(array, item, index + 1) def","len(array) - 1 if len(array) == 0: return None while","length of array by 2 until array length is 1","loop over all array values until item is found for","stack space\"\"\" # TODO: implement binary search recursively here if","1) def binary_search(array, item): \"\"\"return the index of item in","if array[middle] == item: return middle elif item > array[middle]:","value: #O(1) return index # found O(1) return None #","the index of item in sorted array or None if","Complexity: O(log*n) because you are constantly dividing the length of","item) def binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n) because you are","there are n amount of items\"\"\" # loop over all","left + (right - left) // 2 if left >","not found O(1) def linear_search_recursive(array, item, index=0): \"\"\"Time complexity: O(n)","index + 1) def binary_search(array, item): \"\"\"return the index of","return None if array[middle] == item: return middle elif item","left=None, right=None): \"\"\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call","else: right = middle - 1 return None def binary_search_recursive(array,","item, index + 1) def binary_search(array, item): \"\"\"return the index","or None if item is not found\"\"\" return linear_search_iterative(array, item)","found O(1) return None # not found O(1) def linear_search_recursive(array,","def binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n) because you are constantly","index if array[index] == item: return index else: return linear_search_recursive(array,","amount of items\"\"\" # loop over all array values until","here if left is None and right is None: left,","+ (right - left) // 2 if left > right:","you are returning the function continuously until index equals to","item): \"\"\"Time complexity: O(n) because you iterate through n amount","- left) // 2 if left > right: return None","item is not found\"\"\" return binary_search_iterative(array, item) # return binary_search_recursive(array,","right=None): \"\"\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack","middle + 1, right) else: return binary_search_recursive(array, item, left, middle","array length is 1 Space Complexity: O(1) \"\"\" left, right","in array Space Complexity: O(n) because there are n amount","iterate through n amount of items in array Space Complexity:","because you iterate through n amount of items in array","return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item):","return None # not found O(1) def linear_search_recursive(array, item, index=0):","None if item is not found\"\"\" return binary_search_iterative(array, item) #","== item: return middle elif item > array[middle]: return binary_search_recursive(array,","linear_search_recursive(array, item, index + 1) def binary_search(array, item): \"\"\"return the","of item in sorted array or None if item is","// 2 if left > right: return None if array[middle]","def binary_search(array, item): \"\"\"return the index of item in sorted","def linear_search_iterative(array, item): \"\"\"Time complexity: O(n) because you iterate through","1 if len(array) == 0: return None while left <=","the length of array by 2 until array length is","left <= right: middle = left + (right - left)","sorted array or None if item is not found\"\"\" return","index: return index if array[index] == item: return index else:","of items\"\"\" # loop over all array values until item","- 1 middle = left + (right - left) //","n amount of items\"\"\" # loop over all array values","1 middle = left + (right - left) // 2","O(1) \"\"\" left, right = 0, len(array) - 1 if","\"\"\"return the first index of item in array or None","of items in array Space Complexity: O(n) because there are","\"\"\"return the index of item in sorted array or None","SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array, item): \"\"\"return the first index","for index, value in enumerate(array): #O(n) if item == value:","O(n) because there are n amount of items\"\"\" # loop","binary_search_recursive(array, item) def binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n) because you","- 1 return None def binary_search_recursive(array, item, left=None, right=None): \"\"\"Time","+ 1, right) else: return binary_search_recursive(array, item, left, middle -","item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item): \"\"\"Time complexity:","in sorted array or None if item is not found\"\"\"","the first index of item in array or None if","found for index, value in enumerate(array): #O(n) if item ==","0: return None while left <= right: middle = left","return binary_search_recursive(array, item, middle + 1, right) else: return binary_search_recursive(array,","array by 2 until array length is 1 Space Complexity:","length is 1 Space Complexity: O(1) \"\"\" left, right =","ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array,","# found O(1) return None # not found O(1) def","middle + 1 else: right = middle - 1 return","index, value in enumerate(array): #O(n) if item == value: #O(1)","middle - 1 return None def binary_search_recursive(array, item, left=None, right=None):","item: return middle elif item > array[middle]: return binary_search_recursive(array, item,","right: middle = left + (right - left) // 2","\"\"\" def linear_search(array, item): \"\"\"return the first index of item","because you are returning the function continuously until index equals","item in sorted array or None if item is not","// 2 if item == array[middle]: return middle elif item","= middle - 1 return None def binary_search_recursive(array, item, left=None,","return None while left <= right: middle = left +","= left + (right - left) // 2 if item","items in array Space Complexity: O(n) because there are n","left, right = 0, len(array) - 1 if len(array) ==","left) // 2 if left > right: return None if","None if array[middle] == item: return middle elif item >","found\"\"\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array,","+ (right - left) // 2 if item == array[middle]:","\"\"\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack space\"\"\"","item == array[middle]: return middle elif item > array[middle]: left","in array or None if item is not found\"\"\" return","2 until array length is 1 Space Complexity: O(1) \"\"\"","right = middle - 1 return None def binary_search_recursive(array, item,","is None and right is None: left, right = 0,","def linear_search_recursive(array, item, index=0): \"\"\"Time complexity: O(n) because you are","index of item in sorted array or None if item","= 0, len(array) - 1 if len(array) == 0: return","amount of items in array Space Complexity: O(n) because there","return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item):","of item in array or None if item is not","right is None: left, right = 0, len(array) - 1","by 2 until array length is 1 Space Complexity: O(1)","of array by 2 until array length is 1 Space","if array[index] == item: return index else: return linear_search_recursive(array, item,","array or None if item is not found\"\"\" return binary_search_iterative(array,","right: return None if array[middle] == item: return middle elif","item is not found\"\"\" return linear_search_iterative(array, item) # return linear_search_recursive(array,","you are constantly dividing the length of array by 2","if item == value: #O(1) return index # found O(1)","binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n) because you are constantly dividing","index else: return linear_search_recursive(array, item, index + 1) def binary_search(array,","item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item): \"\"\"Time Complexity:","not found\"\"\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def","left > right: return None if array[middle] == item: return","or None if item is not found\"\"\" return binary_search_iterative(array, item)","== value: #O(1) return index # found O(1) return None","return binary_search_recursive(array, item) def binary_search_iterative(array, item): \"\"\"Time Complexity: O(log*n) because","return linear_search_recursive(array, item) def linear_search_iterative(array, item): \"\"\"Time complexity: O(n) because","binary search recursively here if left is None and right","item, left=None, right=None): \"\"\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion","None: left, right = 0, len(array) - 1 middle =","to nth-item \"\"\" if len(array) <= index: return index if","return None def binary_search_recursive(array, item, left=None, right=None): \"\"\"Time Complexity: O(log*n)","# loop over all array values until item is found","is found for index, value in enumerate(array): #O(n) if item","#O(n) if item == value: #O(1) return index # found","return linear_search_recursive(array, item, index + 1) def binary_search(array, item): \"\"\"return","1 Space Complexity: O(1) \"\"\" left, right = 0, len(array)","> array[middle]: return binary_search_recursive(array, item, middle + 1, right) else:","Complexity: O(n) because there are n amount of items\"\"\" #","nth-item \"\"\" if len(array) <= index: return index if array[index]","array Space Complexity: O(n) because there are n amount of","len(array) <= index: return index if array[index] == item: return","+ 1) def binary_search(array, item): \"\"\"return the index of item","len(array) - 1 middle = left + (right - left)","all array values until item is found for index, value","because there are n amount of items\"\"\" # loop over","None def binary_search_recursive(array, item, left=None, right=None): \"\"\"Time Complexity: O(log*n) Space","Space Complexity: O(1) \"\"\" left, right = 0, len(array) -","linear_search(array, item): \"\"\"return the first index of item in array","if item is not found\"\"\" return linear_search_iterative(array, item) # return","O(1) return None # not found O(1) def linear_search_recursive(array, item,","WITH TIME AND SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array, item): \"\"\"return","\"\"\" if len(array) <= index: return index if array[index] ==","= left + (right - left) // 2 if left","> right: return None if array[middle] == item: return middle","left is None and right is None: left, right =","right = 0, len(array) - 1 middle = left +","item, middle + 1, right) else: return binary_search_recursive(array, item, left,","1 else: right = middle - 1 return None def","O(log*n) Space Complexity: 0(log*n) recursion call stack space\"\"\" # TODO:","values until item is found for index, value in enumerate(array):","len(array) == 0: return None while left <= right: middle","elif item > array[middle]: left = middle + 1 else:","Space Complexity: O(n) because there are n amount of items\"\"\"","through n amount of items in array Space Complexity: O(n)","\"\"\" left, right = 0, len(array) - 1 if len(array)","None if item is not found\"\"\" return linear_search_iterative(array, item) #","enumerate(array): #O(n) if item == value: #O(1) return index #","item: return index else: return linear_search_recursive(array, item, index + 1)","if len(array) <= index: return index if array[index] == item:","in enumerate(array): #O(n) if item == value: #O(1) return index","is not found\"\"\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item)","array or None if item is not found\"\"\" return linear_search_iterative(array,","complexity: O(n) because you are returning the function continuously until","<= right: middle = left + (right - left) //","if left > right: return None if array[middle] == item:","is not found\"\"\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item)","AND SPACE COMPLEXITY!!!!! \"\"\" def linear_search(array, item): \"\"\"return the first","if len(array) == 0: return None while left <= right:","binary_search_recursive(array, item, left=None, right=None): \"\"\"Time Complexity: O(log*n) Space Complexity: 0(log*n)","array values until item is found for index, value in","until array length is 1 Space Complexity: O(1) \"\"\" left,","middle elif item > array[middle]: return binary_search_recursive(array, item, middle +","item): \"\"\"Time Complexity: O(log*n) because you are constantly dividing the","= 0, len(array) - 1 middle = left + (right","def linear_search(array, item): \"\"\"return the first index of item in","TODO: implement binary search recursively here if left is None","# return linear_search_recursive(array, item) def linear_search_iterative(array, item): \"\"\"Time complexity: O(n)","linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item): \"\"\"Time","found O(1) def linear_search_recursive(array, item, index=0): \"\"\"Time complexity: O(n) because","- 1 if len(array) == 0: return None while left","left = middle + 1 else: right = middle -","None and right is None: left, right = 0, len(array)","2 if left > right: return None if array[middle] ==","<= index: return index if array[index] == item: return index","index of item in array or None if item is","= middle + 1 else: right = middle - 1","complexity: O(n) because you iterate through n amount of items","None while left <= right: middle = left + (right","item): \"\"\"return the first index of item in array or","index # found O(1) return None # not found O(1)","n amount of items in array Space Complexity: O(n) because","linear_search_recursive(array, item, index=0): \"\"\"Time complexity: O(n) because you are returning","implement binary search recursively here if left is None and","2 if item == array[middle]: return middle elif item >","recursion call stack space\"\"\" # TODO: implement binary search recursively","and right is None: left, right = 0, len(array) -","0, len(array) - 1 middle = left + (right -","is 1 Space Complexity: O(1) \"\"\" left, right = 0,","you iterate through n amount of items in array Space","item, index=0): \"\"\"Time complexity: O(n) because you are returning the","\"\"\"Time complexity: O(n) because you iterate through n amount of","not found\"\"\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def","Complexity: O(1) \"\"\" left, right = 0, len(array) - 1","- left) // 2 if item == array[middle]: return middle","linear_search_iterative(array, item): \"\"\"Time complexity: O(n) because you iterate through n","item is found for index, value in enumerate(array): #O(n) if","if item == array[middle]: return middle elif item > array[middle]:","== array[middle]: return middle elif item > array[middle]: left =","== item: return index else: return linear_search_recursive(array, item, index +","elif item > array[middle]: return binary_search_recursive(array, item, middle + 1,","return middle elif item > array[middle]: return binary_search_recursive(array, item, middle","linear_search_recursive(array, item) def linear_search_iterative(array, item): \"\"\"Time complexity: O(n) because you","item > array[middle]: left = middle + 1 else: right","# not found O(1) def linear_search_recursive(array, item, index=0): \"\"\"Time complexity:","items\"\"\" # loop over all array values until item is","because you are constantly dividing the length of array by","O(log*n) because you are constantly dividing the length of array","call stack space\"\"\" # TODO: implement binary search recursively here","if left is None and right is None: left, right","item in array or None if item is not found\"\"\"","binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item): \"\"\"Time","value in enumerate(array): #O(n) if item == value: #O(1) return","is None: left, right = 0, len(array) - 1 middle"],"string":"[\n \"middle = left + (right - left) // 2 if\",\n \"equals to nth-item \\\"\\\"\\\" if len(array) <= index: return index\",\n \"search recursively here if left is None and right is\",\n \"0, len(array) - 1 if len(array) == 0: return None\",\n \"O(n) because you are returning the function continuously until index\",\n \"return middle elif item > array[middle]: left = middle +\",\n \"1, right) else: return binary_search_recursive(array, item, left, middle - 1)\",\n \"item): \\\"\\\"\\\"return the index of item in sorted array or\",\n \"TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return the\",\n \"+ 1 else: right = middle - 1 return None\",\n \"def binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity:\",\n \"Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack space\\\"\\\"\\\" #\",\n \"FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item):\",\n \"the function continuously until index equals to nth-item \\\"\\\"\\\" if\",\n \"returning the function continuously until index equals to nth-item \\\"\\\"\\\"\",\n \"are returning the function continuously until index equals to nth-item\",\n \"> array[middle]: left = middle + 1 else: right =\",\n \"Space Complexity: 0(log*n) recursion call stack space\\\"\\\"\\\" # TODO: implement\",\n \"0(log*n) recursion call stack space\\\"\\\"\\\" # TODO: implement binary search\",\n \"until item is found for index, value in enumerate(array): #O(n)\",\n \"return index # found O(1) return None # not found\",\n \"left, right = 0, len(array) - 1 middle = left\",\n \"== 0: return None while left <= right: middle =\",\n \"Complexity: 0(log*n) recursion call stack space\\\"\\\"\\\" # TODO: implement binary\",\n \"binary_search(array, item): \\\"\\\"\\\"return the index of item in sorted array\",\n \"#!python \\\"\\\"\\\" ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\"\",\n \"over all array values until item is found for index,\",\n \"are n amount of items\\\"\\\"\\\" # loop over all array\",\n \"while left <= right: middle = left + (right -\",\n \"return index if array[index] == item: return index else: return\",\n \"O(1) def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity: O(n) because you\",\n \"continuously until index equals to nth-item \\\"\\\"\\\" if len(array) <=\",\n \"None # not found O(1) def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time\",\n \"# return binary_search_recursive(array, item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n)\",\n \"# TODO: implement binary search recursively here if left is\",\n \"(right - left) // 2 if item == array[middle]: return\",\n \"item == value: #O(1) return index # found O(1) return\",\n \"index=0): \\\"\\\"\\\"Time complexity: O(n) because you are returning the function\",\n \"COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return the first index of\",\n \"O(n) because you iterate through n amount of items in\",\n \"array[middle]: return middle elif item > array[middle]: left = middle\",\n \"first index of item in array or None if item\",\n \"if item is not found\\\"\\\"\\\" return binary_search_iterative(array, item) # return\",\n \"\\\"\\\"\\\"Time complexity: O(n) because you are returning the function continuously\",\n \"else: return linear_search_recursive(array, item, index + 1) def binary_search(array, item):\",\n \"are constantly dividing the length of array by 2 until\",\n \"array[middle]: left = middle + 1 else: right = middle\",\n \"1 return None def binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time Complexity:\",\n \"space\\\"\\\"\\\" # TODO: implement binary search recursively here if left\",\n \"found\\\"\\\"\\\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array,\",\n \"right = 0, len(array) - 1 if len(array) == 0:\",\n \"index equals to nth-item \\\"\\\"\\\" if len(array) <= index: return\",\n \"left) // 2 if item == array[middle]: return middle elif\",\n \"array[middle]: return binary_search_recursive(array, item, middle + 1, right) else: return\",\n \"until index equals to nth-item \\\"\\\"\\\" if len(array) <= index:\",\n \"\\\"\\\"\\\" ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def\",\n \"function continuously until index equals to nth-item \\\"\\\"\\\" if len(array)\",\n \"binary_search_recursive(array, item, middle + 1, right) else: return binary_search_recursive(array, item,\",\n \"left + (right - left) // 2 if item ==\",\n \"item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n) because you iterate\",\n \"array[middle] == item: return middle elif item > array[middle]: return\",\n \"#O(1) return index # found O(1) return None # not\",\n \"recursively here if left is None and right is None:\",\n \"item > array[middle]: return binary_search_recursive(array, item, middle + 1, right)\",\n \"dividing the length of array by 2 until array length\",\n \"constantly dividing the length of array by 2 until array\",\n \"middle elif item > array[middle]: left = middle + 1\",\n \"array[index] == item: return index else: return linear_search_recursive(array, item, index\",\n \"\\\"\\\"\\\"Time Complexity: O(log*n) because you are constantly dividing the length\",\n \"(right - left) // 2 if left > right: return\",\n \"return index else: return linear_search_recursive(array, item, index + 1) def\",\n \"len(array) - 1 if len(array) == 0: return None while\",\n \"length of array by 2 until array length is 1\",\n \"loop over all array values until item is found for\",\n \"stack space\\\"\\\"\\\" # TODO: implement binary search recursively here if\",\n \"1) def binary_search(array, item): \\\"\\\"\\\"return the index of item in\",\n \"if array[middle] == item: return middle elif item > array[middle]:\",\n \"value: #O(1) return index # found O(1) return None #\",\n \"the index of item in sorted array or None if\",\n \"Complexity: O(log*n) because you are constantly dividing the length of\",\n \"item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n) because you are\",\n \"there are n amount of items\\\"\\\"\\\" # loop over all\",\n \"left + (right - left) // 2 if left >\",\n \"not found O(1) def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity: O(n)\",\n \"index + 1) def binary_search(array, item): \\\"\\\"\\\"return the index of\",\n \"return None if array[middle] == item: return middle elif item\",\n \"left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call\",\n \"else: right = middle - 1 return None def binary_search_recursive(array,\",\n \"item, index + 1) def binary_search(array, item): \\\"\\\"\\\"return the index\",\n \"or None if item is not found\\\"\\\"\\\" return linear_search_iterative(array, item)\",\n \"found O(1) return None # not found O(1) def linear_search_recursive(array,\",\n \"def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n) because you are constantly\",\n \"index if array[index] == item: return index else: return linear_search_recursive(array,\",\n \"amount of items\\\"\\\"\\\" # loop over all array values until\",\n \"here if left is None and right is None: left,\",\n \"+ (right - left) // 2 if left > right:\",\n \"you are returning the function continuously until index equals to\",\n \"item): \\\"\\\"\\\"Time complexity: O(n) because you iterate through n amount\",\n \"- left) // 2 if left > right: return None\",\n \"item is not found\\\"\\\"\\\" return binary_search_iterative(array, item) # return binary_search_recursive(array,\",\n \"right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack\",\n \"middle + 1, right) else: return binary_search_recursive(array, item, left, middle\",\n \"array length is 1 Space Complexity: O(1) \\\"\\\"\\\" left, right\",\n \"in array Space Complexity: O(n) because there are n amount\",\n \"iterate through n amount of items in array Space Complexity:\",\n \"because you iterate through n amount of items in array\",\n \"return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item):\",\n \"return None # not found O(1) def linear_search_recursive(array, item, index=0):\",\n \"None if item is not found\\\"\\\"\\\" return binary_search_iterative(array, item) #\",\n \"== item: return middle elif item > array[middle]: return binary_search_recursive(array,\",\n \"linear_search_recursive(array, item, index + 1) def binary_search(array, item): \\\"\\\"\\\"return the\",\n \"of item in sorted array or None if item is\",\n \"// 2 if left > right: return None if array[middle]\",\n \"def binary_search(array, item): \\\"\\\"\\\"return the index of item in sorted\",\n \"def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n) because you iterate through\",\n \"1 if len(array) == 0: return None while left <=\",\n \"the length of array by 2 until array length is\",\n \"left <= right: middle = left + (right - left)\",\n \"sorted array or None if item is not found\\\"\\\"\\\" return\",\n \"index: return index if array[index] == item: return index else:\",\n \"of items\\\"\\\"\\\" # loop over all array values until item\",\n \"- 1 middle = left + (right - left) //\",\n \"n amount of items\\\"\\\"\\\" # loop over all array values\",\n \"1 middle = left + (right - left) // 2\",\n \"O(1) \\\"\\\"\\\" left, right = 0, len(array) - 1 if\",\n \"\\\"\\\"\\\"return the first index of item in array or None\",\n \"of items in array Space Complexity: O(n) because there are\",\n \"\\\"\\\"\\\"return the index of item in sorted array or None\",\n \"SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return the first index\",\n \"for index, value in enumerate(array): #O(n) if item == value:\",\n \"O(n) because there are n amount of items\\\"\\\"\\\" # loop\",\n \"binary_search_recursive(array, item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n) because you\",\n \"- 1 return None def binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time\",\n \"+ 1, right) else: return binary_search_recursive(array, item, left, middle -\",\n \"item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity:\",\n \"in sorted array or None if item is not found\\\"\\\"\\\"\",\n \"the first index of item in array or None if\",\n \"found for index, value in enumerate(array): #O(n) if item ==\",\n \"0: return None while left <= right: middle = left\",\n \"return binary_search_recursive(array, item, middle + 1, right) else: return binary_search_recursive(array,\",\n \"array by 2 until array length is 1 Space Complexity:\",\n \"length is 1 Space Complexity: O(1) \\\"\\\"\\\" left, right =\",\n \"ANNOTATE FUNCTIONS WITH TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array,\",\n \"# found O(1) return None # not found O(1) def\",\n \"middle + 1 else: right = middle - 1 return\",\n \"index, value in enumerate(array): #O(n) if item == value: #O(1)\",\n \"middle - 1 return None def binary_search_recursive(array, item, left=None, right=None):\",\n \"item: return middle elif item > array[middle]: return binary_search_recursive(array, item,\",\n \"right: middle = left + (right - left) // 2\",\n \"\\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return the first index of item\",\n \"because you are returning the function continuously until index equals\",\n \"item in sorted array or None if item is not\",\n \"// 2 if item == array[middle]: return middle elif item\",\n \"= middle - 1 return None def binary_search_recursive(array, item, left=None,\",\n \"return None while left <= right: middle = left +\",\n \"= left + (right - left) // 2 if item\",\n \"items in array Space Complexity: O(n) because there are n\",\n \"left, right = 0, len(array) - 1 if len(array) ==\",\n \"left) // 2 if left > right: return None if\",\n \"None if array[middle] == item: return middle elif item >\",\n \"found\\\"\\\"\\\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array,\",\n \"+ (right - left) // 2 if item == array[middle]:\",\n \"\\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion call stack space\\\"\\\"\\\"\",\n \"item == array[middle]: return middle elif item > array[middle]: left\",\n \"in array or None if item is not found\\\"\\\"\\\" return\",\n \"2 until array length is 1 Space Complexity: O(1) \\\"\\\"\\\"\",\n \"right = middle - 1 return None def binary_search_recursive(array, item,\",\n \"is None and right is None: left, right = 0,\",\n \"def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity: O(n) because you are\",\n \"index of item in sorted array or None if item\",\n \"= 0, len(array) - 1 if len(array) == 0: return\",\n \"amount of items in array Space Complexity: O(n) because there\",\n \"return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item):\",\n \"of item in array or None if item is not\",\n \"right is None: left, right = 0, len(array) - 1\",\n \"by 2 until array length is 1 Space Complexity: O(1)\",\n \"of array by 2 until array length is 1 Space\",\n \"if array[index] == item: return index else: return linear_search_recursive(array, item,\",\n \"array or None if item is not found\\\"\\\"\\\" return binary_search_iterative(array,\",\n \"right: return None if array[middle] == item: return middle elif\",\n \"item is not found\\\"\\\"\\\" return linear_search_iterative(array, item) # return linear_search_recursive(array,\",\n \"you are constantly dividing the length of array by 2\",\n \"if item == value: #O(1) return index # found O(1)\",\n \"binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n) because you are constantly dividing\",\n \"index else: return linear_search_recursive(array, item, index + 1) def binary_search(array,\",\n \"item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity:\",\n \"not found\\\"\\\"\\\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item) def\",\n \"left > right: return None if array[middle] == item: return\",\n \"or None if item is not found\\\"\\\"\\\" return binary_search_iterative(array, item)\",\n \"== value: #O(1) return index # found O(1) return None\",\n \"return binary_search_recursive(array, item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time Complexity: O(log*n) because\",\n \"return linear_search_recursive(array, item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n) because\",\n \"binary search recursively here if left is None and right\",\n \"item, left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity: 0(log*n) recursion\",\n \"None: left, right = 0, len(array) - 1 middle =\",\n \"to nth-item \\\"\\\"\\\" if len(array) <= index: return index if\",\n \"return None def binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n)\",\n \"# loop over all array values until item is found\",\n \"is found for index, value in enumerate(array): #O(n) if item\",\n \"#O(n) if item == value: #O(1) return index # found\",\n \"return linear_search_recursive(array, item, index + 1) def binary_search(array, item): \\\"\\\"\\\"return\",\n \"1 Space Complexity: O(1) \\\"\\\"\\\" left, right = 0, len(array)\",\n \"> array[middle]: return binary_search_recursive(array, item, middle + 1, right) else:\",\n \"Complexity: O(n) because there are n amount of items\\\"\\\"\\\" #\",\n \"nth-item \\\"\\\"\\\" if len(array) <= index: return index if array[index]\",\n \"array Space Complexity: O(n) because there are n amount of\",\n \"len(array) <= index: return index if array[index] == item: return\",\n \"+ 1) def binary_search(array, item): \\\"\\\"\\\"return the index of item\",\n \"len(array) - 1 middle = left + (right - left)\",\n \"all array values until item is found for index, value\",\n \"because there are n amount of items\\\"\\\"\\\" # loop over\",\n \"None def binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space\",\n \"Space Complexity: O(1) \\\"\\\"\\\" left, right = 0, len(array) -\",\n \"linear_search(array, item): \\\"\\\"\\\"return the first index of item in array\",\n \"if item is not found\\\"\\\"\\\" return linear_search_iterative(array, item) # return\",\n \"O(1) return None # not found O(1) def linear_search_recursive(array, item,\",\n \"WITH TIME AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return\",\n \"\\\"\\\"\\\" if len(array) <= index: return index if array[index] ==\",\n \"= left + (right - left) // 2 if left\",\n \"> right: return None if array[middle] == item: return middle\",\n \"left is None and right is None: left, right =\",\n \"right = 0, len(array) - 1 middle = left +\",\n \"item, middle + 1, right) else: return binary_search_recursive(array, item, left,\",\n \"1 else: right = middle - 1 return None def\",\n \"O(log*n) Space Complexity: 0(log*n) recursion call stack space\\\"\\\"\\\" # TODO:\",\n \"values until item is found for index, value in enumerate(array):\",\n \"len(array) == 0: return None while left <= right: middle\",\n \"elif item > array[middle]: left = middle + 1 else:\",\n \"Space Complexity: O(n) because there are n amount of items\\\"\\\"\\\"\",\n \"through n amount of items in array Space Complexity: O(n)\",\n \"\\\"\\\"\\\" left, right = 0, len(array) - 1 if len(array)\",\n \"None if item is not found\\\"\\\"\\\" return linear_search_iterative(array, item) #\",\n \"enumerate(array): #O(n) if item == value: #O(1) return index #\",\n \"item: return index else: return linear_search_recursive(array, item, index + 1)\",\n \"if len(array) <= index: return index if array[index] == item:\",\n \"in enumerate(array): #O(n) if item == value: #O(1) return index\",\n \"is not found\\\"\\\"\\\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item)\",\n \"array or None if item is not found\\\"\\\"\\\" return linear_search_iterative(array,\",\n \"complexity: O(n) because you are returning the function continuously until\",\n \"<= right: middle = left + (right - left) //\",\n \"if left > right: return None if array[middle] == item:\",\n \"is not found\\\"\\\"\\\" return linear_search_iterative(array, item) # return linear_search_recursive(array, item)\",\n \"AND SPACE COMPLEXITY!!!!! \\\"\\\"\\\" def linear_search(array, item): \\\"\\\"\\\"return the first\",\n \"if len(array) == 0: return None while left <= right:\",\n \"binary_search_recursive(array, item, left=None, right=None): \\\"\\\"\\\"Time Complexity: O(log*n) Space Complexity: 0(log*n)\",\n \"array values until item is found for index, value in\",\n \"until array length is 1 Space Complexity: O(1) \\\"\\\"\\\" left,\",\n \"middle elif item > array[middle]: return binary_search_recursive(array, item, middle +\",\n \"item): \\\"\\\"\\\"Time Complexity: O(log*n) because you are constantly dividing the\",\n \"= 0, len(array) - 1 middle = left + (right\",\n \"def linear_search(array, item): \\\"\\\"\\\"return the first index of item in\",\n \"TODO: implement binary search recursively here if left is None\",\n \"# return linear_search_recursive(array, item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n)\",\n \"linear_search_iterative(array, item) # return linear_search_recursive(array, item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time\",\n \"found O(1) def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity: O(n) because\",\n \"- 1 if len(array) == 0: return None while left\",\n \"left = middle + 1 else: right = middle -\",\n \"None and right is None: left, right = 0, len(array)\",\n \"2 if left > right: return None if array[middle] ==\",\n \"<= index: return index if array[index] == item: return index\",\n \"index of item in array or None if item is\",\n \"= middle + 1 else: right = middle - 1\",\n \"complexity: O(n) because you iterate through n amount of items\",\n \"None while left <= right: middle = left + (right\",\n \"item): \\\"\\\"\\\"return the first index of item in array or\",\n \"index # found O(1) return None # not found O(1)\",\n \"n amount of items in array Space Complexity: O(n) because\",\n \"linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity: O(n) because you are returning\",\n \"implement binary search recursively here if left is None and\",\n \"2 if item == array[middle]: return middle elif item >\",\n \"recursion call stack space\\\"\\\"\\\" # TODO: implement binary search recursively\",\n \"and right is None: left, right = 0, len(array) -\",\n \"0, len(array) - 1 middle = left + (right -\",\n \"is 1 Space Complexity: O(1) \\\"\\\"\\\" left, right = 0,\",\n \"you iterate through n amount of items in array Space\",\n \"item, index=0): \\\"\\\"\\\"Time complexity: O(n) because you are returning the\",\n \"\\\"\\\"\\\"Time complexity: O(n) because you iterate through n amount of\",\n \"not found\\\"\\\"\\\" return binary_search_iterative(array, item) # return binary_search_recursive(array, item) def\",\n \"Complexity: O(1) \\\"\\\"\\\" left, right = 0, len(array) - 1\",\n \"- left) // 2 if item == array[middle]: return middle\",\n \"linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n) because you iterate through n\",\n \"item is found for index, value in enumerate(array): #O(n) if\",\n \"if item == array[middle]: return middle elif item > array[middle]:\",\n \"== array[middle]: return middle elif item > array[middle]: left =\",\n \"== item: return index else: return linear_search_recursive(array, item, index +\",\n \"elif item > array[middle]: return binary_search_recursive(array, item, middle + 1,\",\n \"return middle elif item > array[middle]: return binary_search_recursive(array, item, middle\",\n \"linear_search_recursive(array, item) def linear_search_iterative(array, item): \\\"\\\"\\\"Time complexity: O(n) because you\",\n \"item > array[middle]: left = middle + 1 else: right\",\n \"# not found O(1) def linear_search_recursive(array, item, index=0): \\\"\\\"\\\"Time complexity:\",\n \"items\\\"\\\"\\\" # loop over all array values until item is\",\n \"because you are constantly dividing the length of array by\",\n \"O(log*n) because you are constantly dividing the length of array\",\n \"call stack space\\\"\\\"\\\" # TODO: implement binary search recursively here\",\n \"if left is None and right is None: left, right\",\n \"item in array or None if item is not found\\\"\\\"\\\"\",\n \"binary_search_iterative(array, item) # return binary_search_recursive(array, item) def binary_search_iterative(array, item): \\\"\\\"\\\"Time\",\n \"value in enumerate(array): #O(n) if item == value: #O(1) return\",\n \"is None: left, right = 0, len(array) - 1 middle\"\n]","isConversational":false}}},{"rowIdx":687,"cells":{"ngram":{"kind":"list like","value":["b, c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory","into memory (concept, location, person,DATE) values (?, ?, ?, ?)\",","= ? WHERE concept = ? ''', (b, a)) RM.commit()","from pattern.en import pprint from pattern.en import parsetree from pattern.en","(b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM","what/how is 'concept' self.reference = 'none' # unused self.locality =","3rd-person, singular self.isProperNoun = False # True if proper noun:","# what/how is 'concept' self.reference = 'none' # unused self.locality","a, b): c = '3sg' dt = datetime.now() RM.execute(\"insert into","single concept class conceptClass: def __init__(self, state='none', locality='none'): self.state =","holds raw input and memories in parse taged columns '''","location = ? WHERE concept = ? ''', (b, a))","memory def known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE","conceptClass: def __init__(self, state='none', locality='none'): self.state = state # what/how","datetime from pattern.en import parse from pattern.en import pprint from","= conceptClass() if c in self.person: self.concepts[c].person = self.person[c] else:","def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item =?''',","memory WHERE concept =?''', (c,)) user = cursor.fetchone() # if","# Information about a single concept class conceptClass: def __init__(self,","c): cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''', (c,))","== 'None': return user def add_memory(self, a, b): c =","add oncept to raw_input table in robbie_memory # x= #","(?, ?)\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if c","= RM.cursor() # Information about a single concept class conceptClass:","cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value FROM profile","sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor()","= datetime.now() # RM.execute(\"insert into RAW_INPUT (RAW, DATE) values (?,","'3sg' dt = datetime.now() RM.execute(\"insert into memory (concept, location, person,DATE)","self.locality = locality # where is 'concept' self.person = '3sg'","memory class stored in sqlite data base holds raw input","# where is 'concept' self.person = '3sg' # e.g. a","= sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information about","import sys import re import sqlite3 import os from datetime","{'1sg': 'my', '2sg': 'your', '3sg': 'its' } # Add a","import pprint from pattern.en import parsetree from pattern.en import wordnet","state='none', locality='none'): self.state = state # what/how is 'concept' self.reference","Robbie memory class. Collection of concepts class memoryClass(): def __init__(self):","?)\", (a, b, c, dt)) RM.commit() def update_memory(self, a, b):","from datetime import date, datetime from pattern.en import parse from","RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor","concept,location, person FROM memory WHERE concept =?''', (a,)) user =","RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE","''' memory class stored in sqlite data base holds raw","'2sg': 'your', '3sg': 'its' } # Add a concept to","SET location = ? WHERE concept = ? ''', (b,","from pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir","taged columns ''' import sys import re import sqlite3 import","= state # what/how is 'concept' self.reference = 'none' #","FROM memory WHERE concept =?''', (c,)) user = cursor.fetchone() #","date, datetime from pattern.en import parse from pattern.en import pprint","FROM profile WHERE item =?''', (a,)) user = cursor.fetchone() return","parsetree from pattern.en import wordnet from pattern.en import pluralize, singularize","'3sg' # e.g. a thing is 3rd-person, singular self.isProperNoun =","class conceptClass: def __init__(self, state='none', locality='none'): self.state = state #","'my', '2sg': 'your', '3sg': 'its' } # Add a concept","self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg':","= os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM","pattern.en import parsetree from pattern.en import wordnet from pattern.en import","c = '3sg' dt = datetime.now() RM.execute(\"insert into memory (concept,","(a,)) user = cursor.fetchone() return user def Dump(self): return (self.concepts.state)","lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir","} self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its' }","thing is 3rd-person, singular self.isProperNoun = False # True if","import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/'","def __init__(self, state='none', locality='none'): self.state = state # what/how is","is 'concept' self.person = '3sg' # e.g. a thing is","re import sqlite3 import os from datetime import date, datetime","self.isProperNoun = False # True if proper noun: e.g. Robert","__init__(self, state='none', locality='none'): self.state = state # what/how is 'concept'","Add a concept to memory def add(self, c): # add","raw_input table in robbie_memory # x= # dt = datetime.now()","# unused self.locality = locality # where is 'concept' self.person","= ? ''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT","data base holds raw input and memories in parse taged","import os from datetime import date, datetime from pattern.en import","location FROM memory WHERE concept =?''', (c,)) user = cursor.fetchone()","dt)) # RM.commit() self.concepts[c] = conceptClass() if c in self.person:","table in robbie_memory # x= # dt = datetime.now() #","concepts class memoryClass(): def __init__(self): self.concepts = {} self.person =","unused self.locality = locality # where is 'concept' self.person =","from pattern.en import pluralize, singularize from pattern.en import conjugate, lemma,","'concept' self.reference = 'none' # unused self.locality = locality #","'none' # unused self.locality = locality # where is 'concept'","39, 'color' = 'blue' # Robbie memory class. Collection of","values (?, ?, ?, ?)\", (a, b, c, dt)) RM.commit()","= '3sg' # e.g. a thing is 3rd-person, singular self.isProperNoun","# RM.commit() self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person","'3sg' # Return True if concept 'c' (string) is in","= 'none' # unused self.locality = locality # where is","# True if proper noun: e.g. Robert self.properties = {}","#dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite')","into RAW_INPUT (RAW, DATE) values (?, ?)\",(c, dt)) # RM.commit()","is 'concept' self.reference = 'none' # unused self.locality = locality","? WHERE concept = ? ''', (b, a)) RM.commit() def","''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person","#!/usr/bin/python ''' memory class stored in sqlite data base holds","{} # Dict of custom properties, e.g. 'age' = 39,","'color' = 'blue' # Robbie memory class. Collection of concepts","Robert self.properties = {} # Dict of custom properties, e.g.","= locality # where is 'concept' self.person = '3sg' #","is 3rd-person, singular self.isProperNoun = False # True if proper","memory def add(self, c): # add oncept to raw_input table","self.state = state # what/how is 'concept' self.reference = 'none'","if user == 'None': return user def add_memory(self, a, b):","= False # True if proper noun: e.g. Robert self.properties","person FROM memory WHERE concept =?''', (a,)) user = cursor.fetchone()","class. Collection of concepts class memoryClass(): def __init__(self): self.concepts =","memories in parse taged columns ''' import sys import re","(string) is in memory def known(self, c): cursor.execute('''SELECT concept, location","RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location =","'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg':","c): # add oncept to raw_input table in robbie_memory #","in robbie_memory # x= # dt = datetime.now() # RM.execute(\"insert","memory WHERE concept =?''', (a,)) user = cursor.fetchone() return user","add_memory(self, a, b): c = '3sg' dt = datetime.now() RM.execute(\"insert","sqlite data base holds raw input and memories in parse","from pattern.en import parsetree from pattern.en import wordnet from pattern.en","sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information about a","memory class. Collection of concepts class memoryClass(): def __init__(self): self.concepts","wordnet from pattern.en import pluralize, singularize from pattern.en import conjugate,","pprint from pattern.en import parsetree from pattern.en import wordnet from","lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM =","user def add_memory(self, a, b): c = '3sg' dt =","parse taged columns ''' import sys import re import sqlite3","# add oncept to raw_input table in robbie_memory # x=","self.properties = {} # Dict of custom properties, e.g. 'age'","(a,)) user = cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT","user = cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value","Collection of concepts class memoryClass(): def __init__(self): self.concepts = {}","+ '/data/robbie_memory.db') cursor = RM.cursor() # Information about a single","# e.g. a thing is 3rd-person, singular self.isProperNoun = False","dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location","value FROM profile WHERE item =?''', (a,)) user = cursor.fetchone()","# Add a concept to memory def add(self, c): #","'3sg': 'its' } # Add a concept to memory def","=?''', (a,)) user = cursor.fetchone() return user def Dump(self): return","RAW_INPUT (RAW, DATE) values (?, ?)\",(c, dt)) # RM.commit() self.concepts[c]","user == 'None': return user def add_memory(self, a, b): c","# Return True if concept 'c' (string) is in memory","update_memory(self, a, b): cursor.execute('''UPDATE memory SET location = ? WHERE","?, ?)\", (a, b, c, dt)) RM.commit() def update_memory(self, a,","a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''', (a,))","in memory def known(self, c): cursor.execute('''SELECT concept, location FROM memory","b): cursor.execute('''UPDATE memory SET location = ? WHERE concept =","WHERE concept =?''', (a,)) user = cursor.fetchone() return user def","(concept, location, person,DATE) values (?, ?, ?, ?)\", (a, b,","__init__(self): self.concepts = {} self.person = {'I': '1sg', 'you': '2sg'","'1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your',","sqlite3 import os from datetime import date, datetime from pattern.en","= {} self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns","FROM memory WHERE concept =?''', (a,)) user = cursor.fetchone() return","RM.commit() self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person =","memoryClass(): def __init__(self): self.concepts = {} self.person = {'I': '1sg',","True if proper noun: e.g. Robert self.properties = {} #","'blue' # Robbie memory class. Collection of concepts class memoryClass():","state # what/how is 'concept' self.reference = 'none' # unused","values (?, ?)\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if","concept class conceptClass: def __init__(self, state='none', locality='none'): self.state = state","e.g. 'age' = 39, 'color' = 'blue' # Robbie memory","pluralize, singularize from pattern.en import conjugate, lemma, lexeme #dir =","''' import sys import re import sqlite3 import os from","Dict of custom properties, e.g. 'age' = 39, 'color' =","dt = datetime.now() RM.execute(\"insert into memory (concept, location, person,DATE) values","datetime.now() # RM.execute(\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\",(c,","self.reference = 'none' # unused self.locality = locality # where","item =?''', (a,)) user = cursor.fetchone() return user def Dump(self):","datetime.now() RM.execute(\"insert into memory (concept, location, person,DATE) values (?, ?,","= cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value FROM","where is 'concept' self.person = '3sg' # e.g. a thing","False # True if proper noun: e.g. Robert self.properties =","profile WHERE item =?''', (a,)) user = cursor.fetchone() return user","#RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information","= '3sg' # Return True if concept 'c' (string) is","WHERE concept = ? ''', (b, a)) RM.commit() def search_memory(self,","dt = datetime.now() # RM.execute(\"insert into RAW_INPUT (RAW, DATE) values","# Robbie memory class. Collection of concepts class memoryClass(): def","self.concepts[c].person = '3sg' # Return True if concept 'c' (string)","self.person = '3sg' # e.g. a thing is 3rd-person, singular","(RAW, DATE) values (?, ?)\",(c, dt)) # RM.commit() self.concepts[c] =","search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,))","'its' } # Add a concept to memory def add(self,","# if user == 'None': return user def add_memory(self, a,","cursor.execute('''UPDATE memory SET location = ? WHERE concept = ?","RM.execute(\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\",(c, dt)) #","True if concept 'c' (string) is in memory def known(self,","# dt = datetime.now() # RM.execute(\"insert into RAW_INPUT (RAW, DATE)","pattern.en import pprint from pattern.en import parsetree from pattern.en import","if c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person =","{} self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns =","stored in sqlite data base holds raw input and memories","import re import sqlite3 import os from datetime import date,","= '3sg' dt = datetime.now() RM.execute(\"insert into memory (concept, location,","e.g. Robert self.properties = {} # Dict of custom properties,","= datetime.now() RM.execute(\"insert into memory (concept, location, person,DATE) values (?,","memory SET location = ? WHERE concept = ? ''',","= self.person[c] else: self.concepts[c].person = '3sg' # Return True if","memory (concept, location, person,DATE) values (?, ?, ?, ?)\", (a,","if concept 'c' (string) is in memory def known(self, c):","e.g. a thing is 3rd-person, singular self.isProperNoun = False #","if proper noun: e.g. Robert self.properties = {} # Dict","?)\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if c in","cursor = RM.cursor() # Information about a single concept class","# Dict of custom properties, e.g. 'age' = 39, 'color'","= 'blue' # Robbie memory class. Collection of concepts class","from pattern.en import wordnet from pattern.en import pluralize, singularize from","def known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE concept","(?, ?, ?, ?)\", (a, b, c, dt)) RM.commit() def","return user def add_memory(self, a, b): c = '3sg' dt","else: self.concepts[c].person = '3sg' # Return True if concept 'c'","self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' # Return","c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg'","os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM =","'/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db')","= {} # Dict of custom properties, e.g. 'age' =","c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET","Return True if concept 'c' (string) is in memory def","pattern.en import parse from pattern.en import pprint from pattern.en import","is in memory def known(self, c): cursor.execute('''SELECT concept, location FROM","WHERE concept =?''', (c,)) user = cursor.fetchone() # if user","about a single concept class conceptClass: def __init__(self, state='none', locality='none'):","= '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir +","?, ?, ?)\", (a, b, c, dt)) RM.commit() def update_memory(self,","in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' #","pattern.en import pluralize, singularize from pattern.en import conjugate, lemma, lexeme","oncept to raw_input table in robbie_memory # x= # dt","= 39, 'color' = 'blue' # Robbie memory class. Collection","{'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg':","def add_memory(self, a, b): c = '3sg' dt = datetime.now()","= {'1sg': 'my', '2sg': 'your', '3sg': 'its' } # Add","concept =?''', (c,)) user = cursor.fetchone() # if user ==","Information about a single concept class conceptClass: def __init__(self, state='none',","WHERE item =?''', (a,)) user = cursor.fetchone() return user def","import wordnet from pattern.en import pluralize, singularize from pattern.en import","search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''',","and memories in parse taged columns ''' import sys import","'concept' self.person = '3sg' # e.g. a thing is 3rd-person,","(a, b, c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE","import parsetree from pattern.en import wordnet from pattern.en import pluralize,","=?''', (c,)) user = cursor.fetchone() # if user == 'None':","a single concept class conceptClass: def __init__(self, state='none', locality='none'): self.state","# RM.execute(\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\",(c, dt))","to raw_input table in robbie_memory # x= # dt =","cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''', (a,)) user","custom properties, e.g. 'age' = 39, 'color' = 'blue' #","+'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() #","self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person = self.person[c]","concept 'c' (string) is in memory def known(self, c): cursor.execute('''SELECT","locality='none'): self.state = state # what/how is 'concept' self.reference =","'/data/robbie_memory.db') cursor = RM.cursor() # Information about a single concept","os from datetime import date, datetime from pattern.en import parse","class stored in sqlite data base holds raw input and","cursor.fetchone() # if user == 'None': return user def add_memory(self,","robbie_memory # x= # dt = datetime.now() # RM.execute(\"insert into","=?''', (a,)) user = cursor.fetchone() return user def search_profile(self, a):","? ''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location,","properties, e.g. 'age' = 39, 'color' = 'blue' # Robbie","cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,)) user =","input and memories in parse taged columns ''' import sys","def __init__(self): self.concepts = {} self.person = {'I': '1sg', 'you':","pattern.en import wordnet from pattern.en import pluralize, singularize from pattern.en","# x= # dt = datetime.now() # RM.execute(\"insert into RAW_INPUT","a concept to memory def add(self, c): # add oncept","user = cursor.fetchone() # if user == 'None': return user","self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its' } #","= sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor =","in parse taged columns ''' import sys import re import","a): cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,)) user","RM.cursor() # Information about a single concept class conceptClass: def","locality # where is 'concept' self.person = '3sg' # e.g.","concept =?''', (a,)) user = cursor.fetchone() return user def search_profile(self,","known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''',","a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory","def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept","RM.execute(\"insert into memory (concept, location, person,DATE) values (?, ?, ?,","noun: e.g. Robert self.properties = {} # Dict of custom","singular self.isProperNoun = False # True if proper noun: e.g.","cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''', (c,)) user","from pattern.en import parse from pattern.en import pprint from pattern.en","concept to memory def add(self, c): # add oncept to","pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir =","class memoryClass(): def __init__(self): self.concepts = {} self.person = {'I':","concept, location FROM memory WHERE concept =?''', (c,)) user =","x= # dt = datetime.now() # RM.execute(\"insert into RAW_INPUT (RAW,","a, b): cursor.execute('''UPDATE memory SET location = ? WHERE concept","= {'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my',","add(self, c): # add oncept to raw_input table in robbie_memory","user def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item","of custom properties, e.g. 'age' = 39, 'color' = 'blue'","} # Add a concept to memory def add(self, c):","'None': return user def add_memory(self, a, b): c = '3sg'","DATE) values (?, ?)\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass()","import sqlite3 import os from datetime import date, datetime from","dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir","to memory def add(self, c): # add oncept to raw_input","(c,)) user = cursor.fetchone() # if user == 'None': return","return user def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE","'age' = 39, 'color' = 'blue' # Robbie memory class.","def add(self, c): # add oncept to raw_input table in","of concepts class memoryClass(): def __init__(self): self.concepts = {} self.person","location, person,DATE) values (?, ?, ?, ?)\", (a, b, c,","a thing is 3rd-person, singular self.isProperNoun = False # True","b): c = '3sg' dt = datetime.now() RM.execute(\"insert into memory","import pluralize, singularize from pattern.en import conjugate, lemma, lexeme #dir","self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' # Return True","import parse from pattern.en import pprint from pattern.en import parsetree","raw input and memories in parse taged columns ''' import","parse from pattern.en import pprint from pattern.en import parsetree from","= cursor.fetchone() # if user == 'None': return user def","sys import re import sqlite3 import os from datetime import","person,DATE) values (?, ?, ?, ?)\", (a, b, c, dt))","proper noun: e.g. Robert self.properties = {} # Dict of","import date, datetime from pattern.en import parse from pattern.en import","self.person[c] else: self.concepts[c].person = '3sg' # Return True if concept","conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM","concept = ? ''', (b, a)) RM.commit() def search_memory(self, a):","'2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its'","in sqlite data base holds raw input and memories in","'your', '3sg': 'its' } # Add a concept to memory","base holds raw input and memories in parse taged columns","def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location = ?","conceptClass() if c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person","'c' (string) is in memory def known(self, c): cursor.execute('''SELECT concept,","self.concepts = {} self.person = {'I': '1sg', 'you': '2sg' }","columns ''' import sys import re import sqlite3 import os","datetime import date, datetime from pattern.en import parse from pattern.en","singularize from pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__))"],"string":"[\n \"b, c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory\",\n \"into memory (concept, location, person,DATE) values (?, ?, ?, ?)\\\",\",\n \"= ? WHERE concept = ? ''', (b, a)) RM.commit()\",\n \"from pattern.en import pprint from pattern.en import parsetree from pattern.en\",\n \"(b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM\",\n \"what/how is 'concept' self.reference = 'none' # unused self.locality =\",\n \"3rd-person, singular self.isProperNoun = False # True if proper noun:\",\n \"# what/how is 'concept' self.reference = 'none' # unused self.locality\",\n \"a, b): c = '3sg' dt = datetime.now() RM.execute(\\\"insert into\",\n \"single concept class conceptClass: def __init__(self, state='none', locality='none'): self.state =\",\n \"holds raw input and memories in parse taged columns '''\",\n \"location = ? WHERE concept = ? ''', (b, a))\",\n \"memory def known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE\",\n \"conceptClass: def __init__(self, state='none', locality='none'): self.state = state # what/how\",\n \"datetime from pattern.en import parse from pattern.en import pprint from\",\n \"= conceptClass() if c in self.person: self.concepts[c].person = self.person[c] else:\",\n \"def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item =?''',\",\n \"memory WHERE concept =?''', (c,)) user = cursor.fetchone() # if\",\n \"# Information about a single concept class conceptClass: def __init__(self,\",\n \"c): cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''', (c,))\",\n \"== 'None': return user def add_memory(self, a, b): c =\",\n \"add oncept to raw_input table in robbie_memory # x= #\",\n \"(?, ?)\\\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if c\",\n \"= RM.cursor() # Information about a single concept class conceptClass:\",\n \"cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value FROM profile\",\n \"sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor()\",\n \"= datetime.now() # RM.execute(\\\"insert into RAW_INPUT (RAW, DATE) values (?,\",\n \"'3sg' dt = datetime.now() RM.execute(\\\"insert into memory (concept, location, person,DATE)\",\n \"self.locality = locality # where is 'concept' self.person = '3sg'\",\n \"memory class stored in sqlite data base holds raw input\",\n \"# where is 'concept' self.person = '3sg' # e.g. a\",\n \"= sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information about\",\n \"import sys import re import sqlite3 import os from datetime\",\n \"{'1sg': 'my', '2sg': 'your', '3sg': 'its' } # Add a\",\n \"import pprint from pattern.en import parsetree from pattern.en import wordnet\",\n \"state='none', locality='none'): self.state = state # what/how is 'concept' self.reference\",\n \"Robbie memory class. Collection of concepts class memoryClass(): def __init__(self):\",\n \"?)\\\", (a, b, c, dt)) RM.commit() def update_memory(self, a, b):\",\n \"from datetime import date, datetime from pattern.en import parse from\",\n \"RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor\",\n \"concept,location, person FROM memory WHERE concept =?''', (a,)) user =\",\n \"RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE\",\n \"''' memory class stored in sqlite data base holds raw\",\n \"'2sg': 'your', '3sg': 'its' } # Add a concept to\",\n \"SET location = ? WHERE concept = ? ''', (b,\",\n \"from pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir\",\n \"taged columns ''' import sys import re import sqlite3 import\",\n \"= state # what/how is 'concept' self.reference = 'none' #\",\n \"FROM memory WHERE concept =?''', (c,)) user = cursor.fetchone() #\",\n \"date, datetime from pattern.en import parse from pattern.en import pprint\",\n \"FROM profile WHERE item =?''', (a,)) user = cursor.fetchone() return\",\n \"parsetree from pattern.en import wordnet from pattern.en import pluralize, singularize\",\n \"'3sg' # e.g. a thing is 3rd-person, singular self.isProperNoun =\",\n \"class conceptClass: def __init__(self, state='none', locality='none'): self.state = state #\",\n \"'my', '2sg': 'your', '3sg': 'its' } # Add a concept\",\n \"self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg':\",\n \"= os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM\",\n \"pattern.en import parsetree from pattern.en import wordnet from pattern.en import\",\n \"c = '3sg' dt = datetime.now() RM.execute(\\\"insert into memory (concept,\",\n \"(a,)) user = cursor.fetchone() return user def Dump(self): return (self.concepts.state)\",\n \"lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir\",\n \"} self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its' }\",\n \"thing is 3rd-person, singular self.isProperNoun = False # True if\",\n \"import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/'\",\n \"def __init__(self, state='none', locality='none'): self.state = state # what/how is\",\n \"is 'concept' self.person = '3sg' # e.g. a thing is\",\n \"re import sqlite3 import os from datetime import date, datetime\",\n \"self.isProperNoun = False # True if proper noun: e.g. Robert\",\n \"__init__(self, state='none', locality='none'): self.state = state # what/how is 'concept'\",\n \"Add a concept to memory def add(self, c): # add\",\n \"raw_input table in robbie_memory # x= # dt = datetime.now()\",\n \"# unused self.locality = locality # where is 'concept' self.person\",\n \"= ? ''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT\",\n \"data base holds raw input and memories in parse taged\",\n \"import os from datetime import date, datetime from pattern.en import\",\n \"location FROM memory WHERE concept =?''', (c,)) user = cursor.fetchone()\",\n \"dt)) # RM.commit() self.concepts[c] = conceptClass() if c in self.person:\",\n \"table in robbie_memory # x= # dt = datetime.now() #\",\n \"concepts class memoryClass(): def __init__(self): self.concepts = {} self.person =\",\n \"unused self.locality = locality # where is 'concept' self.person =\",\n \"from pattern.en import pluralize, singularize from pattern.en import conjugate, lemma,\",\n \"'concept' self.reference = 'none' # unused self.locality = locality #\",\n \"'none' # unused self.locality = locality # where is 'concept'\",\n \"39, 'color' = 'blue' # Robbie memory class. Collection of\",\n \"values (?, ?, ?, ?)\\\", (a, b, c, dt)) RM.commit()\",\n \"= '3sg' # e.g. a thing is 3rd-person, singular self.isProperNoun\",\n \"# RM.commit() self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person\",\n \"'3sg' # Return True if concept 'c' (string) is in\",\n \"= 'none' # unused self.locality = locality # where is\",\n \"# True if proper noun: e.g. Robert self.properties = {}\",\n \"#dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite')\",\n \"into RAW_INPUT (RAW, DATE) values (?, ?)\\\",(c, dt)) # RM.commit()\",\n \"is 'concept' self.reference = 'none' # unused self.locality = locality\",\n \"? WHERE concept = ? ''', (b, a)) RM.commit() def\",\n \"''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person\",\n \"#!/usr/bin/python ''' memory class stored in sqlite data base holds\",\n \"{} # Dict of custom properties, e.g. 'age' = 39,\",\n \"'color' = 'blue' # Robbie memory class. Collection of concepts\",\n \"Robert self.properties = {} # Dict of custom properties, e.g.\",\n \"= locality # where is 'concept' self.person = '3sg' #\",\n \"is 3rd-person, singular self.isProperNoun = False # True if proper\",\n \"memory def add(self, c): # add oncept to raw_input table\",\n \"self.state = state # what/how is 'concept' self.reference = 'none'\",\n \"if user == 'None': return user def add_memory(self, a, b):\",\n \"= False # True if proper noun: e.g. Robert self.properties\",\n \"person FROM memory WHERE concept =?''', (a,)) user = cursor.fetchone()\",\n \"class. Collection of concepts class memoryClass(): def __init__(self): self.concepts =\",\n \"memories in parse taged columns ''' import sys import re\",\n \"(string) is in memory def known(self, c): cursor.execute('''SELECT concept, location\",\n \"RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location =\",\n \"'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg':\",\n \"c): # add oncept to raw_input table in robbie_memory #\",\n \"in robbie_memory # x= # dt = datetime.now() # RM.execute(\\\"insert\",\n \"memory WHERE concept =?''', (a,)) user = cursor.fetchone() return user\",\n \"add_memory(self, a, b): c = '3sg' dt = datetime.now() RM.execute(\\\"insert\",\n \"sqlite data base holds raw input and memories in parse\",\n \"from pattern.en import parsetree from pattern.en import wordnet from pattern.en\",\n \"sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information about a\",\n \"memory class. Collection of concepts class memoryClass(): def __init__(self): self.concepts\",\n \"wordnet from pattern.en import pluralize, singularize from pattern.en import conjugate,\",\n \"pprint from pattern.en import parsetree from pattern.en import wordnet from\",\n \"lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM =\",\n \"user def add_memory(self, a, b): c = '3sg' dt =\",\n \"parse taged columns ''' import sys import re import sqlite3\",\n \"# add oncept to raw_input table in robbie_memory # x=\",\n \"self.properties = {} # Dict of custom properties, e.g. 'age'\",\n \"(a,)) user = cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT\",\n \"user = cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value\",\n \"Collection of concepts class memoryClass(): def __init__(self): self.concepts = {}\",\n \"+ '/data/robbie_memory.db') cursor = RM.cursor() # Information about a single\",\n \"# e.g. a thing is 3rd-person, singular self.isProperNoun = False\",\n \"dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location\",\n \"value FROM profile WHERE item =?''', (a,)) user = cursor.fetchone()\",\n \"# Add a concept to memory def add(self, c): #\",\n \"'3sg': 'its' } # Add a concept to memory def\",\n \"=?''', (a,)) user = cursor.fetchone() return user def Dump(self): return\",\n \"RAW_INPUT (RAW, DATE) values (?, ?)\\\",(c, dt)) # RM.commit() self.concepts[c]\",\n \"user == 'None': return user def add_memory(self, a, b): c\",\n \"# Return True if concept 'c' (string) is in memory\",\n \"update_memory(self, a, b): cursor.execute('''UPDATE memory SET location = ? WHERE\",\n \"?, ?)\\\", (a, b, c, dt)) RM.commit() def update_memory(self, a,\",\n \"a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''', (a,))\",\n \"in memory def known(self, c): cursor.execute('''SELECT concept, location FROM memory\",\n \"b): cursor.execute('''UPDATE memory SET location = ? WHERE concept =\",\n \"WHERE concept =?''', (a,)) user = cursor.fetchone() return user def\",\n \"(concept, location, person,DATE) values (?, ?, ?, ?)\\\", (a, b,\",\n \"__init__(self): self.concepts = {} self.person = {'I': '1sg', 'you': '2sg'\",\n \"'1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your',\",\n \"sqlite3 import os from datetime import date, datetime from pattern.en\",\n \"= {} self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns\",\n \"FROM memory WHERE concept =?''', (a,)) user = cursor.fetchone() return\",\n \"RM.commit() self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person =\",\n \"memoryClass(): def __init__(self): self.concepts = {} self.person = {'I': '1sg',\",\n \"True if proper noun: e.g. Robert self.properties = {} #\",\n \"'blue' # Robbie memory class. Collection of concepts class memoryClass():\",\n \"state # what/how is 'concept' self.reference = 'none' # unused\",\n \"values (?, ?)\\\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if\",\n \"concept class conceptClass: def __init__(self, state='none', locality='none'): self.state = state\",\n \"e.g. 'age' = 39, 'color' = 'blue' # Robbie memory\",\n \"pluralize, singularize from pattern.en import conjugate, lemma, lexeme #dir =\",\n \"''' import sys import re import sqlite3 import os from\",\n \"Dict of custom properties, e.g. 'age' = 39, 'color' =\",\n \"dt = datetime.now() RM.execute(\\\"insert into memory (concept, location, person,DATE) values\",\n \"datetime.now() # RM.execute(\\\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\\\",(c,\",\n \"self.reference = 'none' # unused self.locality = locality # where\",\n \"item =?''', (a,)) user = cursor.fetchone() return user def Dump(self):\",\n \"datetime.now() RM.execute(\\\"insert into memory (concept, location, person,DATE) values (?, ?,\",\n \"= cursor.fetchone() return user def search_profile(self, a): cursor.execute('''SELECT value FROM\",\n \"where is 'concept' self.person = '3sg' # e.g. a thing\",\n \"False # True if proper noun: e.g. Robert self.properties =\",\n \"profile WHERE item =?''', (a,)) user = cursor.fetchone() return user\",\n \"#RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() # Information\",\n \"= '3sg' # Return True if concept 'c' (string) is\",\n \"WHERE concept = ? ''', (b, a)) RM.commit() def search_memory(self,\",\n \"dt = datetime.now() # RM.execute(\\\"insert into RAW_INPUT (RAW, DATE) values\",\n \"# Robbie memory class. Collection of concepts class memoryClass(): def\",\n \"self.concepts[c].person = '3sg' # Return True if concept 'c' (string)\",\n \"self.person = '3sg' # e.g. a thing is 3rd-person, singular\",\n \"(RAW, DATE) values (?, ?)\\\",(c, dt)) # RM.commit() self.concepts[c] =\",\n \"search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,))\",\n \"'its' } # Add a concept to memory def add(self,\",\n \"# if user == 'None': return user def add_memory(self, a,\",\n \"cursor.execute('''UPDATE memory SET location = ? WHERE concept = ?\",\n \"RM.execute(\\\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\\\",(c, dt)) #\",\n \"True if concept 'c' (string) is in memory def known(self,\",\n \"# dt = datetime.now() # RM.execute(\\\"insert into RAW_INPUT (RAW, DATE)\",\n \"pattern.en import pprint from pattern.en import parsetree from pattern.en import\",\n \"if c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person =\",\n \"{} self.person = {'I': '1sg', 'you': '2sg' } self.posessivePronouns =\",\n \"stored in sqlite data base holds raw input and memories\",\n \"import re import sqlite3 import os from datetime import date,\",\n \"= '3sg' dt = datetime.now() RM.execute(\\\"insert into memory (concept, location,\",\n \"e.g. Robert self.properties = {} # Dict of custom properties,\",\n \"= datetime.now() RM.execute(\\\"insert into memory (concept, location, person,DATE) values (?,\",\n \"memory SET location = ? WHERE concept = ? ''',\",\n \"= self.person[c] else: self.concepts[c].person = '3sg' # Return True if\",\n \"memory (concept, location, person,DATE) values (?, ?, ?, ?)\\\", (a,\",\n \"if concept 'c' (string) is in memory def known(self, c):\",\n \"e.g. a thing is 3rd-person, singular self.isProperNoun = False #\",\n \"if proper noun: e.g. Robert self.properties = {} # Dict\",\n \"?)\\\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass() if c in\",\n \"cursor = RM.cursor() # Information about a single concept class\",\n \"# Dict of custom properties, e.g. 'age' = 39, 'color'\",\n \"= 'blue' # Robbie memory class. Collection of concepts class\",\n \"from pattern.en import wordnet from pattern.en import pluralize, singularize from\",\n \"def known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE concept\",\n \"(?, ?, ?, ?)\\\", (a, b, c, dt)) RM.commit() def\",\n \"return user def add_memory(self, a, b): c = '3sg' dt\",\n \"else: self.concepts[c].person = '3sg' # Return True if concept 'c'\",\n \"self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' # Return\",\n \"c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg'\",\n \"os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM =\",\n \"'/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db')\",\n \"= {} # Dict of custom properties, e.g. 'age' =\",\n \"c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory SET\",\n \"Return True if concept 'c' (string) is in memory def\",\n \"pattern.en import parse from pattern.en import pprint from pattern.en import\",\n \"is in memory def known(self, c): cursor.execute('''SELECT concept, location FROM\",\n \"WHERE concept =?''', (c,)) user = cursor.fetchone() # if user\",\n \"about a single concept class conceptClass: def __init__(self, state='none', locality='none'):\",\n \"= '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir +\",\n \"?, ?, ?)\\\", (a, b, c, dt)) RM.commit() def update_memory(self,\",\n \"in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' #\",\n \"pattern.en import pluralize, singularize from pattern.en import conjugate, lemma, lexeme\",\n \"oncept to raw_input table in robbie_memory # x= # dt\",\n \"= 39, 'color' = 'blue' # Robbie memory class. Collection\",\n \"{'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my', '2sg':\",\n \"def add_memory(self, a, b): c = '3sg' dt = datetime.now()\",\n \"= {'1sg': 'my', '2sg': 'your', '3sg': 'its' } # Add\",\n \"concept =?''', (c,)) user = cursor.fetchone() # if user ==\",\n \"Information about a single concept class conceptClass: def __init__(self, state='none',\",\n \"WHERE item =?''', (a,)) user = cursor.fetchone() return user def\",\n \"import wordnet from pattern.en import pluralize, singularize from pattern.en import\",\n \"search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''',\",\n \"and memories in parse taged columns ''' import sys import\",\n \"'concept' self.person = '3sg' # e.g. a thing is 3rd-person,\",\n \"(a, b, c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE\",\n \"import parsetree from pattern.en import wordnet from pattern.en import pluralize,\",\n \"=?''', (c,)) user = cursor.fetchone() # if user == 'None':\",\n \"a single concept class conceptClass: def __init__(self, state='none', locality='none'): self.state\",\n \"# RM.execute(\\\"insert into RAW_INPUT (RAW, DATE) values (?, ?)\\\",(c, dt))\",\n \"to raw_input table in robbie_memory # x= # dt =\",\n \"cursor.execute('''SELECT concept,location, person FROM memory WHERE concept =?''', (a,)) user\",\n \"custom properties, e.g. 'age' = 39, 'color' = 'blue' #\",\n \"+'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor = RM.cursor() #\",\n \"self.concepts[c] = conceptClass() if c in self.person: self.concepts[c].person = self.person[c]\",\n \"concept 'c' (string) is in memory def known(self, c): cursor.execute('''SELECT\",\n \"locality='none'): self.state = state # what/how is 'concept' self.reference =\",\n \"'/data/robbie_memory.db') cursor = RM.cursor() # Information about a single concept\",\n \"os from datetime import date, datetime from pattern.en import parse\",\n \"class stored in sqlite data base holds raw input and\",\n \"cursor.fetchone() # if user == 'None': return user def add_memory(self,\",\n \"robbie_memory # x= # dt = datetime.now() # RM.execute(\\\"insert into\",\n \"=?''', (a,)) user = cursor.fetchone() return user def search_profile(self, a):\",\n \"? ''', (b, a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location,\",\n \"properties, e.g. 'age' = 39, 'color' = 'blue' # Robbie\",\n \"cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,)) user =\",\n \"input and memories in parse taged columns ''' import sys\",\n \"def __init__(self): self.concepts = {} self.person = {'I': '1sg', 'you':\",\n \"pattern.en import wordnet from pattern.en import pluralize, singularize from pattern.en\",\n \"# x= # dt = datetime.now() # RM.execute(\\\"insert into RAW_INPUT\",\n \"a concept to memory def add(self, c): # add oncept\",\n \"user = cursor.fetchone() # if user == 'None': return user\",\n \"self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its' } #\",\n \"= sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir + '/data/robbie_memory.db') cursor =\",\n \"in parse taged columns ''' import sys import re import\",\n \"a): cursor.execute('''SELECT value FROM profile WHERE item =?''', (a,)) user\",\n \"RM.cursor() # Information about a single concept class conceptClass: def\",\n \"locality # where is 'concept' self.person = '3sg' # e.g.\",\n \"concept =?''', (a,)) user = cursor.fetchone() return user def search_profile(self,\",\n \"known(self, c): cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''',\",\n \"a)) RM.commit() def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory\",\n \"def search_memory(self, a): cursor.execute('''SELECT concept,location, person FROM memory WHERE concept\",\n \"RM.execute(\\\"insert into memory (concept, location, person,DATE) values (?, ?, ?,\",\n \"noun: e.g. Robert self.properties = {} # Dict of custom\",\n \"singular self.isProperNoun = False # True if proper noun: e.g.\",\n \"cursor.execute('''SELECT concept, location FROM memory WHERE concept =?''', (c,)) user\",\n \"from pattern.en import parse from pattern.en import pprint from pattern.en\",\n \"concept to memory def add(self, c): # add oncept to\",\n \"pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir =\",\n \"class memoryClass(): def __init__(self): self.concepts = {} self.person = {'I':\",\n \"concept, location FROM memory WHERE concept =?''', (c,)) user =\",\n \"x= # dt = datetime.now() # RM.execute(\\\"insert into RAW_INPUT (RAW,\",\n \"a, b): cursor.execute('''UPDATE memory SET location = ? WHERE concept\",\n \"= {'I': '1sg', 'you': '2sg' } self.posessivePronouns = {'1sg': 'my',\",\n \"add(self, c): # add oncept to raw_input table in robbie_memory\",\n \"user def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE item\",\n \"of custom properties, e.g. 'age' = 39, 'color' = 'blue'\",\n \"} # Add a concept to memory def add(self, c):\",\n \"'None': return user def add_memory(self, a, b): c = '3sg'\",\n \"DATE) values (?, ?)\\\",(c, dt)) # RM.commit() self.concepts[c] = conceptClass()\",\n \"import sqlite3 import os from datetime import date, datetime from\",\n \"dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM = sqlite3.connect(dir +'robbie_memory.sqlite') #RM = sqlite3.connect(dir\",\n \"to memory def add(self, c): # add oncept to raw_input\",\n \"(c,)) user = cursor.fetchone() # if user == 'None': return\",\n \"return user def search_profile(self, a): cursor.execute('''SELECT value FROM profile WHERE\",\n \"'age' = 39, 'color' = 'blue' # Robbie memory class.\",\n \"def add(self, c): # add oncept to raw_input table in\",\n \"of concepts class memoryClass(): def __init__(self): self.concepts = {} self.person\",\n \"location, person,DATE) values (?, ?, ?, ?)\\\", (a, b, c,\",\n \"a thing is 3rd-person, singular self.isProperNoun = False # True\",\n \"b): c = '3sg' dt = datetime.now() RM.execute(\\\"insert into memory\",\n \"import pluralize, singularize from pattern.en import conjugate, lemma, lexeme #dir\",\n \"self.concepts[c].person = self.person[c] else: self.concepts[c].person = '3sg' # Return True\",\n \"import parse from pattern.en import pprint from pattern.en import parsetree\",\n \"raw input and memories in parse taged columns ''' import\",\n \"parse from pattern.en import pprint from pattern.en import parsetree from\",\n \"= cursor.fetchone() # if user == 'None': return user def\",\n \"sys import re import sqlite3 import os from datetime import\",\n \"person,DATE) values (?, ?, ?, ?)\\\", (a, b, c, dt))\",\n \"proper noun: e.g. Robert self.properties = {} # Dict of\",\n \"import date, datetime from pattern.en import parse from pattern.en import\",\n \"self.person[c] else: self.concepts[c].person = '3sg' # Return True if concept\",\n \"conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__)) dir = '/home/erni/catkin_ws/src/max_ros/max_ai/src/max_ai/' RM\",\n \"concept = ? ''', (b, a)) RM.commit() def search_memory(self, a):\",\n \"'2sg' } self.posessivePronouns = {'1sg': 'my', '2sg': 'your', '3sg': 'its'\",\n \"in sqlite data base holds raw input and memories in\",\n \"'your', '3sg': 'its' } # Add a concept to memory\",\n \"base holds raw input and memories in parse taged columns\",\n \"def update_memory(self, a, b): cursor.execute('''UPDATE memory SET location = ?\",\n \"conceptClass() if c in self.person: self.concepts[c].person = self.person[c] else: self.concepts[c].person\",\n \"'c' (string) is in memory def known(self, c): cursor.execute('''SELECT concept,\",\n \"self.concepts = {} self.person = {'I': '1sg', 'you': '2sg' }\",\n \"columns ''' import sys import re import sqlite3 import os\",\n \"datetime import date, datetime from pattern.en import parse from pattern.en\",\n \"singularize from pattern.en import conjugate, lemma, lexeme #dir = os.path.dirname(os.path.abspath(__file__))\"\n]","isConversational":false}}},{"rowIdx":688,"cells":{"ngram":{"kind":"list like","value":["= bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z = 10","# to make the bad region transparent. This is the","palette.set* lines, you will see # all the defaults; under","palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could use #","range to which the regular palette color scale is applied.","of BoundaryNorm to get a filled contour effect. ''' from","lines, you will see # all the defaults; under and","contour effect. ''' from pylab import * from numpy import","could use # palette.set_bad(alpha = 0.0) # to make the","> 1.2, Z) # By setting vmin and vmax in","out all the palette.set* lines, you will see # all","norm, we establish the # range to which the regular","over will be colored with the # first and last","interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax = 1.0,","Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm,","1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im,","colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette,","import ma import matplotlib.colors as colors delta = 0.025 x","the # first and last colors in the palette, respectively.","ma import matplotlib.colors as colors delta = 0.025 x =","palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm =","= colors.Normalize(vmin = -1.0, vmax = 1.0, clip = False),","# difference of Gaussians # Set up a colormap: palette","'''imshow with masked array input and out-of-range colors. The second","the default. # If you comment out all the palette.set*","the # range to which the regular palette color scale","-0.2, 0, 0.2, 0.5, 1], ncolors=256, clip = False), origin='lower',","The second subplot illustrates the use of BoundaryNorm to get","norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256,","use # palette.set_bad(alpha = 0.0) # to make the bad","be colored with the # first and last colors in","make the bad region transparent. This is the default. #","colors in the palette, respectively. Zm = ma.masked_where(Z > 1.2,","with masked array input and out-of-range colors. The second subplot","the defaults; under and over will be colored with the","# Alternatively, we could use # palette.set_bad(alpha = 0.0) #","# first and last colors in the palette, respectively. Zm","all the palette.set* lines, you will see # all the","masked array input and out-of-range colors. The second subplot illustrates","# Set up a colormap: palette = cm.gray palette.set_over('r', 1.0)","clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional',","This is the default. # If you comment out all","the bad region transparent. This is the default. # If","imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax =","* (Z2-Z1) # difference of Gaussians # Set up a","origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal', shrink=0.8) show()","1.0) # Alternatively, we could use # palette.set_bad(alpha = 0.0)","transparent. This is the default. # If you comment out","to which the regular palette color scale is applied. #","bad region transparent. This is the default. # If you","0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)","subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5,","setting vmin and vmax in the norm, we establish the","interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5,","y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2","and over will be colored with the # first and","vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high,","<filename>examples/pylab_examples/image_masked.py #!/usr/bin/env python '''imshow with masked array input and out-of-range","use of BoundaryNorm to get a filled contour effect. '''","python '''imshow with masked array input and out-of-range colors. The","you will see # all the defaults; under and over","extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm","ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both',","and vmax in the norm, we establish the # range","range is colored based on palette.set_over, etc. subplot(1,2,1) im =","palette, respectively. Zm = ma.masked_where(Z > 1.2, Z) # By","# If you comment out all the palette.set* lines, you","im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0,","and out-of-range colors. The second subplot illustrates the use of","1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could use","1) Z = 10 * (Z2-Z1) # difference of Gaussians","to make the bad region transparent. This is the default.","10 * (Z2-Z1) # difference of Gaussians # Set up","last colors in the palette, respectively. Zm = ma.masked_where(Z >","is applied. # Anything above that range is colored based","illustrates the use of BoundaryNorm to get a filled contour","colored based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear',","see # all the defaults; under and over will be","import * from numpy import ma import matplotlib.colors as colors","Alternatively, we could use # palette.set_bad(alpha = 0.0) # to","will be colored with the # first and last colors","delta) X, Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y,","the use of BoundaryNorm to get a filled contour effect.","scale is applied. # Anything above that range is colored","ma.masked_where(Z > 1.2, Z) # By setting vmin and vmax","matplotlib.colors as colors delta = 0.025 x = y =","from pylab import * from numpy import ma import matplotlib.colors","0.025 x = y = arange(-3.0, 3.0, delta) X, Y","= y = arange(-3.0, 3.0, delta) X, Y = meshgrid(x,","with the # first and last colors in the palette,","colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip =","from numpy import ma import matplotlib.colors as colors delta =","palette.set_bad('b', 1.0) # Alternatively, we could use # palette.set_bad(alpha =","* from numpy import ma import matplotlib.colors as colors delta","array input and out-of-range colors. The second subplot illustrates the","= meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0,","based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette,","BoundaryNorm to get a filled contour effect. ''' from pylab","first and last colors in the palette, respectively. Zm =","= bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X,","Z) # By setting vmin and vmax in the norm,","= False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal',","clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both',","Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest',","= imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0,","pylab import * from numpy import ma import matplotlib.colors as","1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5,","in the palette, respectively. Zm = ma.masked_where(Z > 1.2, Z)","cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1],","in the norm, we establish the # range to which","Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0,","0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z","= colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip","arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y) Z1 =","1.5, 0.5, 1, 1) Z = 10 * (Z2-Z1) #","y = arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y)","x = y = arange(-3.0, 3.0, delta) X, Y =","vmax in the norm, we establish the # range to","cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we","Z = 10 * (Z2-Z1) # difference of Gaussians #","numpy import ma import matplotlib.colors as colors delta = 0.025","default. # If you comment out all the palette.set* lines,","palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) #","1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1,","''' from pylab import * from numpy import ma import","1, 1) Z = 10 * (Z2-Z1) # difference of","= ma.masked_where(Z > 1.2, Z) # By setting vmin and","which the regular palette color scale is applied. # Anything","= 0.0) # to make the bad region transparent. This","palette.set_bad(alpha = 0.0) # to make the bad region transparent.","Zm = ma.masked_where(Z > 1.2, Z) # By setting vmin","bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z = 10 *","Y, 1.5, 0.5, 1, 1) Z = 10 * (Z2-Z1)","origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2)","filled contour effect. ''' from pylab import * from numpy","colors delta = 0.025 x = y = arange(-3.0, 3.0,","Gaussians # Set up a colormap: palette = cm.gray palette.set_over('r',","we could use # palette.set_bad(alpha = 0.0) # to make","difference of Gaussians # Set up a colormap: palette =","= False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal',","0.0) # to make the bad region transparent. This is","that range is colored based on palette.set_over, etc. subplot(1,2,1) im","of Gaussians # Set up a colormap: palette = cm.gray","colored with the # first and last colors in the","shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1,","Set up a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g',","on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm","etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin","= 10 * (Z2-Z1) # difference of Gaussians # Set","vmin and vmax in the norm, we establish the #","= 0.025 x = y = arange(-3.0, 3.0, delta) X,","a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b',","-0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip = False),","effect. ''' from pylab import * from numpy import ma","Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 =","all the defaults; under and over will be colored with","Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5,","color scale is applied. # Anything above that range is","Anything above that range is colored based on palette.set_over, etc.","orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm =","0.5, 1, 1) Z = 10 * (Z2-Z1) # difference","By setting vmin and vmax in the norm, we establish","(Z2-Z1) # difference of Gaussians # Set up a colormap:","the regular palette color scale is applied. # Anything above","bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y,","1.2, Z) # By setting vmin and vmax in the","delta = 0.025 x = y = arange(-3.0, 3.0, delta)","region transparent. This is the default. # If you comment","# all the defaults; under and over will be colored","# palette.set_bad(alpha = 0.0) # to make the bad region","extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im","norm = colors.Normalize(vmin = -1.0, vmax = 1.0, clip =","you comment out all the palette.set* lines, you will see","= 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad')","title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im =","a filled contour effect. ''' from pylab import * from","0, 0.2, 0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3])","palette color scale is applied. # Anything above that range","cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax = 1.0, clip","respectively. Zm = ma.masked_where(Z > 1.2, Z) # By setting","0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm')","= arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y) Z1","to get a filled contour effect. ''' from pylab import","up a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0)","regular palette color scale is applied. # Anything above that","will see # all the defaults; under and over will","above that range is colored based on palette.set_over, etc. subplot(1,2,1)","= imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax","under and over will be colored with the # first","palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could","defaults; under and over will be colored with the #","False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8)","# Anything above that range is colored based on palette.set_over,","im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2,","3.0, delta) X, Y = meshgrid(x, y) Z1 = bivariate_normal(X,","#!/usr/bin/env python '''imshow with masked array input and out-of-range colors.","Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z =","the palette, respectively. Zm = ma.masked_where(Z > 1.2, Z) #","colors.Normalize(vmin = -1.0, vmax = 1.0, clip = False), origin='lower',","out-of-range colors. The second subplot illustrates the use of BoundaryNorm","we establish the # range to which the regular palette","import matplotlib.colors as colors delta = 0.025 x = y","subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin =","is colored based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm,","colors. The second subplot illustrates the use of BoundaryNorm to","as colors delta = 0.025 x = y = arange(-3.0,","1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im,","0.2, 0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With","establish the # range to which the regular palette color","input and out-of-range colors. The second subplot illustrates the use","# range to which the regular palette color scale is","applied. # Anything above that range is colored based on","the palette.set* lines, you will see # all the defaults;","If you comment out all the palette.set* lines, you will","meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)","colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0)","the norm, we establish the # range to which the","= -1.0, vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3])","False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal', shrink=0.8)","second subplot illustrates the use of BoundaryNorm to get a","and last colors in the palette, respectively. Zm = ma.masked_where(Z","1.0) palette.set_bad('b', 1.0) # Alternatively, we could use # palette.set_bad(alpha","imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2,","get a filled contour effect. ''' from pylab import *","comment out all the palette.set* lines, you will see #","# By setting vmin and vmax in the norm, we","subplot illustrates the use of BoundaryNorm to get a filled","= cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively,","is the default. # If you comment out all the","-1.0, vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low,","X, Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0,"],"string":"[\n \"= bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z = 10\",\n \"# to make the bad region transparent. This is the\",\n \"palette.set* lines, you will see # all the defaults; under\",\n \"palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could use #\",\n \"range to which the regular palette color scale is applied.\",\n \"of BoundaryNorm to get a filled contour effect. ''' from\",\n \"lines, you will see # all the defaults; under and\",\n \"contour effect. ''' from pylab import * from numpy import\",\n \"could use # palette.set_bad(alpha = 0.0) # to make the\",\n \"> 1.2, Z) # By setting vmin and vmax in\",\n \"out all the palette.set* lines, you will see # all\",\n \"norm, we establish the # range to which the regular\",\n \"over will be colored with the # first and last\",\n \"interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax = 1.0,\",\n \"Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm,\",\n \"1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im,\",\n \"colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette,\",\n \"import ma import matplotlib.colors as colors delta = 0.025 x\",\n \"the # first and last colors in the palette, respectively.\",\n \"ma import matplotlib.colors as colors delta = 0.025 x =\",\n \"palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm =\",\n \"= colors.Normalize(vmin = -1.0, vmax = 1.0, clip = False),\",\n \"# difference of Gaussians # Set up a colormap: palette\",\n \"'''imshow with masked array input and out-of-range colors. The second\",\n \"the default. # If you comment out all the palette.set*\",\n \"the # range to which the regular palette color scale\",\n \"-0.2, 0, 0.2, 0.5, 1], ncolors=256, clip = False), origin='lower',\",\n \"The second subplot illustrates the use of BoundaryNorm to get\",\n \"norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256,\",\n \"use # palette.set_bad(alpha = 0.0) # to make the bad\",\n \"be colored with the # first and last colors in\",\n \"make the bad region transparent. This is the default. #\",\n \"colors in the palette, respectively. Zm = ma.masked_where(Z > 1.2,\",\n \"with masked array input and out-of-range colors. The second subplot\",\n \"the defaults; under and over will be colored with the\",\n \"# Alternatively, we could use # palette.set_bad(alpha = 0.0) #\",\n \"# first and last colors in the palette, respectively. Zm\",\n \"all the palette.set* lines, you will see # all the\",\n \"masked array input and out-of-range colors. The second subplot illustrates\",\n \"# Set up a colormap: palette = cm.gray palette.set_over('r', 1.0)\",\n \"clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional',\",\n \"This is the default. # If you comment out all\",\n \"the bad region transparent. This is the default. # If\",\n \"imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax =\",\n \"* (Z2-Z1) # difference of Gaussians # Set up a\",\n \"origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal', shrink=0.8) show()\",\n \"1.0) # Alternatively, we could use # palette.set_bad(alpha = 0.0)\",\n \"transparent. This is the default. # If you comment out\",\n \"to which the regular palette color scale is applied. #\",\n \"bad region transparent. This is the default. # If you\",\n \"0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)\",\n \"subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5,\",\n \"setting vmin and vmax in the norm, we establish the\",\n \"interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5,\",\n \"y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2\",\n \"and over will be colored with the # first and\",\n \"vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high,\",\n \"<filename>examples/pylab_examples/image_masked.py #!/usr/bin/env python '''imshow with masked array input and out-of-range\",\n \"use of BoundaryNorm to get a filled contour effect. '''\",\n \"python '''imshow with masked array input and out-of-range colors. The\",\n \"you will see # all the defaults; under and over\",\n \"extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm\",\n \"ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both',\",\n \"and vmax in the norm, we establish the # range\",\n \"range is colored based on palette.set_over, etc. subplot(1,2,1) im =\",\n \"palette, respectively. Zm = ma.masked_where(Z > 1.2, Z) # By\",\n \"# If you comment out all the palette.set* lines, you\",\n \"im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0,\",\n \"and out-of-range colors. The second subplot illustrates the use of\",\n \"1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could use\",\n \"1) Z = 10 * (Z2-Z1) # difference of Gaussians\",\n \"to make the bad region transparent. This is the default.\",\n \"10 * (Z2-Z1) # difference of Gaussians # Set up\",\n \"last colors in the palette, respectively. Zm = ma.masked_where(Z >\",\n \"is applied. # Anything above that range is colored based\",\n \"illustrates the use of BoundaryNorm to get a filled contour\",\n \"colored based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear',\",\n \"see # all the defaults; under and over will be\",\n \"import * from numpy import ma import matplotlib.colors as colors\",\n \"Alternatively, we could use # palette.set_bad(alpha = 0.0) # to\",\n \"will be colored with the # first and last colors\",\n \"delta) X, Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y,\",\n \"the use of BoundaryNorm to get a filled contour effect.\",\n \"scale is applied. # Anything above that range is colored\",\n \"ma.masked_where(Z > 1.2, Z) # By setting vmin and vmax\",\n \"matplotlib.colors as colors delta = 0.025 x = y =\",\n \"from pylab import * from numpy import ma import matplotlib.colors\",\n \"0.025 x = y = arange(-3.0, 3.0, delta) X, Y\",\n \"= y = arange(-3.0, 3.0, delta) X, Y = meshgrid(x,\",\n \"with the # first and last colors in the palette,\",\n \"colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip =\",\n \"from numpy import ma import matplotlib.colors as colors delta =\",\n \"palette.set_bad('b', 1.0) # Alternatively, we could use # palette.set_bad(alpha =\",\n \"* from numpy import ma import matplotlib.colors as colors delta\",\n \"array input and out-of-range colors. The second subplot illustrates the\",\n \"= meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0,\",\n \"based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette,\",\n \"BoundaryNorm to get a filled contour effect. ''' from pylab\",\n \"first and last colors in the palette, respectively. Zm =\",\n \"= bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X,\",\n \"Z) # By setting vmin and vmax in the norm,\",\n \"= False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal',\",\n \"clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both',\",\n \"Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest',\",\n \"= imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0,\",\n \"pylab import * from numpy import ma import matplotlib.colors as\",\n \"1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5,\",\n \"in the palette, respectively. Zm = ma.masked_where(Z > 1.2, Z)\",\n \"cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1],\",\n \"in the norm, we establish the # range to which\",\n \"Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0,\",\n \"0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z\",\n \"= colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip\",\n \"arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y) Z1 =\",\n \"1.5, 0.5, 1, 1) Z = 10 * (Z2-Z1) #\",\n \"y = arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y)\",\n \"x = y = arange(-3.0, 3.0, delta) X, Y =\",\n \"vmax in the norm, we establish the # range to\",\n \"cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we\",\n \"Z = 10 * (Z2-Z1) # difference of Gaussians #\",\n \"numpy import ma import matplotlib.colors as colors delta = 0.025\",\n \"default. # If you comment out all the palette.set* lines,\",\n \"palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) #\",\n \"1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1,\",\n \"''' from pylab import * from numpy import ma import\",\n \"1, 1) Z = 10 * (Z2-Z1) # difference of\",\n \"= ma.masked_where(Z > 1.2, Z) # By setting vmin and\",\n \"which the regular palette color scale is applied. # Anything\",\n \"= 0.0) # to make the bad region transparent. This\",\n \"palette.set_bad(alpha = 0.0) # to make the bad region transparent.\",\n \"Zm = ma.masked_where(Z > 1.2, Z) # By setting vmin\",\n \"bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z = 10 *\",\n \"Y, 1.5, 0.5, 1, 1) Z = 10 * (Z2-Z1)\",\n \"origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2)\",\n \"filled contour effect. ''' from pylab import * from numpy\",\n \"colors delta = 0.025 x = y = arange(-3.0, 3.0,\",\n \"Gaussians # Set up a colormap: palette = cm.gray palette.set_over('r',\",\n \"we could use # palette.set_bad(alpha = 0.0) # to make\",\n \"difference of Gaussians # Set up a colormap: palette =\",\n \"= False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal',\",\n \"0.0) # to make the bad region transparent. This is\",\n \"that range is colored based on palette.set_over, etc. subplot(1,2,1) im\",\n \"of Gaussians # Set up a colormap: palette = cm.gray\",\n \"colored with the # first and last colors in the\",\n \"shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1,\",\n \"Set up a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g',\",\n \"on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm\",\n \"etc. subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin\",\n \"= 10 * (Z2-Z1) # difference of Gaussians # Set\",\n \"vmin and vmax in the norm, we establish the #\",\n \"= 0.025 x = y = arange(-3.0, 3.0, delta) X,\",\n \"a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b',\",\n \"-0.5, -0.2, 0, 0.2, 0.5, 1], ncolors=256, clip = False),\",\n \"effect. ''' from pylab import * from numpy import ma\",\n \"Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 =\",\n \"all the defaults; under and over will be colored with\",\n \"Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y, 1.5,\",\n \"color scale is applied. # Anything above that range is\",\n \"Anything above that range is colored based on palette.set_over, etc.\",\n \"orientation='horizontal', shrink=0.8) subplot(1,2,2) im = imshow(Zm, interpolation='nearest', cmap=palette, norm =\",\n \"0.5, 1, 1) Z = 10 * (Z2-Z1) # difference\",\n \"By setting vmin and vmax in the norm, we establish\",\n \"(Z2-Z1) # difference of Gaussians # Set up a colormap:\",\n \"the regular palette color scale is applied. # Anything above\",\n \"bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0) Z2 = bivariate_normal(X, Y,\",\n \"1.2, Z) # By setting vmin and vmax in the\",\n \"delta = 0.025 x = y = arange(-3.0, 3.0, delta)\",\n \"region transparent. This is the default. # If you comment\",\n \"# all the defaults; under and over will be colored\",\n \"# palette.set_bad(alpha = 0.0) # to make the bad region\",\n \"extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im\",\n \"norm = colors.Normalize(vmin = -1.0, vmax = 1.0, clip =\",\n \"you comment out all the palette.set* lines, you will see\",\n \"= 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad')\",\n \"title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8) subplot(1,2,2) im =\",\n \"a filled contour effect. ''' from pylab import * from\",\n \"0, 0.2, 0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3])\",\n \"palette color scale is applied. # Anything above that range\",\n \"cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax = 1.0, clip\",\n \"respectively. Zm = ma.masked_where(Z > 1.2, Z) # By setting\",\n \"0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm')\",\n \"= arange(-3.0, 3.0, delta) X, Y = meshgrid(x, y) Z1\",\n \"to get a filled contour effect. ''' from pylab import\",\n \"up a colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0)\",\n \"regular palette color scale is applied. # Anything above that\",\n \"will see # all the defaults; under and over will\",\n \"above that range is colored based on palette.set_over, etc. subplot(1,2,1)\",\n \"= imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin = -1.0, vmax\",\n \"under and over will be colored with the # first\",\n \"palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could\",\n \"defaults; under and over will be colored with the #\",\n \"False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im, extend='both', orientation='horizontal', shrink=0.8)\",\n \"# Anything above that range is colored based on palette.set_over,\",\n \"im = imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2,\",\n \"3.0, delta) X, Y = meshgrid(x, y) Z1 = bivariate_normal(X,\",\n \"#!/usr/bin/env python '''imshow with masked array input and out-of-range colors.\",\n \"Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z =\",\n \"the palette, respectively. Zm = ma.masked_where(Z > 1.2, Z) #\",\n \"colors.Normalize(vmin = -1.0, vmax = 1.0, clip = False), origin='lower',\",\n \"out-of-range colors. The second subplot illustrates the use of BoundaryNorm\",\n \"we establish the # range to which the regular palette\",\n \"import matplotlib.colors as colors delta = 0.025 x = y\",\n \"subplot(1,2,1) im = imshow(Zm, interpolation='bilinear', cmap=palette, norm = colors.Normalize(vmin =\",\n \"is colored based on palette.set_over, etc. subplot(1,2,1) im = imshow(Zm,\",\n \"colors. The second subplot illustrates the use of BoundaryNorm to\",\n \"as colors delta = 0.025 x = y = arange(-3.0,\",\n \"1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low, Red=high, Blue=bad') colorbar(im,\",\n \"0.2, 0.5, 1], ncolors=256, clip = False), origin='lower', extent=[-3,3,-3,3]) title('With\",\n \"establish the # range to which the regular palette color\",\n \"input and out-of-range colors. The second subplot illustrates the use\",\n \"# range to which the regular palette color scale is\",\n \"applied. # Anything above that range is colored based on\",\n \"the palette.set* lines, you will see # all the defaults;\",\n \"If you comment out all the palette.set* lines, you will\",\n \"meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)\",\n \"colormap: palette = cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0)\",\n \"the norm, we establish the # range to which the\",\n \"= -1.0, vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3])\",\n \"False), origin='lower', extent=[-3,3,-3,3]) title('With BoundaryNorm') colorbar(im, extend='both', spacing='proportional', orientation='horizontal', shrink=0.8)\",\n \"second subplot illustrates the use of BoundaryNorm to get a\",\n \"and last colors in the palette, respectively. Zm = ma.masked_where(Z\",\n \"1.0) palette.set_bad('b', 1.0) # Alternatively, we could use # palette.set_bad(alpha\",\n \"imshow(Zm, interpolation='nearest', cmap=palette, norm = colors.BoundaryNorm([-1, -0.5, -0.2, 0, 0.2,\",\n \"get a filled contour effect. ''' from pylab import *\",\n \"comment out all the palette.set* lines, you will see #\",\n \"# By setting vmin and vmax in the norm, we\",\n \"subplot illustrates the use of BoundaryNorm to get a filled\",\n \"= cm.gray palette.set_over('r', 1.0) palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively,\",\n \"is the default. # If you comment out all the\",\n \"-1.0, vmax = 1.0, clip = False), origin='lower', extent=[-3,3,-3,3]) title('Green=low,\",\n \"X, Y = meshgrid(x, y) Z1 = bivariate_normal(X, Y, 1.0,\"\n]","isConversational":false}}},{"rowIdx":689,"cells":{"ngram":{"kind":"list like","value":["List, Dict from datetime import datetime class Authenticate(BaseModel): access_token: str","typing import List, Dict from datetime import datetime class Authenticate(BaseModel):","import List, Dict from datetime import datetime class Authenticate(BaseModel): access_token:","validator, Field from typing import List, Dict from datetime import","from pydantic import BaseModel, validator, Field from typing import List,","Field from typing import List, Dict from datetime import datetime","from typing import List, Dict from datetime import datetime class","pydantic import BaseModel, validator, Field from typing import List, Dict","import BaseModel, validator, Field from typing import List, Dict from","BaseModel, validator, Field from typing import List, Dict from datetime"],"string":"[\n \"List, Dict from datetime import datetime class Authenticate(BaseModel): access_token: str\",\n \"typing import List, Dict from datetime import datetime class Authenticate(BaseModel):\",\n \"import List, Dict from datetime import datetime class Authenticate(BaseModel): access_token:\",\n \"validator, Field from typing import List, Dict from datetime import\",\n \"from pydantic import BaseModel, validator, Field from typing import List,\",\n \"Field from typing import List, Dict from datetime import datetime\",\n \"from typing import List, Dict from datetime import datetime class\",\n \"pydantic import BaseModel, validator, Field from typing import List, Dict\",\n \"import BaseModel, validator, Field from typing import List, Dict from\",\n \"BaseModel, validator, Field from typing import List, Dict from datetime\"\n]","isConversational":false}}},{"rowIdx":690,"cells":{"ngram":{"kind":"list like","value":["\"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ]","requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed =","return elif random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub","Randomly spawns memes. \"\"\" subreddits = [ \"dankmemes\", \"wholesomememes\", \"memes\",","Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On","\"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def __init__(self, bot): self.bot = bot","import discord from discord.ext import commands import requests import random","= bot @commands.Cog.listener() async def on_message(self, message): if message.author ==","\"tumblr\" ] def __init__(self, bot): self.bot = bot @commands.Cog.listener() async","Box class WildMemes(commands.Cog): \"\"\" Randomly spawns memes. \"\"\" subreddits =","elif random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub =","image = data.url upvotes = data.ups title = data.title subreddit","= discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme has appeared!\")","\"\"\" Randomly spawns memes. \"\"\" subreddits = [ \"dankmemes\", \"wholesomememes\",","__init__(self, bot): self.bot = bot @commands.Cog.listener() async def on_message(self, message):","bot): self.bot = bot @commands.Cog.listener() async def on_message(self, message): if","embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\") await message.channel.send(embed=embed) def setup(bot): bot.add_cog(WildMemes(bot))","async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent':","import random from box import Box class WildMemes(commands.Cog): \"\"\" Randomly","message): if message.author == self.bot.user: return elif random.randint(0, 100) <","= Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes =","from box import Box class WildMemes(commands.Cog): \"\"\" Randomly spawns memes.","Bot 9000'}) r = r.json() boxed = Box(r) data =","= data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild","embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme has","memes. \"\"\" subreddits = [ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\",","== self.bot.user: return elif random.randint(0, 100) < 25: async with","on_message(self, message): if message.author == self.bot.user: return elif random.randint(0, 100)","color=0x6bdcd7) embed.set_author(name=\"A wild meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with","r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image = data.url","[ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\"","commands import requests import random from box import Box class","= requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed","class WildMemes(commands.Cog): \"\"\" Randomly spawns memes. \"\"\" subreddits = [","requests import random from box import Box class WildMemes(commands.Cog): \"\"\"","self.bot = bot @commands.Cog.listener() async def on_message(self, message): if message.author","discord from discord.ext import commands import requests import random from","random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits)","upvotes = data.ups title = data.title subreddit = data.subreddit_name_prefixed embed","with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super","bot @commands.Cog.listener() async def on_message(self, message): if message.author == self.bot.user:","{title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit}","data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme","box import Box class WildMemes(commands.Cog): \"\"\" Randomly spawns memes. \"\"\"","@commands.Cog.listener() async def on_message(self, message): if message.author == self.bot.user: return","data.url upvotes = data.ups title = data.title subreddit = data.subreddit_name_prefixed","random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r =","'Super Bot 9000'}) r = r.json() boxed = Box(r) data","import Box class WildMemes(commands.Cog): \"\"\" Randomly spawns memes. \"\"\" subreddits","self.bot.user: return elif random.randint(0, 100) < 25: async with message.channel.typing():","import commands import requests import random from box import Box","= (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title =","25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\",","appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\") await message.channel.send(embed=embed) def","9000'}) r = r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data","= data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}',","r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r = r.json()","data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7)","embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\") await message.channel.send(embed=embed) def setup(bot):","WildMemes(commands.Cog): \"\"\" Randomly spawns memes. \"\"\" subreddits = [ \"dankmemes\",","\"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def __init__(self, bot): self.bot =","data = (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title","from discord.ext import commands import requests import random from box","\"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def __init__(self, bot):","message.author == self.bot.user: return elif random.randint(0, 100) < 25: async","has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\") await message.channel.send(embed=embed)","100) < 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r","boxed = Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes","meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\") await","<reponame>aniket091/modmail-plugins-1 import discord from discord.ext import commands import requests import","\"\"\" subreddits = [ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\",","Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups","def __init__(self, bot): self.bot = bot @commands.Cog.listener() async def on_message(self,","if message.author == self.bot.user: return elif random.randint(0, 100) < 25:","discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A wild meme has appeared!\") embed.set_image(url=image)","\"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def __init__(self,","import requests import random from box import Box class WildMemes(commands.Cog):","= r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image =","chosen_sub = random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'})","spawns memes. \"\"\" subreddits = [ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\",","wild meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes} upvotes.\")","r = r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image","async def on_message(self, message): if message.author == self.bot.user: return elif","] def __init__(self, bot): self.bot = bot @commands.Cog.listener() async def","embed.set_author(name=\"A wild meme has appeared!\") embed.set_image(url=image) embed.set_footer(text=f\"On {subreddit} with {upvotes}","= data.ups title = data.title subreddit = data.subreddit_name_prefixed embed =","title = data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title:","\"fellowkids\", \"tumblr\" ] def __init__(self, bot): self.bot = bot @commands.Cog.listener()","data.ups title = data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme","subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\"A","= [ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\",","\"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def","random from box import Box class WildMemes(commands.Cog): \"\"\" Randomly spawns","def on_message(self, message): if message.author == self.bot.user: return elif random.randint(0,","subreddits = [ \"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\",","= data.url upvotes = data.ups title = data.title subreddit =","< 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r =","message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot","discord.ext import commands import requests import random from box import","= random.choice(self.subreddits) r = requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r","(random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title = data.title","headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed = Box(r)","\"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ] def __init__(self, bot): self.bot"],"string":"[\n \"\\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ]\",\n \"requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed =\",\n \"return elif random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub\",\n \"Randomly spawns memes. \\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\",\",\n \"Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On\",\n \"\\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self, bot): self.bot = bot\",\n \"import discord from discord.ext import commands import requests import random\",\n \"= bot @commands.Cog.listener() async def on_message(self, message): if message.author ==\",\n \"\\\"tumblr\\\" ] def __init__(self, bot): self.bot = bot @commands.Cog.listener() async\",\n \"Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\" subreddits =\",\n \"elif random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub =\",\n \"image = data.url upvotes = data.ups title = data.title subreddit\",\n \"= discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has appeared!\\\")\",\n \"\\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\",\",\n \"__init__(self, bot): self.bot = bot @commands.Cog.listener() async def on_message(self, message):\",\n \"bot): self.bot = bot @commands.Cog.listener() async def on_message(self, message): if\",\n \"embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\") await message.channel.send(embed=embed) def setup(bot): bot.add_cog(WildMemes(bot))\",\n \"async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent':\",\n \"import random from box import Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly\",\n \"message): if message.author == self.bot.user: return elif random.randint(0, 100) <\",\n \"= Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes =\",\n \"from box import Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes.\",\n \"Bot 9000'}) r = r.json() boxed = Box(r) data =\",\n \"= data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild\",\n \"embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has\",\n \"memes. \\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\",\",\n \"== self.bot.user: return elif random.randint(0, 100) < 25: async with\",\n \"on_message(self, message): if message.author == self.bot.user: return elif random.randint(0, 100)\",\n \"color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with\",\n \"r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image = data.url\",\n \"[ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\"\",\n \"commands import requests import random from box import Box class\",\n \"= requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed\",\n \"class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\" subreddits = [\",\n \"requests import random from box import Box class WildMemes(commands.Cog): \\\"\\\"\\\"\",\n \"self.bot = bot @commands.Cog.listener() async def on_message(self, message): if message.author\",\n \"discord from discord.ext import commands import requests import random from\",\n \"random.randint(0, 100) < 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits)\",\n \"upvotes = data.ups title = data.title subreddit = data.subreddit_name_prefixed embed\",\n \"with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super\",\n \"bot @commands.Cog.listener() async def on_message(self, message): if message.author == self.bot.user:\",\n \"{title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit}\",\n \"data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme\",\n \"box import Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\"\",\n \"@commands.Cog.listener() async def on_message(self, message): if message.author == self.bot.user: return\",\n \"data.url upvotes = data.ups title = data.title subreddit = data.subreddit_name_prefixed\",\n \"random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'}) r =\",\n \"'Super Bot 9000'}) r = r.json() boxed = Box(r) data\",\n \"import Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\" subreddits\",\n \"self.bot.user: return elif random.randint(0, 100) < 25: async with message.channel.typing():\",\n \"import commands import requests import random from box import Box\",\n \"= (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title =\",\n \"25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\",\",\n \"appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\") await message.channel.send(embed=embed) def\",\n \"9000'}) r = r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data\",\n \"= data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}',\",\n \"r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'}) r = r.json()\",\n \"data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7)\",\n \"embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\") await message.channel.send(embed=embed) def setup(bot):\",\n \"WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns memes. \\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\",\",\n \"\\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self, bot): self.bot =\",\n \"data = (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title\",\n \"from discord.ext import commands import requests import random from box\",\n \"\\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self, bot):\",\n \"message.author == self.bot.user: return elif random.randint(0, 100) < 25: async\",\n \"has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\") await message.channel.send(embed=embed)\",\n \"100) < 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r\",\n \"boxed = Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes\",\n \"meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\") await\",\n \"<reponame>aniket091/modmail-plugins-1 import discord from discord.ext import commands import requests import\",\n \"\\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\",\",\n \"Box(r) data = (random.choice(boxed.data.children)).data image = data.url upvotes = data.ups\",\n \"def __init__(self, bot): self.bot = bot @commands.Cog.listener() async def on_message(self,\",\n \"if message.author == self.bot.user: return elif random.randint(0, 100) < 25:\",\n \"discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A wild meme has appeared!\\\") embed.set_image(url=image)\",\n \"\\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self,\",\n \"import requests import random from box import Box class WildMemes(commands.Cog):\",\n \"= r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image =\",\n \"chosen_sub = random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'})\",\n \"spawns memes. \\\"\\\"\\\" subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\",\",\n \"wild meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes} upvotes.\\\")\",\n \"r = r.json() boxed = Box(r) data = (random.choice(boxed.data.children)).data image\",\n \"async def on_message(self, message): if message.author == self.bot.user: return elif\",\n \"] def __init__(self, bot): self.bot = bot @commands.Cog.listener() async def\",\n \"embed.set_author(name=\\\"A wild meme has appeared!\\\") embed.set_image(url=image) embed.set_footer(text=f\\\"On {subreddit} with {upvotes}\",\n \"= data.ups title = data.title subreddit = data.subreddit_name_prefixed embed =\",\n \"title = data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title:\",\n \"\\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self, bot): self.bot = bot @commands.Cog.listener()\",\n \"data.ups title = data.title subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme\",\n \"subreddit = data.subreddit_name_prefixed embed = discord.Embed(title=f'Meme Title: {title}', color=0x6bdcd7) embed.set_author(name=\\\"A\",\n \"= [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\",\",\n \"\\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def\",\n \"random from box import Box class WildMemes(commands.Cog): \\\"\\\"\\\" Randomly spawns\",\n \"def on_message(self, message): if message.author == self.bot.user: return elif random.randint(0,\",\n \"subreddits = [ \\\"dankmemes\\\", \\\"wholesomememes\\\", \\\"memes\\\", \\\"terriblefacebookmemes\\\", \\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\",\",\n \"= data.url upvotes = data.ups title = data.title subreddit =\",\n \"< 25: async with message.channel.typing(): chosen_sub = random.choice(self.subreddits) r =\",\n \"message.channel.typing(): chosen_sub = random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot\",\n \"discord.ext import commands import requests import random from box import\",\n \"= random.choice(self.subreddits) r = requests.get(f\\\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\\\", headers={'User-agent': 'Super Bot 9000'}) r\",\n \"(random.choice(boxed.data.children)).data image = data.url upvotes = data.ups title = data.title\",\n \"headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed = Box(r)\",\n \"\\\"historymemes\\\", \\\"me_irl\\\", \\\"2meirl4meirl\\\", \\\"fellowkids\\\", \\\"tumblr\\\" ] def __init__(self, bot): self.bot\"\n]","isConversational":false}}},{"rowIdx":691,"cells":{"ngram":{"kind":"list like","value":["Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output","[r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server","<NAME> # <EMAIL> # <EMAIL> # # 10/24/2014 Original Construction","'Machine') # Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz","10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self): self.output =","def __init__(self): self.output = [] self.is_compliant = False self.uuid =","= \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): # Initialize Compliance self.is_compliant =","self.is_compliant = False return self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item -path","Finding: def __init__(self): self.output = [] self.is_compliant = False self.uuid","self.is_compliant = False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): #","NT\\CurrentVersion\") for sz in multi_sz: if sz.lower() not in rec_multi_sz.lower():","cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] +","################################################################################ # 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL>","#!/usr/bin/python ################################################################################ # 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> #","'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type","-path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type MultiString -value","self.is_compliant = True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths',","-path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\")","# Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz","Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in multi_sz: if sz.lower() not in","= [] self.is_compliant = False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self,","= False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): # Initialize","# 10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self): self.output","Compliance self.is_compliant = True # Get Registry MultiSZ multi_sz =","not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self,","Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ","MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in multi_sz:","= False return self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\")","'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'","__init__(self): self.output = [] self.is_compliant = False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\"","self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): # Initialize Compliance self.is_compliant","# # 10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self):","True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') #","('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows","-path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine'","[] self.is_compliant = False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli):","# <EMAIL> # # 10/24/2014 Original Construction ################################################################################ class Finding:","sz in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant =","def fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item","# Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output","20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL> # #","return self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path","if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant","# Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz #","# 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL> #","# <EMAIL> # <EMAIL> # # 10/24/2014 Original Construction ################################################################################","for sz in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant","Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines","cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty","self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz","Initialize Compliance self.is_compliant = True # Get Registry MultiSZ multi_sz","'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server","<EMAIL> # <EMAIL> # # 10/24/2014 Original Construction ################################################################################ class","cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\")","self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\")","self.output = [] self.is_compliant = False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def","False self.uuid = \"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): # Initialize Compliance","# <NAME> # <EMAIL> # <EMAIL> # # 10/24/2014 Original","Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in","+ multi_sz # Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\")","rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in multi_sz: if","fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path","\"20bdcef0-5cc5-11e4-af55-00155d01fe08\" def check(self, cli): # Initialize Compliance self.is_compliant = True","<EMAIL> # # 10/24/2014 Original Construction ################################################################################ class Finding: def","def check(self, cli): # Initialize Compliance self.is_compliant = True #","cli): # Initialize Compliance self.is_compliant = True # Get Registry","(\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in multi_sz: if sz.lower() not","-path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path","= (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for sz in multi_sz: if sz.lower()","in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self, cli):","= cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')]","cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type MultiString","= True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine')","cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg'\") cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths'\") cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name","Original Construction ################################################################################ class Finding: def __init__(self): self.output = []","Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended","rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item","sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def","class Finding: def __init__(self): self.output = [] self.is_compliant = False","False return self.is_compliant def fix(self, cli): cli.powershell(r\"New-Item -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers'\") cli.powershell(r\"New-Item","################################################################################ class Finding: def __init__(self): self.output = [] self.is_compliant =","MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output =","# # <NAME> # <EMAIL> # <EMAIL> # # 10/24/2014","in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False","cli.powershell(r\"Set-ItemProperty -path 'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows","Construction ################################################################################ class Finding: def __init__(self): self.output = [] self.is_compliant","check(self, cli): # Initialize Compliance self.is_compliant = True # Get","# Initialize Compliance self.is_compliant = True # Get Registry MultiSZ","multi_sz # Recommended MultiSZ rec_multi_sz = (\"System\\CurrentControlSet\\Control\\ProductOptions,System\\CurrentControlSet\\Control\\Server Applications,Software\\Microsoft\\Windows NT\\CurrentVersion\") for","= [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz =","multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths',","multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return"],"string":"[\n \"Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines self.output\",\n \"[r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server\",\n \"<NAME> # <EMAIL> # <EMAIL> # # 10/24/2014 Original Construction\",\n \"'Machine') # Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz\",\n \"10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self): self.output =\",\n \"def __init__(self): self.output = [] self.is_compliant = False self.uuid =\",\n \"= \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): # Initialize Compliance self.is_compliant =\",\n \"self.is_compliant = False return self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item -path\",\n \"Finding: def __init__(self): self.output = [] self.is_compliant = False self.uuid\",\n \"self.is_compliant = False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): #\",\n \"NT\\\\CurrentVersion\\\") for sz in multi_sz: if sz.lower() not in rec_multi_sz.lower():\",\n \"cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] +\",\n \"################################################################################ # 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL>\",\n \"#!/usr/bin/python ################################################################################ # 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> #\",\n \"'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type\",\n \"-path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type MultiString -value\",\n \"self.is_compliant = True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths',\",\n \"-path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\")\",\n \"# Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz\",\n \"Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in multi_sz: if sz.lower() not in\",\n \"= [] self.is_compliant = False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self,\",\n \"= False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): # Initialize\",\n \"# 10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self): self.output\",\n \"Compliance self.is_compliant = True # Get Registry MultiSZ multi_sz =\",\n \"not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self,\",\n \"Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ\",\n \"MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in multi_sz:\",\n \"= False return self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\")\",\n \"'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\",\n \"__init__(self): self.output = [] self.is_compliant = False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\"\",\n \"self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): # Initialize Compliance self.is_compliant\",\n \"# # 10/24/2014 Original Construction ################################################################################ class Finding: def __init__(self):\",\n \"True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') #\",\n \"('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows\",\n \"-path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine'\",\n \"[] self.is_compliant = False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli):\",\n \"# <EMAIL> # # 10/24/2014 Original Construction ################################################################################ class Finding:\",\n \"sz in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant =\",\n \"def fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item\",\n \"# Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output\",\n \"20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL> # #\",\n \"return self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path\",\n \"if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant\",\n \"# Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz #\",\n \"# 20bdcef0-5cc5-11e4-af55-00155d01fe08 # # <NAME> # <EMAIL> # <EMAIL> #\",\n \"# <EMAIL> # <EMAIL> # # 10/24/2014 Original Construction ################################################################################\",\n \"for sz in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant\",\n \"Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines\",\n \"cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty\",\n \"self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz\",\n \"Initialize Compliance self.is_compliant = True # Get Registry MultiSZ multi_sz\",\n \"'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server\",\n \"<EMAIL> # <EMAIL> # # 10/24/2014 Original Construction ################################################################################ class\",\n \"cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\")\",\n \"self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\")\",\n \"self.output = [] self.is_compliant = False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def\",\n \"False self.uuid = \\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): # Initialize Compliance\",\n \"# <NAME> # <EMAIL> # <EMAIL> # # 10/24/2014 Original\",\n \"Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in\",\n \"+ multi_sz # Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\")\",\n \"rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in multi_sz: if\",\n \"fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path\",\n \"\\\"20bdcef0-5cc5-11e4-af55-00155d01fe08\\\" def check(self, cli): # Initialize Compliance self.is_compliant = True\",\n \"<EMAIL> # # 10/24/2014 Original Construction ################################################################################ class Finding: def\",\n \"def check(self, cli): # Initialize Compliance self.is_compliant = True #\",\n \"cli): # Initialize Compliance self.is_compliant = True # Get Registry\",\n \"(\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in multi_sz: if sz.lower() not\",\n \"-path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path\",\n \"= (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for sz in multi_sz: if sz.lower()\",\n \"in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self, cli):\",\n \"= cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')]\",\n \"cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type MultiString\",\n \"= True # Get Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine')\",\n \"cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg'\\\") cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths'\\\") cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name\",\n \"Original Construction ################################################################################ class Finding: def __init__(self): self.output = []\",\n \"Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended\",\n \"rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item\",\n \"sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return self.is_compliant def\",\n \"class Finding: def __init__(self): self.output = [] self.is_compliant = False\",\n \"False return self.is_compliant def fix(self, cli): cli.powershell(r\\\"New-Item -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers'\\\") cli.powershell(r\\\"New-Item\",\n \"################################################################################ class Finding: def __init__(self): self.output = [] self.is_compliant =\",\n \"MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines self.output =\",\n \"# # <NAME> # <EMAIL> # <EMAIL> # # 10/24/2014\",\n \"in multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False\",\n \"cli.powershell(r\\\"Set-ItemProperty -path 'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths' -name 'Machine' -Type MultiString -value System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows\",\n \"Construction ################################################################################ class Finding: def __init__(self): self.output = [] self.is_compliant\",\n \"check(self, cli): # Initialize Compliance self.is_compliant = True # Get\",\n \"# Initialize Compliance self.is_compliant = True # Get Registry MultiSZ\",\n \"multi_sz # Recommended MultiSZ rec_multi_sz = (\\\"System\\\\CurrentControlSet\\\\Control\\\\ProductOptions,System\\\\CurrentControlSet\\\\Control\\\\Server Applications,Software\\\\Microsoft\\\\Windows NT\\\\CurrentVersion\\\") for\",\n \"= [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended MultiSZ rec_multi_sz =\",\n \"multi_sz = cli.get_reg_multi_sz(r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths', 'Machine') # Output Lines self.output = [r'HKLM:\\\\SYSTEM\\\\CurrentControlSet\\\\control\\\\SecurePipeServers\\\\winreg\\\\allowedExactPaths',\",\n \"multi_sz: if sz.lower() not in rec_multi_sz.lower(): self.is_compliant = False return\"\n]","isConversational":false}}},{"rowIdx":692,"cells":{"ngram":{"kind":"list like","value":["MikanException(Exception): \"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot convert a","<reponame>dzzhvks94vd2/mikan class MikanException(Exception): \"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot","class MikanException(Exception): \"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot convert","\"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot convert a string\"\"\""],"string":"[\n \"MikanException(Exception): \\\"\\\"\\\"Generic Mikan exception\\\"\\\"\\\" class ConversionError(MikanException, ValueError): \\\"\\\"\\\"Cannot convert a\",\n \"<reponame>dzzhvks94vd2/mikan class MikanException(Exception): \\\"\\\"\\\"Generic Mikan exception\\\"\\\"\\\" class ConversionError(MikanException, ValueError): \\\"\\\"\\\"Cannot\",\n \"class MikanException(Exception): \\\"\\\"\\\"Generic Mikan exception\\\"\\\"\\\" class ConversionError(MikanException, ValueError): \\\"\\\"\\\"Cannot convert\",\n \"\\\"\\\"\\\"Generic Mikan exception\\\"\\\"\\\" class ConversionError(MikanException, ValueError): \\\"\\\"\\\"Cannot convert a string\\\"\\\"\\\"\"\n]","isConversational":false}}},{"rowIdx":693,"cells":{"ngram":{"kind":"list like","value":["path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/',","name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'),","path from django.contrib.auth import views as auth_views from . import","<reponame>hhhameem/CV-Recommender<gh_stars>1-10 from django.urls import path from django.contrib.auth import views as","auth_views from . import views urlpatterns = [ path('register/', views.register,","path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/',","import views as auth_views from . import views urlpatterns =","views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(),","auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(),","import views urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin,","name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'),","name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'),","from . import views urlpatterns = [ path('register/', views.register, name='register'),","views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(),","django.contrib.auth import views as auth_views from . import views urlpatterns","path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/',","path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/',","views urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'),","views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit,","name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'),","views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(),","auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(),","path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit, name='editrecruiterprofile'), ]","name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'),","name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'),","auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(),","path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/',","as auth_views from . import views urlpatterns = [ path('register/',","name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'),","path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/',","name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'),","django.urls import path from django.contrib.auth import views as auth_views from","name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit, name='editrecruiterprofile'),","import path from django.contrib.auth import views as auth_views from .",". import views urlpatterns = [ path('register/', views.register, name='register'), path('login/',","name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'),","views as auth_views from . import views urlpatterns = [","[ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'),","= [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout,","path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/',","auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard,","path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/',","path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/',","auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit,","from django.contrib.auth import views as auth_views from . import views","auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard,","urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/',","from django.urls import path from django.contrib.auth import views as auth_views","path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/',"],"string":"[\n \"path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/',\",\n \"name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'),\",\n \"path from django.contrib.auth import views as auth_views from . import\",\n \"<reponame>hhhameem/CV-Recommender<gh_stars>1-10 from django.urls import path from django.contrib.auth import views as\",\n \"auth_views from . import views urlpatterns = [ path('register/', views.register,\",\n \"path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/',\",\n \"import views as auth_views from . import views urlpatterns =\",\n \"views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(),\",\n \"auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(),\",\n \"import views urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin,\",\n \"name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'),\",\n \"name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'),\",\n \"from . import views urlpatterns = [ path('register/', views.register, name='register'),\",\n \"views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(),\",\n \"django.contrib.auth import views as auth_views from . import views urlpatterns\",\n \"path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/',\",\n \"path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/',\",\n \"views urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'),\",\n \"views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit,\",\n \"name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'),\",\n \"views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(),\",\n \"auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(),\",\n \"path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit, name='editrecruiterprofile'), ]\",\n \"name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'),\",\n \"name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'),\",\n \"auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(),\",\n \"path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/',\",\n \"as auth_views from . import views urlpatterns = [ path('register/',\",\n \"name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'),\",\n \"path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/',\",\n \"name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'),\",\n \"django.urls import path from django.contrib.auth import views as auth_views from\",\n \"name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit, name='editapplicantprofile'), path('recruiterdashboard/profile-edit/', views.recruiteredit, name='editrecruiterprofile'),\",\n \"import path from django.contrib.auth import views as auth_views from .\",\n \". import views urlpatterns = [ path('register/', views.register, name='register'), path('login/',\",\n \"name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'),\",\n \"views as auth_views from . import views urlpatterns = [\",\n \"[ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'),\",\n \"= [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout,\",\n \"path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/',\",\n \"auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard,\",\n \"path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/', views.userlogout, name='logout'), path('password_change/',\",\n \"path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/',\",\n \"auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard, name='recruiterdashboard'), path('applicantdashboard/profile-edit/', views.applicantedit,\",\n \"from django.contrib.auth import views as auth_views from . import views\",\n \"auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), path('applicantdashboard/', views.applicantdashboard, name='applicantdashboard'), path('recruiterdashboard/', views.recruiterdashboard,\",\n \"urlpatterns = [ path('register/', views.register, name='register'), path('login/', views.userlogin, name='login'), path('logout/',\",\n \"from django.urls import path from django.contrib.auth import views as auth_views\",\n \"path('password_change/', auth_views.PasswordChangeView.as_view(), name='password_change'), path('password_change/done/', auth_views.PasswordChangeDoneView.as_view(), name='password_change_done'), path('password_reset/', auth_views.PasswordResetView.as_view(), name='password_reset'), path('password_reset/done/',\"\n]","isConversational":false}}},{"rowIdx":694,"cells":{"ngram":{"kind":"list like","value":["def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg =","_to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container)","def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container,","/ Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version is","def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path),","def setup_shibboleth_part(local_path, **params): if params is None or len(params) ==","== conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML","len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container,","yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in params and","relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p","-d --remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the","'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to","container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path,","datetime import datetime from difflib import unified_diff from logging import","relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f:","local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path,","= 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version =","return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest =","and 'container_path' in params and params['container'] == container and params['container_path']","return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host,","{dest} from {src} to {host_1}') def restart_container(host, container): cmd =","extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env","x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host,","'-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from","= getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret =","import basicConfig, getLogger, INFO import os from pathlib import Path","ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host,","from pathlib import Path import shutil import subprocess import sys","in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}):","check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml'","generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return","[ 'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for","= Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container, str(conf_path),","and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir =","update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path =","out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest def","f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf =","create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path,","in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}'])","'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2],","def make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy","ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return","src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def","ret = Path(WORKDIR).absolute() / host if version is None: ret","= [ 'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg]","= Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest}","host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is","if version is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /=","**params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg","def get_local_path(host, container, conf_path, version=None): if version is None: version","edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path,","dest.touch() return dest def _to_backup(conf): return conf.parent / (conf.name +","local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path)","= '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger =","f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in ENV_INHERIT])","f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy',","pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path):","= f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host,","f: lines_a = f.readlines() with path_b.open() as f: lines_b =","host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True)","yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p =","conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path)","['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def","ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x","f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if params","sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest","make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path':","container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as","import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR =","'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from","show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container, conf_path, version)","'-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path}","dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args","dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path,","download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return","restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent), version, restart)","'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir /","quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container,","return dest def _to_backup(conf): return conf.parent / (conf.name + '.orig')","(conf.name + '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if","0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path,","['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k,","make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path,","'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc'","= { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'],","= parent / META_YML if not p.exists(): return False with","generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' /","host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return","get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path,","make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars,","lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name))","conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None,","+ '/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host, remote_path)","= download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to)","container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path):","list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f,","find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path,","save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f) params =","def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template}","params and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir","remote_path): p = parent / META_YML if not p.exists(): return","version=None): if version is None: version = find_latest_version(host, container, conf_path)","relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version) backup_path =","path_a.open() as f: lines_a = f.readlines() with path_b.open() as f:","= generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run(","ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}')","= _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org)","str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w')","container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart:","ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env,","'hostname': hostname, }), } vars_path = conf_dir / 'shibboleth.yml' with","template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x,","conf_path): p = parent / META_YML if not p.exists(): return","= (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>')","remote_path) local_path = ( Path(WORKDIR).absolute() / host / version /","local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path,","= {} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth'])","host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out =","yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in params and","subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes'","dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = [ 'ansible',","version=None): local_path = get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path)","for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in","host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg","remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path,","return vars_path def setup_shibboleth_part(local_path, **params): if params is None or","= MOODLE_DIR + '/docker-compose.yml' if version is None: version =","def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart","= yaml.safe_load(f) params = {} if 'shibboleth' in data['services']: params['shibboleth_container']","dict) and 'container' in params and 'container_path' in params and","= get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container,","'-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for","from {local_path} to {host_1}') if not apply: return ansible_arg =","lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container,","with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path,","'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), }","['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out","local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f)","remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f:","== container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p","show_diff(path_a, path_b): lines_a = [] lines_b = [] with path_a.open()","None: version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() /","version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version is","data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml'","conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env","version) with conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path,","version) def _match_metainfo(parent, container, conf_path): p = parent / META_YML","host return sorted([ x.name for x in pdir.glob('*') if x.is_dir()","k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file:","generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host,","conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path =","not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans'","template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x,","params = yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in","extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env =","def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose(","{container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True,","version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data","'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with","{ 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version':","IPython.core.display import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR","conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host,","if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d","and 'remote_path' in params and params['remote_path'] == remote_path) def get_versions(host,","= ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4',","= list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return","container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars)","import yaml from urllib.parse import urlparse from notebook import notebookapp","'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to","{ 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w')","extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host, version=None,","p.exists(): return False with p.open() as f: params = yaml.safe_load(f)","ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x","= list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def","from notebook import notebookapp from IPython.core.display import HTML WORKDIR =","make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path,","/ META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf","None: version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version)","match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([ x.name for","version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out =","relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a',","conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version) backup_path","upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version","host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path","args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path =","conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path","}), } vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as","conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions(","as f: lines_a = f.readlines() with path_b.open() as f: lines_b","for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def","MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN',","= yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in params","if x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path): return","= [] with path_a.open() as f: lines_a = f.readlines() with","WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE","len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def","+ '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf =","['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading","def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container,","def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit'","= ['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for","container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1]","find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host / version","generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible',","subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False):","conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent","with path_a.open() as f: lines_a = f.readlines() with path_b.open() as","show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data =","conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True)","= setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return","extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path","f) return vars_path def setup_shibboleth_part(local_path, **params): if params is None","with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def","local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to)","def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to)","{src} to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose","= get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to)","f: data = yaml.safe_load(f) params = {} if 'shibboleth' in","'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir / 'shibboleth.yml'","def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host if","moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars,","docker-compose up -d --remove-orphans' args = ['ansible', host, '-a', ansible_arg]","'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host,","make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf}","restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path","relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True):","= conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f)","for x in pdir.glob('*') if x.is_dir() and match(x, *args)]) def","_to_backup(conf): return conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False):","conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir","'.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host,","conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path,","make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars])","isinstance(params, dict) and 'container' in params and 'container_path' in params","subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg","check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy'","restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host,","'container' in params and 'container_path' in params and params['container'] ==","generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff =","if not p.exists(): return False with p.open() as f: params","'/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path","save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file):","local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path):","args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env,","'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth',","urllib.parse import urlparse from notebook import notebookapp from IPython.core.display import","params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports':","params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent","local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container,","conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src,","if restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):","conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) with","generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version)","subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR","'-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output(","for x in ENV_INHERIT]) args = ['ansible', host, '-m', 'template',","getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute()","as f: params = yaml.safe_load(f) return ( isinstance(params, dict) and","['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path","get_local_path(host, container, conf_path, version=None): if version is None: version =","flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg])","container, conf_path, version=None): if version is None: version = find_latest_version(host,","pathlib import Path import shutil import subprocess import sys import","path_b.open() as f: lines_b = f.readlines() diff = list(unified_diff( lines_a,","return link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host,","f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes","container, conf_path, version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path),","shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image':","yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path':","and 'container' in params and 'container_path' in params and params['container']","params and params['container'] == container and params['container_path'] == conf_path) def","container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None):","vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params,","download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host,","with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir,","= out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest","as f: lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b,","if conf_path is None: conf_path = Path(remote_path).name dest = generate_local_path(host,","conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False):","host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k, v","f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy',","print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host,","return sorted([ x.name for x in pdir.glob('*') if x.is_dir() and","conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def","'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from","ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible',","conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host,","/ container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path,","volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname':","import os from pathlib import Path import shutil import subprocess","update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host,","shutil import subprocess import sys import yaml from urllib.parse import","list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a","urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, )","hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes':","local_path = ( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml')","= yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w')","make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container,","return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host,","= generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path,","in params and params['container'] == container and params['container_path'] == conf_path)","remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path)","'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path,","container, conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host, container,","conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def","apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container,","return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes)","from {src} to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR}","container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host,","local_path = get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path,","'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k,","link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy',","(local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path,","def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link","if version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path =","docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd])","get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open()","yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if params is","= get_local_path(host, container, conf_path, version) with conf.open() as f: print(f.read())","f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p","'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to","data = yaml.safe_load(f) params = {} if 'shibboleth' in data['services']:","def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2],","make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host,","container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","with p.open() as f: params = yaml.safe_load(f) return ( isinstance(params,","(local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf","version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path =","return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff","subprocess import sys import yaml from urllib.parse import urlparse from","conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) return","{dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src}","lines_a = [] lines_b = [] with path_a.open() as f:","logging import basicConfig, getLogger, INFO import os from pathlib import","with conf_path.open() as f: data = yaml.safe_load(f) params = {}","conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container,","version is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version","host if version is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret","*args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([ x.name","extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host,","datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version ret /= Path(conf_path).name return ret","'-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args,","( isinstance(params, dict) and 'remote_path' in params and params['remote_path'] ==","not p.exists(): return False with p.open() as f: params =","out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg])","in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent /","create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path)","ret /= version ret /= Path(conf_path).name return ret def generate_remote_path(container,","False with p.open() as f: params = yaml.safe_load(f) return (","'.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR',","container, conf_path): p = parent / META_YML if not p.exists():","target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b = []","= create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return","f.readlines() with path_b.open() as f: lines_b = f.readlines() diff =","as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes):","as f: data = yaml.safe_load(f) params = {} if 'shibboleth'","check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg =","p.open() as f: params = yaml.safe_load(f) return ( isinstance(params, dict)","container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True)","f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path =","relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest =","conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible',","= download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params)","return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = []","container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version)","def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf'","'/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE']","org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container':","= out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host,","'-m', 'template', '-c', 'local', '-a', ansible_arg] for k, v in","'container_path' in params and params['container'] == container and params['container_path'] ==","container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path)","conf_path.open() as f: data = yaml.safe_load(f) params = {} if","dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = ['ansible', host,","local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url,","f: params = yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path'","= _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host,","create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container,","= ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s')","return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path =","host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if","vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname,","cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host,","def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container,","{host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up","args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path","relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host,","remote_path, conf_path=None): if conf_path is None: conf_path = Path(remote_path).name dest","conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path, 'remote_path':","logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host, container): cmd","ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host,","['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args,","= find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host,","Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent))","META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT","def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return","for x in ENV_INHERIT]) args = [ 'ansible', host, '-m',","'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent","not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container,","version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link","in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={},","pdir = Path(WORKDIR).absolute() / host return sorted([ x.name for x","= f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible', host,","generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template}","{remote_path} from {local_path} to {host_1}') if not apply: return ansible_arg","get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path,","conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False): org =","show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent","return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container,","logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path,","conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent","'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR","difflib import unified_diff from logging import basicConfig, getLogger, INFO import","subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0]","return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container","/ conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a","is None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg","fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path,","host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path}","exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent / (conf.name","/ 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output(","= f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m',","restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container,","**params): if params is None or len(params) == 0: return","= dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = [","relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path,","version=None): conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def","is None: version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path,","conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):","href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b =","import datetime from difflib import unified_diff from logging import basicConfig,","= f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff)","Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path}","def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host,","default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version':","append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host,","from urllib.parse import urlparse from notebook import notebookapp from IPython.core.display","generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return","the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars):","def download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path =","conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest","if params is None or len(params) == 0: return save_shibboleth_part(local_path)","backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent /","download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path)","extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False):","conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/'","def show_diff(path_a, path_b): lines_a = [] lines_b = [] with","[] lines_b = [] with path_a.open() as f: lines_a =","import urlparse from notebook import notebookapp from IPython.core.display import HTML","remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is None:","list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src,","remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None):","version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def","conf_path, version) def _match_metainfo(parent, container, conf_path): p = parent /","container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return","version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version)","container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], }","from difflib import unified_diff from logging import basicConfig, getLogger, INFO","*args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def","['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0]","== remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() /","return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def","= find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host /","= Path(WORKDIR).absolute() / host return sorted([ x.name for x in","relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host,","generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host if version","import Path import shutil import subprocess import sys import yaml","f: params = yaml.safe_load(f) return ( isinstance(params, dict) and 'container'","= f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m',","else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template","host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name +","( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg =","Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version is None:","local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host,","return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template =","conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return","yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as","setup_shibboleth_part(local_path, **params): if params is None or len(params) == 0:","--remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes","x in ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template',","f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True)","setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path)","tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE):","is None or len(params) == 0: return save_shibboleth_part(local_path) else: return","'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname,","logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret","conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p","params = {} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump(","/ host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path}","version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent), version,","= Path(WORKDIR).absolute() / host if version is None: ret /=","= find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent,","restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path =","+ '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars","changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template","in params and 'container_path' in params and params['container'] == container","conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to))","and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent","in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template =","basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() /","vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params","container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path)","'-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path}","to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart","version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container,","in ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template', '-c',","env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR +","return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p =","def _to_backup(conf): return conf.parent / (conf.name + '.orig') def make_backup(conf,","if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path,","conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if","conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host, container, conf_path,","x in pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host,","as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf","unified_diff from logging import basicConfig, getLogger, INFO import os from","if version is None: version = find_latest_version(host, container, conf_path) return","def _match_metainfo(parent, container, conf_path): p = parent / META_YML if","def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if","params = { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path,","def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container,","get_local_path(host, container, conf_path, version) with conf.open() as f: print(f.read()) def","diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff)","_to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def","return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if","restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def","container, conf_path, version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path,","conf_path, version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host,","conf_path, version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path)","ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template', '-c', 'local',","container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}')","vars_path def setup_shibboleth_part(local_path, **params): if params is None or len(params)","from datetime import datetime from difflib import unified_diff from logging","in params and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo):","generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path):","remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path),","conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def","match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path","show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container, conf_path, version)","'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for","conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version)","with conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","_match_metainfo(parent, container, conf_path): p = parent / META_YML if not","upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to)","f: lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name,","notebook import notebookapp from IPython.core.display import HTML WORKDIR = 'edit'","/= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR)","CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger","update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link =","version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf')","= '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH',","'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def","extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host,","conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path}","'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as","'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def","generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output(",") def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml'","apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent),","MOODLE_DIR + '/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host,","out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1","remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with","= [] lines_b = [] with path_a.open() as f: lines_a","init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = {","for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True)","version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path = (","notebookapp from IPython.core.display import HTML WORKDIR = 'edit' META_YML =","= f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in","find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path,","logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not apply: return","Path(WORKDIR).absolute() / host if version is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\")","None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg =","Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes'","import shutil import subprocess import sys import yaml from urllib.parse","container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None):","org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf,","conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return","import unified_diff from logging import basicConfig, getLogger, INFO import os","link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container,","import notebookapp from IPython.core.display import HTML WORKDIR = 'edit' META_YML","'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version","'-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest}","params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute()","container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg =","generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container,","return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host,","{conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params","hostname, }), } vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w')","env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args =","get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path","'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname':","'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path =","= { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)),","generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' /","vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params,","apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version is None:","fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path","in ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c', 'local',","env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container =","generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def","[] with path_a.open() as f: lines_a = f.readlines() with path_b.open()","conf_path is None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path)","= ( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg","f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible', host, '-a',","= fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path))","as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params =","is None: version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute()","backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to)","/ host return sorted([ x.name for x in pdir.glob('*') if","src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b',","container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container, conf_path,","lines_a = f.readlines() with path_b.open() as f: lines_b = f.readlines()","container = 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version","def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch()","show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path,","+ '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if not","default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) /","stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path,","version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path =","f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = {","f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in ENV_INHERIT])","relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container,","version ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE):","logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest def download_conf_file(host,","= ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.')","get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([","src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b',","if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path =","up -d --remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply","hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container':","fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path,","relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path","container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to)","{remote_path} from {host_1} to {dest}') return dest def download_conf_file(host, container,","dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a',","= f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m',","version is None: version = find_latest_version(host, container, conf_path) return generate_local_path(host,","container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path,","p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p}","volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({","'/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars =","return False with p.open() as f: params = yaml.safe_load(f) return","'template', '-c', 'local', '-a', ansible_arg] for k, v in extra_vars.items():","'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host,","x in ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c',","getLogger, INFO import os from pathlib import Path import shutil","Path(WORKDIR).absolute() / host return sorted([ x.name for x in pdir.glob('*')","local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname =","conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest","-p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest}","src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True,","conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return","( isinstance(params, dict) and 'container' in params and 'container_path' in","args = ['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg]","'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes,","{host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}'","lines_b = [] with path_a.open() as f: lines_a = f.readlines()","/ 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path","os.environ[x]) for x in ENV_INHERIT]) args = ['ansible', host, '-m',","remote_path = MOODLE_DIR + '/docker-compose.yml' if version is None: version","def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if not","return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc","'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None):","params['container'] == container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path):","dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent /","str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host,","/ META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path):","host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading","= f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in","match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1]","format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host","conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host,","get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path,","download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path = Path(remote_path).name","return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host,","ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg])","/= version ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path,","f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a',","yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name':","['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged':","return ( isinstance(params, dict) and 'container' in params and 'container_path'","out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host, container):","'remote_path' in params and params['remote_path'] == remote_path) def get_versions(host, *args,","quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}')","basicConfig, getLogger, INFO import os from pathlib import Path import","= { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent /","(local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf):","/ 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path","container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path =","urlparse from notebook import notebookapp from IPython.core.display import HTML WORKDIR","{host_1} to {dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE):","import subprocess import sys import yaml from urllib.parse import urlparse","host, container, conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host,","META_YML if not p.exists(): return False with p.open() as f:","relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container,","**shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def","def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container, conf_path,","dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def","version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host, container)","in pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host, container,","data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f:","extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env =","extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[],","sorted([ x.name for x in pdir.glob('*') if x.is_dir() and match(x,","relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) with conf.open()","True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }),","volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR +","{local_path} to {host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle","out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not apply:","hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path =","ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host,","= generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host,","_match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if not p.exists():","= generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf):","ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c', 'local', '-a',","parent / META_YML if not p.exists(): return False with p.open()","conf_path, version=None): if version is None: version = find_latest_version(host, container,","'-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def","x.name for x in pdir.glob('*') if x.is_dir() and match(x, *args)])","def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params =","sys import yaml from urllib.parse import urlparse from notebook import","f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return","[shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}):","= 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x])","params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent /","} vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f:","= yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in params","relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def","version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path,","datetime from difflib import unified_diff from logging import basicConfig, getLogger,","nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir']))","Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) /","Path import shutil import subprocess import sys import yaml from","dict) and 'remote_path' in params and params['remote_path'] == remote_path) def","conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def","/ (conf.name + '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf)","else: ret /= version ret /= Path(conf_path).name return ret def","p = parent / META_YML if not p.exists(): return False","/= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version ret /= Path(conf_path).name return","relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path)","ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container /","META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf =","HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b","logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container,","def save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f) params","from {host_1} to {dest}') return dest def download_conf_file(host, container, conf_path,","} with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False)","stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url'])","relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path","quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path,","f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup']","params is None or len(params) == 0: return save_shibboleth_part(local_path) else:","'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for","{} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path","return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars,","docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template'","conf_path, version=None): ret = Path(WORKDIR).absolute() / host if version is","version=None): conf = get_local_path(host, container, conf_path, version) with conf.open() as","create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return","(Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container,","'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1}","from IPython.core.display import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml'","{org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params =","as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0]","relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf)","container, conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path","conf_path=None): if conf_path is None: conf_path = Path(remote_path).name dest =","yaml.safe_load(f) params = {} if 'shibboleth' in data['services']: params['shibboleth_container'] =","os from pathlib import Path import shutil import subprocess import","version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def","logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE):","ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version ret /= Path(conf_path).name","generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host,","from logging import basicConfig, getLogger, INFO import os from pathlib","path_b): lines_a = [] lines_b = [] with path_a.open() as","str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host,","show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname","{container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host,","os.environ[x]) for x in ENV_INHERIT]) args = [ 'ansible', host,","backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a',","v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host,","= subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 =","f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch',","dest def _to_backup(conf): return conf.parent / (conf.name + '.orig') def","'/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__)","(Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def","to {host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose","str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path,","container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p =","extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args,","find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container,","return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args =","upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name","def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host,","def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container,","'-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path =","host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None,","= 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x])","subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 =","and params['container'] == container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent,","container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert'))","relative_to) if restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE,","subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container","= get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path)","ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}')","local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path,","= f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg","revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path,","dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}'","== 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host,","= dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = ['ansible',","and match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container,","ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO,","extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}):","x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host,","import sys import yaml from urllib.parse import urlparse from notebook","generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf'","find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path):","None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version ret /=","remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host","'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible',","conf = get_local_path(host, container, conf_path, version) with conf.open() as f:","isinstance(params, dict) and 'remote_path' in params and params['remote_path'] == remote_path)","f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg =","def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path","cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible',","extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname,","def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path)","as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if","k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def","shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path","yaml from urllib.parse import urlparse from notebook import notebookapp from","diff = show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container,","container, conf_path, version) with conf.open() as f: print(f.read()) def edit_local_conf(host,","HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle'","= generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out =","yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes =","/ META_YML if not p.exists(): return False with p.open() as","with path_b.open() as f: lines_b = f.readlines() diff = list(unified_diff(","ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible',","ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host,","args = [ 'ansible', host, '-m', 'template', '-c', 'local', '-a',","or len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params)","/ version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out","container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path,","/ 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if","def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host,","= MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path,","dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args","= out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not","ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x","f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf')","/ 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a,","= get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with","generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p = parent","= f.readlines() with path_b.open() as f: lines_b = f.readlines() diff","None or len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent,","show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True):","apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args","args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in","shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir","params and 'container_path' in params and params['container'] == container and","params = yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in","return conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False): org","'.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet:","ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}')","/ (local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):","init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml'","host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def","return ( isinstance(params, dict) and 'remote_path' in params and params['remote_path']","make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], }","= subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1","'local', '-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}'])","to {dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src","{dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src =","= show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container, conf_path,","conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as","= conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f)","= urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes},","'-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\"\\n\")[0].split()[0] logger.info(f'Uploading {dest} from {src}","container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version)","restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container","conf_path, version) with conf.open() as f: print(f.read()) def edit_local_conf(host, container,","is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else: ret /= version ret","'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b):","create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path,","= 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE =","host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes'","upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host, container) def","{ 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes':","INFO import os from pathlib import Path import shutil import","'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params,","v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e',","/ host if version is None: ret /= datetime.now().strftime(\"%Y%m%d%H%M%S%f\") else:","conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container,","params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with","shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path)","container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host,","vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params):","get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if","dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out","version=None): ret = Path(WORKDIR).absolute() / host if version is None:","lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host,","conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f:","'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent /","META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params","'-c', 'local', '-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e',","version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host","conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\"_blank\">{p.name}</a>') def show_diff(path_a, path_b): lines_a =","def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path,","shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = {","download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return","container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host,","= '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT ="],"string":"[\n \"def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg =\",\n \"_to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container)\",\n \"def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container,\",\n \"/ Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version is\",\n \"def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path),\",\n \"def setup_shibboleth_part(local_path, **params): if params is None or len(params) ==\",\n \"== conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML\",\n \"len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container,\",\n \"yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in params and\",\n \"relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p\",\n \"-d --remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the\",\n \"'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to\",\n \"container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path,\",\n \"datetime import datetime from difflib import unified_diff from logging import\",\n \"relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f:\",\n \"local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path,\",\n \"= 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version =\",\n \"return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest =\",\n \"and 'container_path' in params and params['container'] == container and params['container_path']\",\n \"return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host,\",\n \"{dest} from {src} to {host_1}') def restart_container(host, container): cmd =\",\n \"extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env\",\n \"x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host,\",\n \"'-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from\",\n \"= getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret =\",\n \"import basicConfig, getLogger, INFO import os from pathlib import Path\",\n \"ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host,\",\n \"from pathlib import Path import shutil import subprocess import sys\",\n \"in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}):\",\n \"check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml'\",\n \"generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return\",\n \"[ 'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for\",\n \"= Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container, str(conf_path),\",\n \"and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir =\",\n \"update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path =\",\n \"out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest def\",\n \"f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf =\",\n \"create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path,\",\n \"in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}'])\",\n \"'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2],\",\n \"def make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy\",\n \"ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return\",\n \"src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def\",\n \"ret = Path(WORKDIR).absolute() / host if version is None: ret\",\n \"= [ 'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg]\",\n \"= Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest}\",\n \"host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is\",\n \"if version is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /=\",\n \"**params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg\",\n \"def get_local_path(host, container, conf_path, version=None): if version is None: version\",\n \"edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path,\",\n \"dest.touch() return dest def _to_backup(conf): return conf.parent / (conf.name +\",\n \"local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path)\",\n \"= '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger =\",\n \"f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in ENV_INHERIT])\",\n \"f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy',\",\n \"pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path):\",\n \"= f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host,\",\n \"f: lines_a = f.readlines() with path_b.open() as f: lines_b =\",\n \"host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True)\",\n \"yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p =\",\n \"conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path)\",\n \"['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def\",\n \"ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x\",\n \"f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if params\",\n \"sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest\",\n \"make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path':\",\n \"container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as\",\n \"import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR =\",\n \"'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from\",\n \"show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container, conf_path, version)\",\n \"'-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path}\",\n \"dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args\",\n \"dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path,\",\n \"download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return\",\n \"restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent), version, restart)\",\n \"'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir /\",\n \"quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container,\",\n \"return dest def _to_backup(conf): return conf.parent / (conf.name + '.orig')\",\n \"(conf.name + '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if\",\n \"0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path,\",\n \"['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k,\",\n \"make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path,\",\n \"'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc'\",\n \"= { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'],\",\n \"= parent / META_YML if not p.exists(): return False with\",\n \"generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' /\",\n \"host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return\",\n \"get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path,\",\n \"make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars,\",\n \"lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name))\",\n \"conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None,\",\n \"+ '/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host, remote_path)\",\n \"= download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to)\",\n \"container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path):\",\n \"list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f,\",\n \"find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path,\",\n \"save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f) params =\",\n \"def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template}\",\n \"params and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir\",\n \"remote_path): p = parent / META_YML if not p.exists(): return\",\n \"version=None): if version is None: version = find_latest_version(host, container, conf_path)\",\n \"relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version) backup_path =\",\n \"path_a.open() as f: lines_a = f.readlines() with path_b.open() as f:\",\n \"= generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run(\",\n \"ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}')\",\n \"= _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org)\",\n \"str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w')\",\n \"container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart:\",\n \"ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env,\",\n \"'hostname': hostname, }), } vars_path = conf_dir / 'shibboleth.yml' with\",\n \"template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x,\",\n \"conf_path): p = parent / META_YML if not p.exists(): return\",\n \"= (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>')\",\n \"remote_path) local_path = ( Path(WORKDIR).absolute() / host / version /\",\n \"local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path,\",\n \"= {} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth'])\",\n \"host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out =\",\n \"yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in params and\",\n \"subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes'\",\n \"dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = [ 'ansible',\",\n \"version=None): local_path = get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path)\",\n \"for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in\",\n \"host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg\",\n \"remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path,\",\n \"return vars_path def setup_shibboleth_part(local_path, **params): if params is None or\",\n \"= MOODLE_DIR + '/docker-compose.yml' if version is None: version =\",\n \"def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart\",\n \"= yaml.safe_load(f) params = {} if 'shibboleth' in data['services']: params['shibboleth_container']\",\n \"dict) and 'container' in params and 'container_path' in params and\",\n \"= get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container,\",\n \"'-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for\",\n \"from {local_path} to {host_1}') if not apply: return ansible_arg =\",\n \"lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container,\",\n \"with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path,\",\n \"'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), }\",\n \"['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out\",\n \"local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f)\",\n \"remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f:\",\n \"== container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p\",\n \"show_diff(path_a, path_b): lines_a = [] lines_b = [] with path_a.open()\",\n \"None: version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() /\",\n \"version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version is\",\n \"data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml'\",\n \"conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env\",\n \"version) with conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path,\",\n \"version) def _match_metainfo(parent, container, conf_path): p = parent / META_YML\",\n \"host return sorted([ x.name for x in pdir.glob('*') if x.is_dir()\",\n \"k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file:\",\n \"generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host,\",\n \"conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path =\",\n \"not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans'\",\n \"template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x,\",\n \"params = yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in\",\n \"extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env =\",\n \"def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose(\",\n \"{container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True,\",\n \"version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data\",\n \"'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with\",\n \"{ 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version':\",\n \"IPython.core.display import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR\",\n \"conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host,\",\n \"if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d\",\n \"and 'remote_path' in params and params['remote_path'] == remote_path) def get_versions(host,\",\n \"= ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4',\",\n \"= list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return\",\n \"container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars)\",\n \"import yaml from urllib.parse import urlparse from notebook import notebookapp\",\n \"'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to\",\n \"{ 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w')\",\n \"extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host, version=None,\",\n \"p.exists(): return False with p.open() as f: params = yaml.safe_load(f)\",\n \"ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x\",\n \"= list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def\",\n \"from notebook import notebookapp from IPython.core.display import HTML WORKDIR =\",\n \"make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path,\",\n \"/ META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf\",\n \"None: version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version)\",\n \"match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([ x.name for\",\n \"version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out =\",\n \"relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a',\",\n \"conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version) backup_path\",\n \"upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version\",\n \"host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path\",\n \"args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path =\",\n \"conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path\",\n \"}), } vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as\",\n \"conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions(\",\n \"as f: lines_a = f.readlines() with path_b.open() as f: lines_b\",\n \"for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def\",\n \"MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN',\",\n \"= yaml.safe_load(f) return ( isinstance(params, dict) and 'container' in params\",\n \"if x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path): return\",\n \"= [] with path_a.open() as f: lines_a = f.readlines() with\",\n \"WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE\",\n \"len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def\",\n \"+ '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf =\",\n \"['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading\",\n \"def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container,\",\n \"def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit'\",\n \"= ['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for\",\n \"container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1]\",\n \"find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host / version\",\n \"generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible',\",\n \"subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False):\",\n \"conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent\",\n \"with path_a.open() as f: lines_a = f.readlines() with path_b.open() as\",\n \"show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f: data =\",\n \"conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True)\",\n \"= setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return\",\n \"extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path\",\n \"f) return vars_path def setup_shibboleth_part(local_path, **params): if params is None\",\n \"with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def\",\n \"local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to)\",\n \"def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to)\",\n \"{src} to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose\",\n \"= get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to)\",\n \"f: data = yaml.safe_load(f) params = {} if 'shibboleth' in\",\n \"'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir / 'shibboleth.yml'\",\n \"def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host if\",\n \"moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars,\",\n \"docker-compose up -d --remove-orphans' args = ['ansible', host, '-a', ansible_arg]\",\n \"'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host,\",\n \"make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf}\",\n \"restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path\",\n \"relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True):\",\n \"= conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f)\",\n \"for x in pdir.glob('*') if x.is_dir() and match(x, *args)]) def\",\n \"_to_backup(conf): return conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False):\",\n \"conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir\",\n \"'.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host,\",\n \"conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path,\",\n \"make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars])\",\n \"isinstance(params, dict) and 'container' in params and 'container_path' in params\",\n \"subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg\",\n \"check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy'\",\n \"restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host,\",\n \"'container' in params and 'container_path' in params and params['container'] ==\",\n \"generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff =\",\n \"if not p.exists(): return False with p.open() as f: params\",\n \"'/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path\",\n \"save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file):\",\n \"local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path):\",\n \"args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env,\",\n \"'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth',\",\n \"urllib.parse import urlparse from notebook import notebookapp from IPython.core.display import\",\n \"params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports':\",\n \"params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent\",\n \"local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container,\",\n \"conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src,\",\n \"if restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):\",\n \"conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) with\",\n \"generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version)\",\n \"subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR\",\n \"'-a', ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output(\",\n \"for x in ENV_INHERIT]) args = ['ansible', host, '-m', 'template',\",\n \"getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute()\",\n \"as f: params = yaml.safe_load(f) return ( isinstance(params, dict) and\",\n \"['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path\",\n \"get_local_path(host, container, conf_path, version=None): if version is None: version =\",\n \"flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg])\",\n \"container, conf_path, version=None): if version is None: version = find_latest_version(host,\",\n \"pathlib import Path import shutil import subprocess import sys import\",\n \"path_b.open() as f: lines_b = f.readlines() diff = list(unified_diff( lines_a,\",\n \"return link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host,\",\n \"f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes\",\n \"container, conf_path, version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path),\",\n \"shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image':\",\n \"yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path':\",\n \"and 'container' in params and 'container_path' in params and params['container']\",\n \"params and params['container'] == container and params['container_path'] == conf_path) def\",\n \"container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None):\",\n \"vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params,\",\n \"download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host,\",\n \"with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir,\",\n \"= out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest\",\n \"as f: lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b,\",\n \"if conf_path is None: conf_path = Path(remote_path).name dest = generate_local_path(host,\",\n \"conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False):\",\n \"host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k, v\",\n \"f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy',\",\n \"print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host,\",\n \"return sorted([ x.name for x in pdir.glob('*') if x.is_dir() and\",\n \"conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def\",\n \"'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from\",\n \"ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible',\",\n \"conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host,\",\n \"/ container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path,\",\n \"volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname':\",\n \"import os from pathlib import Path import shutil import subprocess\",\n \"update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host,\",\n \"shutil import subprocess import sys import yaml from urllib.parse import\",\n \"list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a\",\n \"urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, )\",\n \"hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes':\",\n \"local_path = ( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml')\",\n \"= yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w')\",\n \"make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container,\",\n \"return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host,\",\n \"= generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path,\",\n \"in params and params['container'] == container and params['container_path'] == conf_path)\",\n \"remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path)\",\n \"'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path,\",\n \"container, conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host, container,\",\n \"conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def\",\n \"apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container,\",\n \"return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes)\",\n \"from {src} to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR}\",\n \"container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host,\",\n \"local_path = get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path,\",\n \"'ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg] for k,\",\n \"link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy',\",\n \"(local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path,\",\n \"def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link\",\n \"if version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path =\",\n \"docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd])\",\n \"get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open()\",\n \"yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if params is\",\n \"= get_local_path(host, container, conf_path, version) with conf.open() as f: print(f.read())\",\n \"f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p\",\n \"'-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to\",\n \"data = yaml.safe_load(f) params = {} if 'shibboleth' in data['services']:\",\n \"def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2],\",\n \"make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host,\",\n \"container, conf_path, relative_to) return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"with p.open() as f: params = yaml.safe_load(f) return ( isinstance(params,\",\n \"(local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf\",\n \"version) return generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path =\",\n \"return generate_edit_link(local_path) def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff\",\n \"subprocess import sys import yaml from urllib.parse import urlparse from\",\n \"conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) return\",\n \"{dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest} src={src}\",\n \"lines_a = [] lines_b = [] with path_a.open() as f:\",\n \"logging import basicConfig, getLogger, INFO import os from pathlib import\",\n \"with conf_path.open() as f: data = yaml.safe_load(f) params = {}\",\n \"conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container,\",\n \"version is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version\",\n \"host if version is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret\",\n \"*args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([ x.name\",\n \"extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host,\",\n \"datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version ret /= Path(conf_path).name return ret\",\n \"'-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args,\",\n \"( isinstance(params, dict) and 'remote_path' in params and params['remote_path'] ==\",\n \"not p.exists(): return False with p.open() as f: params =\",\n \"out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg])\",\n \"in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent /\",\n \"create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path)\",\n \"ret /= version ret /= Path(conf_path).name return ret def generate_remote_path(container,\",\n \"False with p.open() as f: params = yaml.safe_load(f) return (\",\n \"'.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR',\",\n \"container, conf_path): p = parent / META_YML if not p.exists():\",\n \"target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b = []\",\n \"= create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return\",\n \"f.readlines() with path_b.open() as f: lines_b = f.readlines() diff =\",\n \"as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes):\",\n \"as f: data = yaml.safe_load(f) params = {} if 'shibboleth'\",\n \"check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg =\",\n \"p.open() as f: params = yaml.safe_load(f) return ( isinstance(params, dict)\",\n \"container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True)\",\n \"f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path =\",\n \"relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest =\",\n \"conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible',\",\n \"= download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params)\",\n \"return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = []\",\n \"container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version)\",\n \"def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf'\",\n \"'/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE']\",\n \"org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = { 'container':\",\n \"= out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host,\",\n \"'-m', 'template', '-c', 'local', '-a', ansible_arg] for k, v in\",\n \"'container_path' in params and params['container'] == container and params['container_path'] ==\",\n \"container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path)\",\n \"conf_path.open() as f: data = yaml.safe_load(f) params = {} if\",\n \"dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = ['ansible', host,\",\n \"local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url,\",\n \"f: params = yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path'\",\n \"= _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host,\",\n \"create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container,\",\n \"= ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s')\",\n \"return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path =\",\n \"host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if\",\n \"vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname,\",\n \"cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host,\",\n \"def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container,\",\n \"{host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up\",\n \"args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path\",\n \"relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host,\",\n \"remote_path, conf_path=None): if conf_path is None: conf_path = Path(remote_path).name dest\",\n \"conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path, 'remote_path':\",\n \"logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host, container): cmd\",\n \"ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host,\",\n \"['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args,\",\n \"= find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host,\",\n \"Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent))\",\n \"META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT\",\n \"def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return\",\n \"for x in ENV_INHERIT]) args = [ 'ansible', host, '-m',\",\n \"'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent\",\n \"not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container,\",\n \"version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link\",\n \"in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={},\",\n \"pdir = Path(WORKDIR).absolute() / host return sorted([ x.name for x\",\n \"= f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible', host,\",\n \"generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template}\",\n \"{remote_path} from {local_path} to {host_1}') if not apply: return ansible_arg\",\n \"get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path,\",\n \"conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False): org =\",\n \"show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent\",\n \"return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container,\",\n \"logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path,\",\n \"conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent\",\n \"'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR\",\n \"difflib import unified_diff from logging import basicConfig, getLogger, INFO import\",\n \"subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0]\",\n \"return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container\",\n \"/ conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a\",\n \"is None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg\",\n \"fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path,\",\n \"host, '-m', 'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path}\",\n \"exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent / (conf.name\",\n \"/ 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output(\",\n \"= f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m',\",\n \"restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container,\",\n \"**params): if params is None or len(params) == 0: return\",\n \"= dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = [\",\n \"relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path,\",\n \"version=None): conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def\",\n \"is None: version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path,\",\n \"conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):\",\n \"href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b =\",\n \"import datetime from difflib import unified_diff from logging import basicConfig,\",\n \"= f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff)\",\n \"Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path}\",\n \"def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host,\",\n \"default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version':\",\n \"append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host,\",\n \"from urllib.parse import urlparse from notebook import notebookapp from IPython.core.display\",\n \"generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return\",\n \"the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars):\",\n \"def download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path =\",\n \"conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest\",\n \"if params is None or len(params) == 0: return save_shibboleth_part(local_path)\",\n \"backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent /\",\n \"download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path)\",\n \"extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False):\",\n \"conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/'\",\n \"def show_diff(path_a, path_b): lines_a = [] lines_b = [] with\",\n \"[] lines_b = [] with path_a.open() as f: lines_a =\",\n \"import urlparse from notebook import notebookapp from IPython.core.display import HTML\",\n \"remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is None:\",\n \"list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src,\",\n \"remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None):\",\n \"version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def\",\n \"conf_path, version) def _match_metainfo(parent, container, conf_path): p = parent /\",\n \"container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return\",\n \"version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version)\",\n \"container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], }\",\n \"from difflib import unified_diff from logging import basicConfig, getLogger, INFO\",\n \"*args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def\",\n \"['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0]\",\n \"== remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() /\",\n \"return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def\",\n \"= find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host /\",\n \"= Path(WORKDIR).absolute() / host return sorted([ x.name for x in\",\n \"relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host,\",\n \"generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host if version\",\n \"import Path import shutil import subprocess import sys import yaml\",\n \"f: params = yaml.safe_load(f) return ( isinstance(params, dict) and 'container'\",\n \"= f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m',\",\n \"else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template\",\n \"host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name +\",\n \"( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg =\",\n \"Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version is None:\",\n \"local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host,\",\n \"return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template =\",\n \"conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return\",\n \"yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as\",\n \"setup_shibboleth_part(local_path, **params): if params is None or len(params) == 0:\",\n \"--remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes\",\n \"x in ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template',\",\n \"f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True)\",\n \"setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path)\",\n \"tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE):\",\n \"is None or len(params) == 0: return save_shibboleth_part(local_path) else: return\",\n \"'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname,\",\n \"logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret\",\n \"conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p\",\n \"params = {} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump(\",\n \"/ host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path}\",\n \"version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent), version,\",\n \"= Path(WORKDIR).absolute() / host if version is None: ret /=\",\n \"= find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent,\",\n \"restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path =\",\n \"+ '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars\",\n \"changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template\",\n \"in params and 'container_path' in params and params['container'] == container\",\n \"conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to))\",\n \"and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent\",\n \"in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template =\",\n \"basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() /\",\n \"vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params\",\n \"container, conf_path, relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path)\",\n \"'-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path}\",\n \"to {host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart\",\n \"version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container,\",\n \"in ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template', '-c',\",\n \"env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR +\",\n \"return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p =\",\n \"def _to_backup(conf): return conf.parent / (conf.name + '.orig') def make_backup(conf,\",\n \"if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path,\",\n \"conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if\",\n \"conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host, container, conf_path,\",\n \"x in pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host,\",\n \"as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf\",\n \"unified_diff from logging import basicConfig, getLogger, INFO import os from\",\n \"if version is None: version = find_latest_version(host, container, conf_path) return\",\n \"def _match_metainfo(parent, container, conf_path): p = parent / META_YML if\",\n \"def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if\",\n \"params = { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path,\",\n \"def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container,\",\n \"get_local_path(host, container, conf_path, version) with conf.open() as f: print(f.read()) def\",\n \"diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff)\",\n \"_to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def\",\n \"return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if\",\n \"restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def\",\n \"container, conf_path, version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path,\",\n \"conf_path, version) local_path = get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host,\",\n \"conf_path, version=None): local_path = get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path)\",\n \"ENV_INHERIT]) args = [ 'ansible', host, '-m', 'template', '-c', 'local',\",\n \"container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}')\",\n \"vars_path def setup_shibboleth_part(local_path, **params): if params is None or len(params)\",\n \"from datetime import datetime from difflib import unified_diff from logging\",\n \"in params and params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo):\",\n \"generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host, conf_path):\",\n \"remote_path) shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path),\",\n \"conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path, relative_to) return generate_edit_link(local_path) def\",\n \"match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path\",\n \"show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container, conf_path, version)\",\n \"'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for\",\n \"conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version)\",\n \"with conf.open() as f: print(f.read()) def edit_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"_match_metainfo(parent, container, conf_path): p = parent / META_YML if not\",\n \"upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to)\",\n \"f: lines_b = f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name,\",\n \"notebook import notebookapp from IPython.core.display import HTML WORKDIR = 'edit'\",\n \"/= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR)\",\n \"CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger\",\n \"update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container = 'proxy' link =\",\n \"version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf')\",\n \"= '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH',\",\n \"'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def\",\n \"extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host,\",\n \"conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path}\",\n \"'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as\",\n \"'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def\",\n \"generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output(\",\n \") def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml'\",\n \"apply_proxy_conf(host, version=None, restart=True): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') apply_conf(host, 'proxy', str(conf_path), str(conf_path.parent),\",\n \"MOODLE_DIR + '/docker-compose.yml' if version is None: version = find_latest_version_by_remote_path(host,\",\n \"out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1\",\n \"remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with\",\n \"= [] lines_b = [] with path_a.open() as f: lines_a\",\n \"init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = {\",\n \"for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True)\",\n \"version is None: version = find_latest_version_by_remote_path(host, remote_path) local_path = (\",\n \"notebookapp from IPython.core.display import HTML WORKDIR = 'edit' META_YML =\",\n \"= f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in\",\n \"find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path,\",\n \"logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not apply: return\",\n \"Path(WORKDIR).absolute() / host if version is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\")\",\n \"None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg =\",\n \"Path(remote_path).name dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes'\",\n \"import shutil import subprocess import sys import yaml from urllib.parse\",\n \"container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None):\",\n \"org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf,\",\n \"conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return\",\n \"import unified_diff from logging import basicConfig, getLogger, INFO import os\",\n \"link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container,\",\n \"import notebookapp from IPython.core.display import HTML WORKDIR = 'edit' META_YML\",\n \"'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if version\",\n \"'-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest}\",\n \"params['remote_path'] == remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute()\",\n \"container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path, relative_to) ansible_arg =\",\n \"generate_edit_link(conf) def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container,\",\n \"return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host,\",\n \"{conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params\",\n \"hostname, }), } vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w')\",\n \"env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args =\",\n \"get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def download_file(host, remote_path, conf_path=None): if conf_path\",\n \"'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname':\",\n \"'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path =\",\n \"= { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container, conf_path, relative_to)),\",\n \"generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' /\",\n \"vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params,\",\n \"apply=False): remote_path = MOODLE_DIR + '/docker-compose.yml' if version is None:\",\n \"fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path\",\n \"in ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c', 'local',\",\n \"env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container =\",\n \"generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host, container, conf_path, version) return link def\",\n \"[] with path_a.open() as f: lines_a = f.readlines() with path_b.open()\",\n \"conf_path is None: conf_path = Path(remote_path).name dest = generate_local_path(host, conf_path)\",\n \"= ( Path(WORKDIR).absolute() / host / version / 'docker-compose.yml') ansible_arg\",\n \"f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args = ['ansible', host, '-a',\",\n \"= fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path))\",\n \"as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params =\",\n \"is None: version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute()\",\n \"backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to)\",\n \"/ host return sorted([ x.name for x in pdir.glob('*') if\",\n \"src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b',\",\n \"container) local_path.rename(local_path.parent / (local_path.name + '.revert')) def show_local_conf(host, container, conf_path,\",\n \"lines_a = f.readlines() with path_b.open() as f: lines_b = f.readlines()\",\n \"container = 'proxy' link = fetch_conf(host, container, str(conf_path), str(conf_path.parent)) version\",\n \"def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch()\",\n \"show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path,\",\n \"+ '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if not\",\n \"default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) /\",\n \"stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path,\",\n \"version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path =\",\n \"f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params = {\",\n \"f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for x in ENV_INHERIT])\",\n \"relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path, 'remote_path': str(generate_remote_path(container,\",\n \"version ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE):\",\n \"logger.info(f'Downloading {remote_path} from {host_1} to {dest}') return dest def download_conf_file(host,\",\n \"= ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.')\",\n \"get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host return sorted([\",\n \"src={src} backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b',\",\n \"if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path =\",\n \"up -d --remove-orphans' args = ['ansible', host, '-a', ansible_arg] logger.info('Apply\",\n \"hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container':\",\n \"fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path,\",\n \"relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path, version) local_path\",\n \"container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container, conf_path, relative_to)\",\n \"{remote_path} from {host_1} to {dest}') return dest def download_conf_file(host, container,\",\n \"dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a',\",\n \"= f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible', host, '-m',\",\n \"version is None: version = find_latest_version(host, container, conf_path) return generate_local_path(host,\",\n \"container, conf_path, relative_to=CONF_RELATIVE): params = { 'container': container, 'container_path': conf_path,\",\n \"p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p}\",\n \"volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({\",\n \"'/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path, remote_path) shibboleth_vars =\",\n \"return False with p.open() as f: params = yaml.safe_load(f) return\",\n \"'template', '-c', 'local', '-a', ansible_arg] for k, v in extra_vars.items():\",\n \"'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host,\",\n \"x in ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c',\",\n \"getLogger, INFO import os from pathlib import Path import shutil\",\n \"Path(WORKDIR).absolute() / host return sorted([ x.name for x in pdir.glob('*')\",\n \"local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname =\",\n \"conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest\",\n \"-p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg = f'dest={dest}\",\n \"src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True,\",\n \"conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return\",\n \"( isinstance(params, dict) and 'container' in params and 'container_path' in\",\n \"args = ['ansible', host, '-m', 'template', '-c', 'local', '-a', ansible_arg]\",\n \"'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes,\",\n \"{host_1}') def restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}'\",\n \"lines_b = [] with path_a.open() as f: lines_a = f.readlines()\",\n \"/ 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path\",\n \"os.environ[x]) for x in ENV_INHERIT]) args = ['ansible', host, '-m',\",\n \"remote_path = MOODLE_DIR + '/docker-compose.yml' if version is None: version\",\n \"def _match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if not\",\n \"return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname = urlparse(moodle_url).netloc\",\n \"'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO, format='%(message)s') def generate_local_path(host, conf_path, version=None):\",\n \"params['container'] == container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path):\",\n \"dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf): return conf.parent /\",\n \"str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host,\",\n \"/ META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path):\",\n \"host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading\",\n \"= f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in\",\n \"match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1]\",\n \"format='%(message)s') def generate_local_path(host, conf_path, version=None): ret = Path(WORKDIR).absolute() / host\",\n \"conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf) def show_local_conf_diff(host,\",\n \"get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path,\",\n \"download_file(host, remote_path, conf_path=None): if conf_path is None: conf_path = Path(remote_path).name\",\n \"return download_file(host, src, conf_path) def create_conf_file(host, conf_path): dest = generate_local_path(host,\",\n \"ansible_arg = f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg])\",\n \"/= version ret /= Path(conf_path).name return ret def generate_remote_path(container, conf_path,\",\n \"f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a',\",\n \"yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name':\",\n \"['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params = { 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged':\",\n \"return ( isinstance(params, dict) and 'container' in params and 'container_path'\",\n \"out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def restart_container(host, container):\",\n \"'remote_path' in params and params['remote_path'] == remote_path) def get_versions(host, *args,\",\n \"quiet=False): org = _to_backup(conf) if not quiet: logger.info(f'Copy {conf} {org}')\",\n \"basicConfig, getLogger, INFO import os from pathlib import Path import\",\n \"= { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], } with (local_path.parent /\",\n \"(local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf):\",\n \"/ 'shibboleth.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path\",\n \"container, str(conf_path), str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path =\",\n \"urlparse from notebook import notebookapp from IPython.core.display import HTML WORKDIR\",\n \"{host_1} to {dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE):\",\n \"import subprocess import sys import yaml from urllib.parse import urlparse\",\n \"host, container, conf_path, relative_to) if restart: restart_container(host, container) def revert_conf(host,\",\n \"META_YML if not p.exists(): return False with p.open() as f:\",\n \"relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container,\",\n \"**shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def\",\n \"def show_local_conf_diff(host, container, conf_path, version=None): local_path = get_local_path(host, container, conf_path,\",\n \"dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def\",\n \"version) upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host, container)\",\n \"in pdir.glob('*') if x.is_dir() and match(x, *args)]) def find_latest_version(host, container,\",\n \"data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f:\",\n \"extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env =\",\n \"extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[],\",\n \"sorted([ x.name for x in pdir.glob('*') if x.is_dir() and match(x,\",\n \"relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) with conf.open()\",\n \"True, 'ports': ['443:443'], 'volumes': shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }),\",\n \"volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path = MOODLE_DIR +\",\n \"{local_path} to {host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle\",\n \"out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not apply:\",\n \"hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path =\",\n \"ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible', host,\",\n \"= generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path) def create_conf_file(host,\",\n \"_match_metainfo_by_remote_path(parent, remote_path): p = parent / META_YML if not p.exists():\",\n \"= generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return dest def _to_backup(conf):\",\n \"ENV_INHERIT]) args = ['ansible', host, '-m', 'template', '-c', 'local', '-a',\",\n \"parent / META_YML if not p.exists(): return False with p.open()\",\n \"conf_path, version=None): if version is None: version = find_latest_version(host, container,\",\n \"'-a', ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def\",\n \"x.name for x in pdir.glob('*') if x.is_dir() and match(x, *args)])\",\n \"def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup'] shibboleth_volumes.extend(volumes) params =\",\n \"sys import yaml from urllib.parse import urlparse from notebook import\",\n \"f.readlines() diff = list(unified_diff( lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return\",\n \"[shibboleth_vars]) show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}):\",\n \"= 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x])\",\n \"params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p = parent /\",\n \"} vars_path = conf_dir / 'shibboleth.yml' with vars_path.open(mode='w') as f:\",\n \"= yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in params\",\n \"relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def\",\n \"version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path,\",\n \"datetime from difflib import unified_diff from logging import basicConfig, getLogger,\",\n \"nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir']))\",\n \"Path(conf_path).name return ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) /\",\n \"Path import shutil import subprocess import sys import yaml from\",\n \"dict) and 'remote_path' in params and params['remote_path'] == remote_path) def\",\n \"conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path): return get_versions( host, remote_path, match=_match_metainfo_by_remote_path)[-1] def\",\n \"/ (conf.name + '.orig') def make_backup(conf, quiet=False): org = _to_backup(conf)\",\n \"else: ret /= version ret /= Path(conf_path).name return ret def\",\n \"p = parent / META_YML if not p.exists(): return False\",\n \"/= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version ret /= Path(conf_path).name return\",\n \"relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host, src, conf_path)\",\n \"ret def generate_remote_path(container, conf_path, relative_to=CONF_RELATIVE): return (Path(MOODLE_DIR) / container /\",\n \"META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf =\",\n \"HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a = [] lines_b\",\n \"logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host, container,\",\n \"def save_shibboleth_part(conf_path): with conf_path.open() as f: data = yaml.safe_load(f) params\",\n \"from {host_1} to {dest}') return dest def download_conf_file(host, container, conf_path,\",\n \"} with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False)\",\n \"stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0] p = (Path(nb_conf['base_url'])\",\n \"relative_to) return generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path\",\n \"quiet: logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path,\",\n \"f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes = ['/sys/fs/cgroup:/sys/fs/cgroup']\",\n \"params is None or len(params) == 0: return save_shibboleth_part(local_path) else:\",\n \"'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x]) for\",\n \"{} if 'shibboleth' in data['services']: params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path\",\n \"return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars,\",\n \"docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template'\",\n \"conf_path, version=None): ret = Path(WORKDIR).absolute() / host if version is\",\n \"version=None): conf = get_local_path(host, container, conf_path, version) with conf.open() as\",\n \"create_conf_file(host, conf_path): dest = generate_local_path(host, conf_path) dest.parent.mkdir(parents=True, exist_ok=True) dest.touch() return\",\n \"(Path(MOODLE_DIR) / container / 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container,\",\n \"'fetch', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Downloading {remote_path} from {host_1}\",\n \"from IPython.core.display import HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml'\",\n \"{org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params =\",\n \"as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def generate_edit_link(conf): nb_conf = list(notebookapp.list_running_servers())[0]\",\n \"relative_to=CONF_RELATIVE, version=None): conf = get_local_path(host, container, conf_path, version) return generate_edit_link(conf)\",\n \"container, conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True): conf_path\",\n \"conf_path=None): if conf_path is None: conf_path = Path(remote_path).name dest =\",\n \"yaml.safe_load(f) params = {} if 'shibboleth' in data['services']: params['shibboleth_container'] =\",\n \"os from pathlib import Path import shutil import subprocess import\",\n \"version = find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def\",\n \"logger.info(f'Copy {conf} {org}') shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE):\",\n \"ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version ret /= Path(conf_path).name\",\n \"generate_edit_link(local_path) def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host,\",\n \"from logging import basicConfig, getLogger, INFO import os from pathlib\",\n \"path_b): lines_a = [] lines_b = [] with path_a.open() as\",\n \"str(conf_path)) local_path = generate_local_path(host, conf_path, version) generate_proxy_conf(host, local_path, extra_vars) show_local_conf_diff(host,\",\n \"show_diff(_to_backup(local_path), local_path) return generate_edit_link(local_path) def append_shibboleth_container(host, moodle_url, volumes=[], extra_vars={}): hostname\",\n \"{container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible', host, '-a', cmd]) def fetch_conf(host,\",\n \"os.environ[x]) for x in ENV_INHERIT]) args = [ 'ansible', host,\",\n \"backup=yes' out = subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a',\",\n \"v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host,\",\n \"= subprocess.check_output( ['ansible', host, '-m', 'fetch', '-a', ansible_arg]) host_1 =\",\n \"f'src={remote_path} dest={dest} flat=yes' out = subprocess.check_output( ['ansible', host, '-m', 'fetch',\",\n \"dest def _to_backup(conf): return conf.parent / (conf.name + '.orig') def\",\n \"'/etc' ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__)\",\n \"(Path(nb_conf['base_url']) / 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def\",\n \"to {host_1}') if not apply: return ansible_arg = f'chdir=/opt/moodle docker-compose\",\n \"str(conf_path.parent)) version = find_latest_version(host, container, str(conf_path)) local_path = generate_local_path(host, conf_path,\",\n \"container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent, remote_path): p =\",\n \"extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args,\",\n \"find_latest_version(host, container, conf_path) return generate_local_path(host, conf_path, version) def _match_metainfo(parent, container,\",\n \"return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args =\",\n \"upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name\",\n \"def find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host,\",\n \"def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = download_conf_file(host, container,\",\n \"'-a', cmd]) def fetch_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path =\",\n \"host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None,\",\n \"= 'template/docker/compose/docker-compose.yml' ansible_arg = f'src={template} dest={conf_path.parent}/' env = dict([(x, os.environ[x])\",\n \"subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1 =\",\n \"and params['container'] == container and params['container_path'] == conf_path) def _match_metainfo_by_remote_path(parent,\",\n \"container, conf_path, relative_to) restart_container(host, container) local_path.rename(local_path.parent / (local_path.name + '.revert'))\",\n \"relative_to) if restart: restart_container(host, container) def revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE,\",\n \"subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf') container\",\n \"= get_local_path(host, container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path)\",\n \"ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}')\",\n \"local_path = download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path,\",\n \"= f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg\",\n \"revert_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path,\",\n \"dest = generate_remote_path(container, conf_path, relative_to) ansible_arg = f'mkdir -p {dest.parent}'\",\n \"== 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host,\",\n \"= dict([(x, os.environ[x]) for x in ENV_INHERIT]) args = ['ansible',\",\n \"and match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host, container,\",\n \"ENV_INHERIT = ['VAULT_ADDR', 'VAULT_TOKEN', 'PATH', 'REQUESTS_CA_BUNDLE'] logger = getLogger(__name__) basicConfig(level=INFO,\",\n \"extra_vars_file: args.extend(['-e', f'@{str(x)}']) subprocess.run(args=args, env=env, check=True) def update_docker_compose(host, extra_vars={}, shibboleth_params={}):\",\n \"x.is_dir() and match(x, *args)]) def find_latest_version(host, container, conf_path): return get_versions(host,\",\n \"import sys import yaml from urllib.parse import urlparse from notebook\",\n \"generate_proxy_conf(host, conf_path, extra_vars): template = 'template/docker/compose/moodle-proxy.conf.template' ansible_arg = f'src={template} dest={conf_path.parent}/moodle-proxy.conf'\",\n \"find_latest_version(host, container, conf_path): return get_versions(host, container, conf_path)[-1] def find_latest_version_by_remote_path(host, remote_path):\",\n \"None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version ret /=\",\n \"remote_path) def get_versions(host, *args, match=_match_metainfo): pdir = Path(WORKDIR).absolute() / host\",\n \"'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out = subprocess.check_output( ['ansible',\",\n \"conf = get_local_path(host, container, conf_path, version) with conf.open() as f:\",\n \"isinstance(params, dict) and 'remote_path' in params and params['remote_path'] == remote_path)\",\n \"f'mkdir -p {dest.parent}' subprocess.run( ['ansible', host, '-a', ansible_arg]) ansible_arg =\",\n \"def update_docker_compose(host, extra_vars={}, shibboleth_params={}): remote_path = MOODLE_DIR + '/docker-compose.yml' local_path\",\n \"cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container {container}') subprocess.check_call(['ansible',\",\n \"extra_vars={}): hostname = urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname,\",\n \"def create_conf(host, container, conf_path, relative_to=CONF_RELATIVE, create=False): local_path = create_conf_file(host, conf_path)\",\n \"as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params): if\",\n \"k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def\",\n \"shibboleth_params={'hostname': hostname, 'volumes': volumes}, ) def upload_docker_compose(host, version=None, apply=False): remote_path\",\n \"yaml from urllib.parse import urlparse from notebook import notebookapp from\",\n \"diff = show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container,\",\n \"container, conf_path, version) with conf.open() as f: print(f.read()) def edit_local_conf(host,\",\n \"HTML WORKDIR = 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle'\",\n \"= generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out =\",\n \"yaml.safe_dump(params, f) return vars_path def init_shibboleth_part(conf_dir, hostname, volumes): shibboleth_volumes =\",\n \"/ META_YML if not p.exists(): return False with p.open() as\",\n \"with path_b.open() as f: lines_b = f.readlines() diff = list(unified_diff(\",\n \"ansible_arg]) ansible_arg = f'dest={dest} src={src} backup=yes' out = subprocess.check_output( ['ansible',\",\n \"ansible_arg] logger.info('Apply the changes in docker-compose.yml.') subprocess.run(args=args, check=True) def generate_proxy_conf(host,\",\n \"args = [ 'ansible', host, '-m', 'template', '-c', 'local', '-a',\",\n \"or len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent, **params)\",\n \"/ version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes' out\",\n \"container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host, container, conf_path,\",\n \"/ 'conf' / Path(conf_path).relative_to(relative_to)) def get_local_path(host, container, conf_path, version=None): if\",\n \"def apply_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None, restart=True): diff = show_local_conf_diff(host,\",\n \"= MOODLE_DIR + '/docker-compose.yml' local_path = download_file(host, remote_path) make_backup(local_path) make_simple_metainfo(local_path,\",\n \"dest={conf_path.parent}/moodle-proxy.conf' env = dict([(x, os.environ[x]) for x in ENV_INHERIT]) args\",\n \"= out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {remote_path} from {local_path} to {host_1}') if not\",\n \"ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x\",\n \"f'{k}={v}']) subprocess.run(args=args, env=env, check=True) def update_proxy_conf(host, extra_vars={}): conf_path = Path('/usr/local/apache2/conf/moodle-proxy.conf')\",\n \"/ 'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a,\",\n \"= get_local_path(host, container, conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with\",\n \"generate_local_path(host, conf_path, version) def _match_metainfo(parent, container, conf_path): p = parent\",\n \"= f.readlines() with path_b.open() as f: lines_b = f.readlines() diff\",\n \"None or len(params) == 0: return save_shibboleth_part(local_path) else: return init_shibboleth_part(local_path.parent,\",\n \"show_local_conf_diff(host, container, conf_path, version) return link def apply_proxy_conf(host, version=None, restart=True):\",\n \"apply: return ansible_arg = f'chdir=/opt/moodle docker-compose up -d --remove-orphans' args\",\n \"args = ['ansible', host, '-a', ansible_arg] logger.info('Apply the changes in\",\n \"shibboleth_volumes, 'container_name': 'shibboleth', 'hostname': hostname, }), } vars_path = conf_dir\",\n \"params and 'container_path' in params and params['container'] == container and\",\n \"params = yaml.safe_load(f) return ( isinstance(params, dict) and 'remote_path' in\",\n \"return conf.parent / (conf.name + '.orig') def make_backup(conf, quiet=False): org\",\n \"'.orig') def make_backup(conf, quiet=False): org = _to_backup(conf) if not quiet:\",\n \"ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}')\",\n \"/ (local_path.name + '.revert')) def show_local_conf(host, container, conf_path, relative_to=CONF_RELATIVE, version=None):\",\n \"init_shibboleth_part(local_path.parent, **params) def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml'\",\n \"host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src} to {host_1}') def\",\n \"return ( isinstance(params, dict) and 'remote_path' in params and params['remote_path']\",\n \"make_simple_metainfo(local_path, remote_path): params = { 'remote_path': remote_path, 'version': list(local_path.parts)[-2], }\",\n \"= subprocess.check_output( ['ansible', host, '-m', 'copy', '-b', '-a', ansible_arg]) host_1\",\n \"'local', '-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e', f'{k}={v}'])\",\n \"to {dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src\",\n \"{dest}') return dest def download_conf_file(host, container, conf_path, relative_to=CONF_RELATIVE): src =\",\n \"= show_local_conf_diff(host, container, conf_path, version) local_path = get_local_path(host, container, conf_path,\",\n \"conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as\",\n \"= conf_path.parent / 'extra_vars.yml' with vars_path.open(mode='w') as f: yaml.safe_dump(params, f)\",\n \"= urlparse(moodle_url).netloc return update_docker_compose( host, extra_vars, shibboleth_params={'hostname': hostname, 'volumes': volumes},\",\n \"'-b', '-a', ansible_arg]) host_1 = out.decode('utf-8').split(\\\"\\\\n\\\")[0].split()[0] logger.info(f'Uploading {dest} from {src}\",\n \"container, conf_path, relative_to=CONF_RELATIVE, version=None): local_path = get_local_path(host, container, conf_path, version)\",\n \"restart_container(host, container): cmd = f'chdir={MOODLE_DIR} docker-compose restart {container}' logger.info(f'Restart container\",\n \"conf_path, version) with conf.open() as f: print(f.read()) def edit_local_conf(host, container,\",\n \"is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else: ret /= version ret\",\n \"'edit' / conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b):\",\n \"create=False): local_path = create_conf_file(host, conf_path) make_backup(local_path, quiet=True) make_metainfo(local_path, container, conf_path,\",\n \"= 'edit' META_YML = '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE =\",\n \"host / version / 'docker-compose.yml') ansible_arg = f'dest={remote_path} src={local_path} backup=yes'\",\n \"upload_conf_file(local_path, host, container, conf_path, relative_to) if restart: restart_container(host, container) def\",\n \"{ 'shibboleth_container': yaml.safe_dump({ 'image': 'harbor.vcloud.nii.ac.jp/vcp/moodle:shibboleth-3.0.4', 'privileged': True, 'ports': ['443:443'], 'volumes':\",\n \"INFO import os from pathlib import Path import shutil import\",\n \"'version': list(local_path.parts)[-2], } with (local_path.parent / META_YML).open(mode='w') as f: yaml.safe_dump(params,\",\n \"v in extra_vars.items(): args.extend(['-e', f'{k}={v}']) for x in extra_vars_file: args.extend(['-e',\",\n \"/ host if version is None: ret /= datetime.now().strftime(\\\"%Y%m%d%H%M%S%f\\\") else:\",\n \"conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container,\",\n \"params['shibboleth_container'] = yaml.safe_dump( data['services']['shibboleth']) vars_path = conf_path.parent / 'extra_vars.yml' with\",\n \"shibboleth_vars = setup_shibboleth_part(local_path, **shibboleth_params) generate_docker_compose(host, local_path, extra_vars, [shibboleth_vars]) show_diff(_to_backup(local_path), local_path)\",\n \"container, conf_path, version) backup_path = _to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host,\",\n \"vars_path.open(mode='w') as f: yaml.safe_dump(params, f) return vars_path def setup_shibboleth_part(local_path, **params):\",\n \"get_local_path(host, container, conf_path, version) upload_conf_file(local_path, host, container, conf_path, relative_to) if\",\n \"dest = generate_local_path(host, conf_path) ansible_arg = f'src={remote_path} dest={dest} flat=yes' out\",\n \"version=None): ret = Path(WORKDIR).absolute() / host if version is None:\",\n \"lines_a, lines_b, fromfile=path_a.name, tofile=path_b.name)) sys.stdout.writelines(diff) return len(diff) def upload_conf_file(src, host,\",\n \"conf_path, version) show_diff(_to_backup(local_path), local_path) def save_shibboleth_part(conf_path): with conf_path.open() as f:\",\n \"'remote_path': str(generate_remote_path(container, conf_path, relative_to)), 'version': list(local_path.parts)[-2], } with (local_path.parent /\",\n \"META_YML).open(mode='w') as f: yaml.safe_dump(params, stream=f, default_flow_style=False) def make_simple_metainfo(local_path, remote_path): params\",\n \"'-c', 'local', '-a', ansible_arg] for k, v in extra_vars.items(): args.extend(['-e',\",\n \"version = find_latest_version_by_remote_path(host, remote_path) local_path = ( Path(WORKDIR).absolute() / host\",\n \"conf.absolute().relative_to(nb_conf['notebook_dir'])) return HTML(f'<a href={p} target=\\\"_blank\\\">{p.name}</a>') def show_diff(path_a, path_b): lines_a =\",\n \"def upload_conf_file(src, host, container, conf_path, relative_to=CONF_RELATIVE): dest = generate_remote_path(container, conf_path,\",\n \"shutil.copy2(conf, org) def make_metainfo(local_path, container, conf_path, relative_to=CONF_RELATIVE): params = {\",\n \"download_conf_file(host, container, conf_path, relative_to) make_backup(local_path) make_metainfo(local_path, container, conf_path, relative_to) return\",\n \"container, conf_path, relative_to=CONF_RELATIVE): src = generate_remote_path(container, conf_path, relative_to) return download_file(host,\",\n \"= '.vcp-meta.yml' MOODLE_DIR = '/opt/moodle' CONF_RELATIVE = '/etc' ENV_INHERIT =\"\n]","isConversational":false}}},{"rowIdx":695,"cells":{"ngram":{"kind":"list like","value":["edit distance computes the cost it takes to get from","len(source) m = len(target) D = {} # Initialization for","j in range(0, m+1): D[0,j] = j for i in","in range(1, m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1,","i in range(0, n+1): D[i,0] = i for j in","computes the cost it takes to get from one string","for insertions or deletions and a cost of 2 for","= D[i-1, j-1] else: D[i,j] = min( D[i-1, j] +","else: D[i,j] = min( D[i-1, j] + 1, D[i, j-1]","to another string. This implementation uses the Levenshtein distance with","get from one string to another string. This implementation uses","= len(target) D = {} # Initialization for i in","for j in range(1, m+1): if source[i-1] == target[j-1]: D[i,j]","deletions and a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance","it takes to get from one string to another string.","substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \"intention\" to \"execution\"","This implementation uses the Levenshtein distance with a cost of","uses the Levenshtein distance with a cost of 1 for","m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else:","min( D[i-1, j] + 1, D[i, j-1] + 1, D[i-1,","source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] =","takes to get from one string to another string. This","another string. This implementation uses the Levenshtein distance with a","distance with a cost of 1 for insertions or deletions","= len(source) m = len(target) D = {} # Initialization","== target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] = min(","Initialization for i in range(0, n+1): D[i,0] = i for","if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j]","one string to another string. This implementation uses the Levenshtein","Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \"intention\" to \"execution\" is","2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \"intention\"","len(target) D = {} # Initialization for i in range(0,","# Initialization for i in range(0, n+1): D[i,0] = i","D[i-1, j-1] else: D[i,j] = min( D[i-1, j] + 1,","string. This implementation uses the Levenshtein distance with a cost","is a cost of 8. minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\"","a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example,","minimum_edit_distance(source, target): n = len(source) m = len(target) D =","1, D[i, j-1] + 1, D[i-1, j-1] + 2 )","in range(0, m+1): D[0,j] = j for i in range(1,","in range(0, n+1): D[i,0] = i for j in range(0,","getting from \"intention\" to \"execution\" is a cost of 8.","n = len(source) m = len(target) D = {} #","target): n = len(source) m = len(target) D = {}","and a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For","{} # Initialization for i in range(0, n+1): D[i,0] =","\"execution\") # 8 \"\"\" def minimum_edit_distance(source, target): n = len(source)","\"\"\" Minimum edit distance computes the cost it takes to","for i in range(0, n+1): D[i,0] = i for j","\"intention\" to \"execution\" is a cost of 8. minimum_edit_distance(\"intention\", \"execution\")","def minimum_edit_distance(source, target): n = len(source) m = len(target) D","of 1 for insertions or deletions and a cost of","insertions or deletions and a cost of 2 for substitutions.","minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\" def minimum_edit_distance(source, target): n =","# 8 \"\"\" def minimum_edit_distance(source, target): n = len(source) m","for j in range(0, m+1): D[0,j] = j for i","target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] = min( D[i-1,","from one string to another string. This implementation uses the","to \"execution\" is a cost of 8. minimum_edit_distance(\"intention\", \"execution\") #","j for i in range(1, n+1): for j in range(1,","to get from one string to another string. This implementation","Minimum edit distance computes the cost it takes to get","range(0, n+1): D[i,0] = i for j in range(0, m+1):","implementation uses the Levenshtein distance with a cost of 1","cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting","from \"intention\" to \"execution\" is a cost of 8. minimum_edit_distance(\"intention\",","the Levenshtein distance with a cost of 1 for insertions","1 for insertions or deletions and a cost of 2","j in range(1, m+1): if source[i-1] == target[j-1]: D[i,j] =","D[0,j] = j for i in range(1, n+1): for j","range(1, m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1]","distance computes the cost it takes to get from one","range(1, n+1): for j in range(1, m+1): if source[i-1] ==","\"\"\" def minimum_edit_distance(source, target): n = len(source) m = len(target)","D[i, j-1] + 1, D[i-1, j-1] + 2 ) return","with a cost of 1 for insertions or deletions and","a cost of 8. minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\" def","D[i,j] = D[i-1, j-1] else: D[i,j] = min( D[i-1, j]","a cost of 1 for insertions or deletions and a","i in range(1, n+1): for j in range(1, m+1): if","= min( D[i-1, j] + 1, D[i, j-1] + 1,","cost of 1 for insertions or deletions and a cost","cost it takes to get from one string to another","https://en.wikipedia.org/wiki/Edit_distance For example, getting from \"intention\" to \"execution\" is a","for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \"intention\" to","string to another string. This implementation uses the Levenshtein distance","cost of 8. minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\" def minimum_edit_distance(source,","+ 1, D[i, j-1] + 1, D[i-1, j-1] + 2","Levenshtein distance with a cost of 1 for insertions or","= {} # Initialization for i in range(0, n+1): D[i,0]","the cost it takes to get from one string to","example, getting from \"intention\" to \"execution\" is a cost of","D = {} # Initialization for i in range(0, n+1):","range(0, m+1): D[0,j] = j for i in range(1, n+1):","n+1): for j in range(1, m+1): if source[i-1] == target[j-1]:","of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from","D[i,j] = min( D[i-1, j] + 1, D[i, j-1] +","8 \"\"\" def minimum_edit_distance(source, target): n = len(source) m =","of 8. minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\" def minimum_edit_distance(source, target):","i for j in range(0, m+1): D[0,j] = j for","8. minimum_edit_distance(\"intention\", \"execution\") # 8 \"\"\" def minimum_edit_distance(source, target): n","D[i,0] = i for j in range(0, m+1): D[0,j] =","j-1] else: D[i,j] = min( D[i-1, j] + 1, D[i,","= j for i in range(1, n+1): for j in","D[i-1, j] + 1, D[i, j-1] + 1, D[i-1, j-1]","n+1): D[i,0] = i for j in range(0, m+1): D[0,j]","or deletions and a cost of 2 for substitutions. Resource:","+ 1, D[i-1, j-1] + 2 ) return D[n-1, m-1]","\"execution\" is a cost of 8. minimum_edit_distance(\"intention\", \"execution\") # 8","in range(1, n+1): for j in range(1, m+1): if source[i-1]","for i in range(1, n+1): for j in range(1, m+1):","j] + 1, D[i, j-1] + 1, D[i-1, j-1] +","m = len(target) D = {} # Initialization for i","j-1] + 1, D[i-1, j-1] + 2 ) return D[n-1,","= i for j in range(0, m+1): D[0,j] = j","m+1): D[0,j] = j for i in range(1, n+1): for","For example, getting from \"intention\" to \"execution\" is a cost"],"string":"[\n \"edit distance computes the cost it takes to get from\",\n \"len(source) m = len(target) D = {} # Initialization for\",\n \"j in range(0, m+1): D[0,j] = j for i in\",\n \"in range(1, m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1,\",\n \"i in range(0, n+1): D[i,0] = i for j in\",\n \"computes the cost it takes to get from one string\",\n \"for insertions or deletions and a cost of 2 for\",\n \"= D[i-1, j-1] else: D[i,j] = min( D[i-1, j] +\",\n \"else: D[i,j] = min( D[i-1, j] + 1, D[i, j-1]\",\n \"to another string. This implementation uses the Levenshtein distance with\",\n \"get from one string to another string. This implementation uses\",\n \"= len(target) D = {} # Initialization for i in\",\n \"for j in range(1, m+1): if source[i-1] == target[j-1]: D[i,j]\",\n \"deletions and a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance\",\n \"it takes to get from one string to another string.\",\n \"substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \\\"intention\\\" to \\\"execution\\\"\",\n \"This implementation uses the Levenshtein distance with a cost of\",\n \"uses the Levenshtein distance with a cost of 1 for\",\n \"m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else:\",\n \"min( D[i-1, j] + 1, D[i, j-1] + 1, D[i-1,\",\n \"source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] =\",\n \"takes to get from one string to another string. This\",\n \"another string. This implementation uses the Levenshtein distance with a\",\n \"distance with a cost of 1 for insertions or deletions\",\n \"= len(source) m = len(target) D = {} # Initialization\",\n \"== target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] = min(\",\n \"Initialization for i in range(0, n+1): D[i,0] = i for\",\n \"if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j]\",\n \"one string to another string. This implementation uses the Levenshtein\",\n \"Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \\\"intention\\\" to \\\"execution\\\" is\",\n \"2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \\\"intention\\\"\",\n \"len(target) D = {} # Initialization for i in range(0,\",\n \"# Initialization for i in range(0, n+1): D[i,0] = i\",\n \"D[i-1, j-1] else: D[i,j] = min( D[i-1, j] + 1,\",\n \"string. This implementation uses the Levenshtein distance with a cost\",\n \"is a cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\"\",\n \"a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example,\",\n \"minimum_edit_distance(source, target): n = len(source) m = len(target) D =\",\n \"1, D[i, j-1] + 1, D[i-1, j-1] + 2 )\",\n \"in range(0, m+1): D[0,j] = j for i in range(1,\",\n \"in range(0, n+1): D[i,0] = i for j in range(0,\",\n \"getting from \\\"intention\\\" to \\\"execution\\\" is a cost of 8.\",\n \"n = len(source) m = len(target) D = {} #\",\n \"target): n = len(source) m = len(target) D = {}\",\n \"and a cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For\",\n \"{} # Initialization for i in range(0, n+1): D[i,0] =\",\n \"\\\"execution\\\") # 8 \\\"\\\"\\\" def minimum_edit_distance(source, target): n = len(source)\",\n \"\\\"\\\"\\\" Minimum edit distance computes the cost it takes to\",\n \"for i in range(0, n+1): D[i,0] = i for j\",\n \"\\\"intention\\\" to \\\"execution\\\" is a cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\")\",\n \"def minimum_edit_distance(source, target): n = len(source) m = len(target) D\",\n \"of 1 for insertions or deletions and a cost of\",\n \"insertions or deletions and a cost of 2 for substitutions.\",\n \"minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\" def minimum_edit_distance(source, target): n =\",\n \"# 8 \\\"\\\"\\\" def minimum_edit_distance(source, target): n = len(source) m\",\n \"for j in range(0, m+1): D[0,j] = j for i\",\n \"target[j-1]: D[i,j] = D[i-1, j-1] else: D[i,j] = min( D[i-1,\",\n \"from one string to another string. This implementation uses the\",\n \"to \\\"execution\\\" is a cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") #\",\n \"j for i in range(1, n+1): for j in range(1,\",\n \"to get from one string to another string. This implementation\",\n \"Minimum edit distance computes the cost it takes to get\",\n \"range(0, n+1): D[i,0] = i for j in range(0, m+1):\",\n \"implementation uses the Levenshtein distance with a cost of 1\",\n \"cost of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting\",\n \"from \\\"intention\\\" to \\\"execution\\\" is a cost of 8. minimum_edit_distance(\\\"intention\\\",\",\n \"the Levenshtein distance with a cost of 1 for insertions\",\n \"1 for insertions or deletions and a cost of 2\",\n \"j in range(1, m+1): if source[i-1] == target[j-1]: D[i,j] =\",\n \"D[0,j] = j for i in range(1, n+1): for j\",\n \"range(1, m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1, j-1]\",\n \"distance computes the cost it takes to get from one\",\n \"range(1, n+1): for j in range(1, m+1): if source[i-1] ==\",\n \"\\\"\\\"\\\" def minimum_edit_distance(source, target): n = len(source) m = len(target)\",\n \"D[i, j-1] + 1, D[i-1, j-1] + 2 ) return\",\n \"with a cost of 1 for insertions or deletions and\",\n \"a cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\" def\",\n \"D[i,j] = D[i-1, j-1] else: D[i,j] = min( D[i-1, j]\",\n \"a cost of 1 for insertions or deletions and a\",\n \"i in range(1, n+1): for j in range(1, m+1): if\",\n \"= min( D[i-1, j] + 1, D[i, j-1] + 1,\",\n \"cost of 1 for insertions or deletions and a cost\",\n \"cost it takes to get from one string to another\",\n \"https://en.wikipedia.org/wiki/Edit_distance For example, getting from \\\"intention\\\" to \\\"execution\\\" is a\",\n \"for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from \\\"intention\\\" to\",\n \"string to another string. This implementation uses the Levenshtein distance\",\n \"cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\" def minimum_edit_distance(source,\",\n \"+ 1, D[i, j-1] + 1, D[i-1, j-1] + 2\",\n \"Levenshtein distance with a cost of 1 for insertions or\",\n \"= {} # Initialization for i in range(0, n+1): D[i,0]\",\n \"the cost it takes to get from one string to\",\n \"example, getting from \\\"intention\\\" to \\\"execution\\\" is a cost of\",\n \"D = {} # Initialization for i in range(0, n+1):\",\n \"range(0, m+1): D[0,j] = j for i in range(1, n+1):\",\n \"n+1): for j in range(1, m+1): if source[i-1] == target[j-1]:\",\n \"of 2 for substitutions. Resource: https://en.wikipedia.org/wiki/Edit_distance For example, getting from\",\n \"D[i,j] = min( D[i-1, j] + 1, D[i, j-1] +\",\n \"8 \\\"\\\"\\\" def minimum_edit_distance(source, target): n = len(source) m =\",\n \"of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\" def minimum_edit_distance(source, target):\",\n \"i for j in range(0, m+1): D[0,j] = j for\",\n \"8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8 \\\"\\\"\\\" def minimum_edit_distance(source, target): n\",\n \"D[i,0] = i for j in range(0, m+1): D[0,j] =\",\n \"j-1] else: D[i,j] = min( D[i-1, j] + 1, D[i,\",\n \"= j for i in range(1, n+1): for j in\",\n \"D[i-1, j] + 1, D[i, j-1] + 1, D[i-1, j-1]\",\n \"n+1): D[i,0] = i for j in range(0, m+1): D[0,j]\",\n \"or deletions and a cost of 2 for substitutions. Resource:\",\n \"+ 1, D[i-1, j-1] + 2 ) return D[n-1, m-1]\",\n \"\\\"execution\\\" is a cost of 8. minimum_edit_distance(\\\"intention\\\", \\\"execution\\\") # 8\",\n \"in range(1, n+1): for j in range(1, m+1): if source[i-1]\",\n \"for i in range(1, n+1): for j in range(1, m+1):\",\n \"j] + 1, D[i, j-1] + 1, D[i-1, j-1] +\",\n \"m = len(target) D = {} # Initialization for i\",\n \"j-1] + 1, D[i-1, j-1] + 2 ) return D[n-1,\",\n \"= i for j in range(0, m+1): D[0,j] = j\",\n \"m+1): D[0,j] = j for i in range(1, n+1): for\",\n \"For example, getting from \\\"intention\\\" to \\\"execution\\\" is a cost\"\n]","isConversational":false}}},{"rowIdx":696,"cells":{"ngram":{"kind":"list like","value":["outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm",": new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in","new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in vcf_records:","os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True,","addition to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile,","this is a deletion. So need to # get the","well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start","\".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] ) ) elif variant.var_type ==","logging.info(\"Finish run command: \" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff(","variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] ) ) elif","maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as f_out: for","the VCF has the correct REF and ALT sequences if","def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ):","import shutil import subprocess import pyfastaq import pymummer from cluster_vcf_records","--threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta}","mummer. Need to do the appropriate # reverse (complement) fixes","elif variant.var_type == pymummer.variant.DEL: # The query has sequence missing,","from operator import attrgetter import logging import os import shutil","outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader =","a new VCF file unmerged records.\"\"\" vcf_records = {} variants","raise Exception(\"Unknown variant type: \" + str(variant)) assert ( new_record.REF","qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base","delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\",","We're making VCF records w.r.t. the # query, so this","for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as f:","print(\"##fileformat=VCFv4.2\", file=f) for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f)","def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the .snps file made by","\"1/1\", ] ) ) else: raise Exception(\"Unknown variant type: \"","[ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start","VCF file unmerged records.\"\"\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file)","vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for","snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if","import os import shutil import subprocess import pyfastaq import pymummer","vcf_record from varifier import utils # We only want the","if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq))","the deletion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name,","vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with","open(outfile, \"wb\") as f_out: for snp_file in tmp_snp_files: with open(snp_file,","The query has sequence missing, compared to the # reference.","f\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1","f: print(\"##fileformat=VCFv4.2\", file=f) for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\",","ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = []","query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] ) if new_record.CHROM","> {delta_1}\", f\"show-snps -rlTHC {delta_1} > {outfile}\", ] for command","of query sequences. Writes a new VCF file unmerged records.\"\"\"","variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base =","+ str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS","vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False,","str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\",","\" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile,","f\"delta-filter -1 {delta} > {delta_1}\", f\"show-snps -rlTHC {delta_1} > {outfile}\",","{delta} > {delta_1}\", f\"show-snps -rlTHC {delta_1} > {outfile}\", ] for","threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f","not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the .snps","had to be # reverse complemented when aligned by mummer.","= pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: #","delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True) maxmatch_opt =","[] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix =","need to # get the nucleotide before the insertion as","prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as","\"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", variant.qry_base, variant.ref_base, \".\",","has the correct REF and ALT sequences if variant.reverse: qry_seq","snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads,","file unmerged records.\"\"\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs","variants: # If the variant is reversed, it means that","snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as",") ) elif variant.var_type == pymummer.variant.DEL: # The query has","for command in commands: logging.info(\"Start run command: \" + command)","str(variant.qry_start + 1), \".\", variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\",","str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS +","VCF records w.r.t. the # query, so this is a","made by dnadiff. query_fasta = fasta file of query sequences.","runs several other commands # in addition to making the","command) subprocess.check_output(command, shell=True) logging.info(\"Finish run command: \" + command) os.unlink(delta)","query, so this is an insertion. So need to #","variant.qry_name, str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\",","truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\"","missing, compared to the # reference. We're making VCF records","If the variant is reversed, it means that either the","[ variant.qry_name, str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base,","cluster_vcf_records import vcf_record from varifier import utils # We only","record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile,","-1 out.delta > out.1delta # show-snps -rlTHC out.1delta > out.snps","w.r.t. the # query, so this is an insertion. So","threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as f_out:","from MUMmer. From reading # the docs inspecting that script,","if maxmatch else \"\" commands = [ f\"nucmer --threads {threads}","run command: \" + command) subprocess.check_output(command, shell=True) logging.info(\"Finish run command:","variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] ) )","so the VCF has the correct REF and ALT sequences","# delta-filter -1 out.delta > out.1delta # show-snps -rlTHC out.1delta","in commands: logging.info(\"Start run command: \" + command) subprocess.check_output(command, shell=True)","= vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start],","file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\"","tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\")","file of query sequences. Writes a new VCF file unmerged","query has sequence missing, compared to the # reference. We're","means that either the ref or query had to be","deletion. So need to # get the nucleotide before the","reverse complemented when aligned by mummer. Need to do the","query sequences. Writes a new VCF file unmerged records.\"\"\" vcf_records","open(tmp_fasta, \"w\") as f: print(seq, file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split(","1), \".\", variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ]","): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch","type: \" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS",") ) elif variant.var_type == pymummer.variant.INS: # The ref has","compared to the # query. We're making VCF records w.r.t.","--delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta} > {delta_1}\", f\"show-snps","to # get the nucleotide before the deletion as well.","sequence missing, compared to the # reference. We're making VCF","= f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch,","shell=True) logging.info(\"Finish run command: \" + command) os.unlink(delta) os.unlink(delta_1) def","the # query. We're making VCF records w.r.t. the #","split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files","subprocess import pyfastaq import pymummer from cluster_vcf_records import vcf_record from","that script, we need to run these commands: # #","# The ref has sequence missing, compared to the #","debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta,","insertion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start","attrgetter import logging import os import shutil import subprocess import","to the # reference. We're making VCF records w.r.t. the","has sequence missing, compared to the # reference. We're making","\"w\") as f: print(\"##fileformat=VCFv4.2\", file=f) for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\",","qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq =","insertion. So need to # get the nucleotide before the","\"w\") as f: print(seq, file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta,","\"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] ) ) elif variant.var_type == pymummer.variant.INS:","\" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS :","not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else:","for seq in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\"",") elif variant.var_type == pymummer.variant.DEL: # The query has sequence","ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file =","vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\", file=f) for seq","instead of just running show-snps, which runs several other commands","if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch )","os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as f_out: for snp_file in","= [ f\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\",","variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start","== pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start +","variant.qry_name, str(variant.qry_start + 1), \".\", variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\",","making VCF records w.r.t. the # query, so this is","_run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file,","key, vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f)","file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True,","is a deletion. So need to # get the nucleotide","+ len(new_record.REF) ] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM]","or query had to be # reverse complemented when aligned","{} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in","# get the nucleotide before the deletion as well. new_record","{delta_1} > {outfile}\", ] for command in commands: logging.info(\"Start run","is reversed, it means that either the ref or query","we need to run these commands: # # nucmer --maxmatch","+ 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\",","f: print(seq, file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file,","debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta,","debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the .snps file","to do the appropriate # reverse (complement) fixes so the","get the nucleotide before the deletion as well. new_record =",".snps file made by dnadiff. query_fasta = fasta file of","new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list","by dnadiff. query_fasta = fasta file of query sequences. Writes","sequences. Writes a new VCF file unmerged records.\"\"\" vcf_records =","pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP:","commands # in addition to making the snps file. def","\".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] ) ) elif variant.var_type","vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\",","with open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\", file=f) for seq in",") else: raise Exception(\"Unknown variant type: \" + str(variant)) assert","len(new_record.REF) ] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] =","variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp()","to be # reverse complemented when aligned by mummer. Need","1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\",","as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start +","# query. We're making VCF records w.r.t. the # query,","the docs inspecting that script, we need to run these","out.snps # # This is instead of just running show-snps,","query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\",","tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader:","from the dnadiff script from MUMmer. From reading # the","1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\",","> out.1delta # show-snps -rlTHC out.1delta > out.snps # #","--maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta >","os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the .snps file made","seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta, \"w\")","variant in variants: # If the variant is reversed, it","pymummer.variant.INS: # The ref has sequence missing, compared to the","f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\" if","ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile,","debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not debug:","file made by dnadiff. query_fasta = fasta file of query","f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as f: print(seq,","= f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\"","query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list in sorted(vcf_records.items()):","os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1,","in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def make_truth_vcf(","truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta,","{delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta} > {delta_1}\", f\"show-snps -rlTHC","import pymummer from cluster_vcf_records import vcf_record from varifier import utils","fixes so the VCF has the correct REF and ALT","an insertion. So need to # get the nucleotide before","open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file)","new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] )","\".\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\",","by mummer. Need to do the appropriate # reverse (complement)","vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start]","so this is an insertion. So need to # get","file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list in sorted(vcf_records.items()): for record","if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name,","sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta,",") ) else: raise Exception(\"Unknown variant type: \" + str(variant))","\"\" commands = [ f\"nucmer --threads {threads} {maxmatch_opt} --delta {delta}","file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch","f_out: for snp_file in tmp_snp_files: with open(snp_file, \"rb\") as f_in:","-1 {delta} > {delta_1}\", f\"show-snps -rlTHC {delta_1} > {outfile}\", ]","\".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\",","for snp_file in tmp_snp_files: with open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in,","run these commands: # # nucmer --maxmatch --delta out.delta ref.fasta","# the docs inspecting that script, we need to run","# get the nucleotide before the insertion as well. new_record","variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] ) ) elif","{outfile}\", ] for command in commands: logging.info(\"Start run command: \"","# If the variant is reversed, it means that either","= vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1]","= pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta","== query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] ) if","tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as f_out: for snp_file in tmp_snp_files:","+ variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] ) )","This is instead of just running show-snps, which runs several","maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta}","the # reference. We're making VCF records w.r.t. the #","command: \" + command) subprocess.check_output(command, shell=True) logging.info(\"Finish run command: \"","new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\",","ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join( [","not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in","command in commands: logging.info(\"Start run command: \" + command) subprocess.check_output(command,","missing, compared to the # query. We're making VCF records","utils # We only want the .snps file from the","pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1),","import utils # We only want the .snps file from","ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base","split_ref=False, threads=1, maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta,","threads=1, maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file,","before the deletion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [","is an insertion. So need to # get the nucleotide","snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta =","import subprocess import pyfastaq import pymummer from cluster_vcf_records import vcf_record","\"--maxmatch\" if maxmatch else \"\" commands = [ f\"nucmer --threads","Writes a new VCF file unmerged records.\"\"\" vcf_records = {}","well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1),","to the # query. We're making VCF records w.r.t. the","dnadiff script from MUMmer. From reading # the docs inspecting","show-snps -rlTHC out.1delta > out.snps # # This is instead","records w.r.t. the # query, so this is a deletion.",") else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq","f\"show-snps -rlTHC {delta_1} > {outfile}\", ] for command in commands:","= pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\",","before the insertion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [","# nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1","threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta)","which runs several other commands # in addition to making","vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", variant.qry_base, variant.ref_base,","several other commands # in addition to making the snps","_run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if","tmp_snp_files: with open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in, f_out) if not","print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list in sorted(vcf_records.items()): for record in","these commands: # # nucmer --maxmatch --delta out.delta ref.fasta query.fasta","snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch )","[ variant.qry_name, str(variant.qry_start + 1), \".\", variant.qry_base, variant.ref_base, \".\", \".\",","f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta,","] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = []","script from MUMmer. From reading # the docs inspecting that","+ variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\",","the variant is reversed, it means that either the ref","print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list in sorted(vcf_records.items()): for","in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\", file=f)","pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If","VCF has the correct REF and ALT sequences if variant.reverse:","reversed, it means that either the ref or query had","else \"\" commands = [ f\"nucmer --threads {threads} {maxmatch_opt} --delta","the nucleotide before the insertion as well. new_record = vcf_record.VcfRecord(","variant is reversed, it means that either the ref or","do the appropriate # reverse (complement) fixes so the VCF","variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord(","to run these commands: # # nucmer --maxmatch --delta out.delta","tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as f: print(seq, file=f)","{delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\" if maxmatch else \"\" commands","unmerged records.\"\"\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs =","-rlTHC {delta_1} > {outfile}\", ] for command in commands: logging.info(\"Start","Exception(\"Unknown variant type: \" + str(variant)) assert ( new_record.REF ==","\"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base,","ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type","_run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files =","file from the dnadiff script from MUMmer. From reading #","\"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] ) ) else: raise Exception(\"Unknown variant","dnadiff. query_fasta = fasta file of query sequences. Writes a","in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values():","shell=True) maxmatch_opt = \"--maxmatch\" if maxmatch else \"\" commands =","with open(tmp_fasta, \"w\") as f: print(seq, file=f) snp_file = f\"{prefix}.snps\"","str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1],","variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants:","that either the ref or query had to be #","for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key,","variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record","- 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] ) )","--delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta","] for command in commands: logging.info(\"Start run command: \" +","import logging import os import shutil import subprocess import pyfastaq","_snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the .snps file made by dnadiff.","the insertion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name,","MUMmer. From reading # the docs inspecting that script, we","= f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True)","import attrgetter import logging import os import shutil import subprocess","when aligned by mummer. Need to do the appropriate #","pymummer from cluster_vcf_records import vcf_record from varifier import utils #","REF and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base)","delta-filter -1 out.delta > out.1delta # show-snps -rlTHC out.1delta >","pyfastaq import pymummer from cluster_vcf_records import vcf_record from varifier import","variant type: \" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][","sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base =","\".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] ) ) else: raise",") if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record)","pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta =","# reverse complemented when aligned by mummer. Need to do","# reverse (complement) fixes so the VCF has the correct",".snps file from the dnadiff script from MUMmer. From reading","> out.snps # # This is instead of just running",") os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \"wb\") as f_out: for snp_file","= vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", variant.qry_base,","if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for","the # query, so this is an insertion. So need","to # get the nucleotide before the insertion as well.","# # nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter","def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1","= f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta)","need to run these commands: # # nucmer --maxmatch --delta","\"\\t\".join( [ variant.qry_name, str(variant.qry_start + 1), \".\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] +","vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\", file=f) for","\" + command) subprocess.check_output(command, shell=True) logging.info(\"Finish run command: \" +","new VCF file unmerged records.\"\"\" vcf_records = {} variants =","variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ]","query. We're making VCF records w.r.t. the # query, so","# query, so this is a deletion. So need to","\"\"\"Loads the .snps file made by dnadiff. query_fasta = fasta","\".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] ) ) else: raise Exception(\"Unknown","query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the","\"1/1\", ] ) ) elif variant.var_type == pymummer.variant.INS: # The","outfile): \"\"\"Loads the .snps file made by dnadiff. query_fasta =","be # reverse complemented when aligned by mummer. Need to","appropriate # reverse (complement) fixes so the VCF has the","_run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with","# in addition to making the snps file. def _run_dnadiff_one_split(ref_fasta,","command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False,","] ) ) elif variant.var_type == pymummer.variant.INS: # The ref","nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta","making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True):","only want the .snps file from the dnadiff script from","fasta file of query sequences. Writes a new VCF file","1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] ) ) else:","reference. We're making VCF records w.r.t. the # query, so","the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta","query_fasta, outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\"","os import shutil import subprocess import pyfastaq import pymummer from","vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def","a deletion. So need to # get the nucleotide before","shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile):","for variant in variants: # If the variant is reversed,","snp_file in tmp_snp_files: with open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in, f_out)","inspecting that script, we need to run these commands: #","the appropriate # reverse (complement) fixes so the VCF has","import vcf_record from varifier import utils # We only want","\"GT\", \"1/1\", ] ) ) elif variant.var_type == pymummer.variant.INS: #","= ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \"\\t\".join(","= {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant","new_record.POS : new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not","ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record =","variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start -","( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ]","logging.info(\"Start run command: \" + command) subprocess.check_output(command, shell=True) logging.info(\"Finish run","in variants: # If the variant is reversed, it means","elif variant.var_type == pymummer.variant.INS: # The ref has sequence missing,","threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile,","query_fasta = fasta file of query sequences. Writes a new","+ command) subprocess.check_output(command, shell=True) logging.info(\"Finish run command: \" + command)","query had to be # reverse complemented when aligned by","shutil import subprocess import pyfastaq import pymummer from cluster_vcf_records import","vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start - 1] +","need to # get the nucleotide before the deletion as","varifier import utils # We only want the .snps file","= pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type ==","{ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta} > {delta_1}\", f\"show-snps -rlTHC {delta_1}","out.delta > out.1delta # show-snps -rlTHC out.1delta > out.snps #","vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\"))","is instead of just running show-snps, which runs several other","= [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile,","- 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\",","has sequence missing, compared to the # query. We're making","sequence missing, compared to the # query. We're making VCF","if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads the","the # query, so this is a deletion. So need","{delta} {delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\" if maxmatch else \"\"","Need to do the appropriate # reverse (complement) fixes so","new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\", query_seqs[variant.qry_name][variant.qry_start -","> {outfile}\", ] for command in commands: logging.info(\"Start run command:","query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ]","] ) ) else: raise Exception(\"Unknown variant type: \" +","in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False,","records w.r.t. the # query, so this is an insertion.","in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list in","reading # the docs inspecting that script, we need to","for record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta,","_run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 =","+ command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False,","outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split(","complemented when aligned by mummer. Need to do the appropriate","threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not debug: os.unlink(snps_file)","variant.var_type == pymummer.variant.INS: # The ref has sequence missing, compared","From reading # the docs inspecting that script, we need","docs inspecting that script, we need to run these commands:","= \"--maxmatch\" if maxmatch else \"\" commands = [ f\"nucmer","assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF)","the nucleotide before the deletion as well. new_record = vcf_record.VcfRecord(","pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base)","ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile)","command: \" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta,","= fasta file of query sequences. Writes a new VCF","query_fasta, outfile): \"\"\"Loads the .snps file made by dnadiff. query_fasta","\"GT\", \"1/1\", ] ) ) else: raise Exception(\"Unknown variant type:","vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\") as","file=f) for key, vcf_list in sorted(vcf_records.items()): for record in vcf_list:","operator import attrgetter import logging import os import shutil import","else: raise Exception(\"Unknown variant type: \" + str(variant)) assert (","query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query:","run command: \" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta,","VCF records w.r.t. the # query, so this is an","# query, so this is an insertion. So need to","] ) ) elif variant.var_type == pymummer.variant.DEL: # The query","maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads,","query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader","the ref or query had to be # reverse complemented","query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] )","So need to # get the nucleotide before the deletion","\"rb\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def","== pymummer.variant.DEL: # The query has sequence missing, compared to","as f: print(\"##fileformat=VCFv4.2\", file=f) for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f)","{delta_1}\", f\"show-snps -rlTHC {delta_1} > {outfile}\", ] for command in","commands: logging.info(\"Start run command: \" + command) subprocess.check_output(command, shell=True) logging.info(\"Finish","f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \"\"\"Loads","utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the variant is","variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \"\".join(reversed(qry_seq.seq)) ref_seq","else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in","subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\" if maxmatch","-f {delta} {delta_1}\", shell=True) maxmatch_opt = \"--maxmatch\" if maxmatch else","maxmatch_opt = \"--maxmatch\" if maxmatch else \"\" commands = [","running show-snps, which runs several other commands # in addition","maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for","== pymummer.variant.INS: # The ref has sequence missing, compared to","split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta,","records.\"\"\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta)","{query_fasta}\", f\"delta-filter -1 {delta} > {delta_1}\", f\"show-snps -rlTHC {delta_1} >","We only want the .snps file from the dnadiff script","# # This is instead of just running show-snps, which","aligned by mummer. Need to do the appropriate # reverse","the .snps file from the dnadiff script from MUMmer. From","in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta,","# The query has sequence missing, compared to the #","def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ):","variant.var_type == pymummer.variant.DEL: # The query has sequence missing, compared","[] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\"POS\")) with open(outfile, \"w\")","correct REF and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\",","the correct REF and ALT sequences if variant.reverse: qry_seq =","show-snps, which runs several other commands # in addition to","as f: print(seq, file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta,","as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file,","nucleotide before the deletion as well. new_record = vcf_record.VcfRecord( \"\\t\".join(","ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not","the dnadiff script from MUMmer. From reading # the docs","other commands # in addition to making the snps file.","nucleotide before the insertion as well. new_record = vcf_record.VcfRecord( \"\\t\".join(","= [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix","query_seqs[variant.qry_name][variant.qry_start - 1], \".\", \".\", \"SVTYPE=DNADIFF_DEL\", \"GT\", \"1/1\", ] )","for key, vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record,","out.1delta # show-snps -rlTHC out.1delta > out.snps # # This","outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff(","= utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the variant","from varifier import utils # We only want the .snps","maxmatch else \"\" commands = [ f\"nucmer --threads {threads} {maxmatch_opt}","(complement) fixes so the VCF has the correct REF and","The ref has sequence missing, compared to the # query.","out.delta ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta #","seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\"","from cluster_vcf_records import vcf_record from varifier import utils # We","f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm -f {delta} {delta_1}\", shell=True) maxmatch_opt","get the nucleotide before the insertion as well. new_record =","f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file)","+ 1), \".\", variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\",","want the .snps file from the dnadiff script from MUMmer.","ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta # show-snps","as f_out: for snp_file in tmp_snp_files: with open(snp_file, \"rb\") as","with open(outfile, \"wb\") as f_out: for snp_file in tmp_snp_files: with","import pyfastaq import pymummer from cluster_vcf_records import vcf_record from varifier","\".\", variant.qry_base, variant.ref_base, \".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] )","So need to # get the nucleotide before the insertion","deletion as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start),","compared to the # reference. We're making VCF records w.r.t.","reverse (complement) fixes so the VCF has the correct REF","): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug,","so this is a deletion. So need to # get","\"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] ) ) elif variant.var_type == pymummer.variant.DEL:","query, so this is a deletion. So need to #","seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for key, vcf_list","subprocess.check_output(command, shell=True) logging.info(\"Finish run command: \" + command) os.unlink(delta) os.unlink(delta_1)","commands: # # nucmer --maxmatch --delta out.delta ref.fasta query.fasta #","-rlTHC out.1delta > out.snps # # This is instead of","f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as f: print(seq, file=f) snp_file =","out.1delta > out.snps # # This is instead of just","print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1,","# show-snps -rlTHC out.1delta > out.snps # # This is","# This is instead of just running show-snps, which runs","\"1/1\", ] ) ) elif variant.var_type == pymummer.variant.DEL: # The","this is an insertion. So need to # get the","commands = [ f\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta}","ref has sequence missing, compared to the # query. We're","\"wb\") as f_out: for snp_file in tmp_snp_files: with open(snp_file, \"rb\")","to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1,","ref or query had to be # reverse complemented when",") elif variant.var_type == pymummer.variant.INS: # The ref has sequence","in tmp_snp_files: with open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in, f_out) if","and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\"x\", variant.qry_base) qry_seq.revcomp()","print(seq, file=f) snp_file = f\"{prefix}.snps\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads,","\"GT\", \"1/1\", ] ) ) elif variant.var_type == pymummer.variant.DEL: #","seq in seq_reader: prefix = f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with","the .snps file made by dnadiff. query_fasta = fasta file","with open(snp_file, \"rb\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug:","# We only want the .snps file from the dnadiff","[ f\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter","f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, )","either the ref or query had to be # reverse","file=f) for seq in query_seqs.values(): print(f\"##contig=<ID={seq.id},length={len(seq)}>\", file=f) print(\"#CHROM\\tPOS\\tID\\tREF\\tALT\\tQUAL\\tFILTER\\tINFO\\tFORMAT\\tsample\", file=f) for","query.fasta # delta-filter -1 out.delta > out.1delta # show-snps -rlTHC","it means that either the ref or query had to","pymummer.variant.DEL: # The query has sequence missing, compared to the","script, we need to run these commands: # # nucmer","of just running show-snps, which runs several other commands #","just running show-snps, which runs several other commands # in","# reference. We're making VCF records w.r.t. the # query,","\".\", \"SVTYPE=DNADIFF_INS\", \"GT\", \"1/1\", ] ) ) elif variant.var_type ==","\"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if","as well. new_record = vcf_record.VcfRecord( \"\\t\".join( [ variant.qry_name, str(variant.qry_start), \".\",","open(outfile, \"w\") as f: print(\"##fileformat=VCFv4.2\", file=f) for seq in query_seqs.values():","maxmatch=True, ): snps_file = f\"{outfile}.tmp.snps\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref,","make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file","{threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta} >","in addition to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta,","= f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as f: print(seq, file=f) snp_file","logging import os import shutil import subprocess import pyfastaq import","w.r.t. the # query, so this is a deletion. So","= f\"{outfile}.tmp.split.{len(tmp_snp_files)}\" tmp_fasta = f\"{prefix}.fasta\" with open(tmp_fasta, \"w\") as f:","{maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\", f\"delta-filter -1 {delta} > {delta_1}\",","= \"\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\"x\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq","split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not","\".\", \".\", \"SVTYPE=DNADIFF_SNP\", \"GT\", \"1/1\", ] ) ) elif variant.var_type"],"string":"[\n \"outfile, threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm\",\n \": new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in\",\n \"new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in vcf_records:\",\n \"os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True,\",\n \"addition to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile,\",\n \"this is a deletion. So need to # get the\",\n \"well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start\",\n \"\\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type ==\",\n \"logging.info(\\\"Finish run command: \\\" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff(\",\n \"variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif\",\n \"maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as f_out: for\",\n \"the VCF has the correct REF and ALT sequences if\",\n \"def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ):\",\n \"import shutil import subprocess import pyfastaq import pymummer from cluster_vcf_records\",\n \"--threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta}\",\n \"mummer. Need to do the appropriate # reverse (complement) fixes\",\n \"elif variant.var_type == pymummer.variant.DEL: # The query has sequence missing,\",\n \"from operator import attrgetter import logging import os import shutil\",\n \"outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader =\",\n \"a new VCF file unmerged records.\\\"\\\"\\\" vcf_records = {} variants\",\n \"raise Exception(\\\"Unknown variant type: \\\" + str(variant)) assert ( new_record.REF\",\n \"qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base\",\n \"delta = f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\",\",\n \"We're making VCF records w.r.t. the # query, so this\",\n \"for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as f:\",\n \"print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f)\",\n \"def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the .snps file made by\",\n \"\\\"1/1\\\", ] ) ) else: raise Exception(\\\"Unknown variant type: \\\"\",\n \"[ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start\",\n \"VCF file unmerged records.\\\"\\\"\\\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file)\",\n \"vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for\",\n \"snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if\",\n \"import os import shutil import subprocess import pyfastaq import pymummer\",\n \"vcf_record from varifier import utils # We only want the\",\n \"if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq))\",\n \"the deletion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name,\",\n \"vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with\",\n \"open(outfile, \\\"wb\\\") as f_out: for snp_file in tmp_snp_files: with open(snp_file,\",\n \"The query has sequence missing, compared to the # reference.\",\n \"f\\\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1\",\n \"f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\",\",\n \"ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = []\",\n \"query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] ) if new_record.CHROM\",\n \"> {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\", ] for command\",\n \"of query sequences. Writes a new VCF file unmerged records.\\\"\\\"\\\"\",\n \"variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base =\",\n \"+ str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS\",\n \"vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False,\",\n \"str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\",\",\n \"\\\" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile,\",\n \"f\\\"delta-filter -1 {delta} > {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\",\",\n \"{delta} > {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\", ] for\",\n \"threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f\",\n \"not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the .snps\",\n \"had to be # reverse complemented when aligned by mummer.\",\n \"= pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: #\",\n \"delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True) maxmatch_opt =\",\n \"[] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix =\",\n \"need to # get the nucleotide before the insertion as\",\n \"prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as\",\n \"\\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\",\",\n \"has the correct REF and ALT sequences if variant.reverse: qry_seq\",\n \"snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads,\",\n \"file unmerged records.\\\"\\\"\\\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs\",\n \"variants: # If the variant is reversed, it means that\",\n \"snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as\",\n \") ) elif variant.var_type == pymummer.variant.DEL: # The query has\",\n \"for command in commands: logging.info(\\\"Start run command: \\\" + command)\",\n \"str(variant.qry_start + 1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\",\",\n \"str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS +\",\n \"VCF records w.r.t. the # query, so this is a\",\n \"made by dnadiff. query_fasta = fasta file of query sequences.\",\n \"runs several other commands # in addition to making the\",\n \"command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command: \\\" + command) os.unlink(delta)\",\n \"query, so this is an insertion. So need to #\",\n \"variant.qry_name, str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\",\",\n \"truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\"\",\n \"missing, compared to the # reference. We're making VCF records\",\n \"If the variant is reversed, it means that either the\",\n \"[ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base,\",\n \"cluster_vcf_records import vcf_record from varifier import utils # We only\",\n \"record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile,\",\n \"-1 out.delta > out.1delta # show-snps -rlTHC out.1delta > out.snps\",\n \"w.r.t. the # query, so this is an insertion. So\",\n \"threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as f_out:\",\n \"from MUMmer. From reading # the docs inspecting that script,\",\n \"if maxmatch else \\\"\\\" commands = [ f\\\"nucmer --threads {threads}\",\n \"run command: \\\" + command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command:\",\n \"variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] ) )\",\n \"so the VCF has the correct REF and ALT sequences\",\n \"# delta-filter -1 out.delta > out.1delta # show-snps -rlTHC out.1delta\",\n \"in commands: logging.info(\\\"Start run command: \\\" + command) subprocess.check_output(command, shell=True)\",\n \"= vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start],\",\n \"file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\"\",\n \"tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\")\",\n \"file of query sequences. Writes a new VCF file unmerged\",\n \"query has sequence missing, compared to the # reference. We're\",\n \"means that either the ref or query had to be\",\n \"deletion. So need to # get the nucleotide before the\",\n \"reverse complemented when aligned by mummer. Need to do the\",\n \"query sequences. Writes a new VCF file unmerged records.\\\"\\\"\\\" vcf_records\",\n \"open(tmp_fasta, \\\"w\\\") as f: print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split(\",\n \"1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ]\",\n \"): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch\",\n \"type: \\\" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS\",\n \") ) elif variant.var_type == pymummer.variant.INS: # The ref has\",\n \"compared to the # query. We're making VCF records w.r.t.\",\n \"--delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta} > {delta_1}\\\", f\\\"show-snps\",\n \"to # get the nucleotide before the deletion as well.\",\n \"sequence missing, compared to the # reference. We're making VCF\",\n \"= f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch,\",\n \"shell=True) logging.info(\\\"Finish run command: \\\" + command) os.unlink(delta) os.unlink(delta_1) def\",\n \"the # query. We're making VCF records w.r.t. the #\",\n \"split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files\",\n \"subprocess import pyfastaq import pymummer from cluster_vcf_records import vcf_record from\",\n \"that script, we need to run these commands: # #\",\n \"# The ref has sequence missing, compared to the #\",\n \"debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta,\",\n \"insertion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start\",\n \"attrgetter import logging import os import shutil import subprocess import\",\n \"to the # reference. We're making VCF records w.r.t. the\",\n \"has sequence missing, compared to the # reference. We're making\",\n \"\\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\",\",\n \"qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq =\",\n \"insertion. So need to # get the nucleotide before the\",\n \"\\\"w\\\") as f: print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta,\",\n \"\\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.INS:\",\n \"\\\" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS :\",\n \"not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else:\",\n \"for seq in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\"\",\n \") elif variant.var_type == pymummer.variant.DEL: # The query has sequence\",\n \"ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file =\",\n \"vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq\",\n \"instead of just running show-snps, which runs several other commands\",\n \"if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch )\",\n \"os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as f_out: for snp_file in\",\n \"= [ f\\\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\",\",\n \"variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start\",\n \"== pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start +\",\n \"variant.qry_name, str(variant.qry_start + 1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\",\",\n \"making VCF records w.r.t. the # query, so this is\",\n \"_run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file,\",\n \"key, vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f)\",\n \"file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True,\",\n \"is a deletion. So need to # get the nucleotide\",\n \"+ len(new_record.REF) ] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM]\",\n \"or query had to be # reverse complemented when aligned\",\n \"{} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in\",\n \"# get the nucleotide before the deletion as well. new_record\",\n \"{delta_1} > {outfile}\\\", ] for command in commands: logging.info(\\\"Start run\",\n \"is reversed, it means that either the ref or query\",\n \"we need to run these commands: # # nucmer --maxmatch\",\n \"+ 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\",\",\n \"f: print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file,\",\n \"debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta,\",\n \"debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the .snps file\",\n \"to do the appropriate # reverse (complement) fixes so the\",\n \"get the nucleotide before the deletion as well. new_record =\",\n \".snps file made by dnadiff. query_fasta = fasta file of\",\n \"new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list\",\n \"by dnadiff. query_fasta = fasta file of query sequences. Writes\",\n \"sequences. Writes a new VCF file unmerged records.\\\"\\\"\\\" vcf_records =\",\n \"pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP:\",\n \"commands # in addition to making the snps file. def\",\n \"\\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type\",\n \"vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\",\",\n \"with open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in\",\n \") else: raise Exception(\\\"Unknown variant type: \\\" + str(variant)) assert\",\n \"len(new_record.REF) ] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] =\",\n \"variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp()\",\n \"to be # reverse complemented when aligned by mummer. Need\",\n \"1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\",\",\n \"as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start +\",\n \"# query. We're making VCF records w.r.t. the # query,\",\n \"the docs inspecting that script, we need to run these\",\n \"out.snps # # This is instead of just running show-snps,\",\n \"query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\",\",\n \"tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader:\",\n \"from the dnadiff script from MUMmer. From reading # the\",\n \"1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\",\",\n \"> out.1delta # show-snps -rlTHC out.1delta > out.snps # #\",\n \"--maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta >\",\n \"os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the .snps file made\",\n \"seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\")\",\n \"variant in variants: # If the variant is reversed, it\",\n \"pymummer.variant.INS: # The ref has sequence missing, compared to the\",\n \"f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\" if\",\n \"ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile,\",\n \"debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not debug:\",\n \"file made by dnadiff. query_fasta = fasta file of query\",\n \"f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as f: print(seq,\",\n \"= f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\"\",\n \"query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list in sorted(vcf_records.items()):\",\n \"os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1,\",\n \"in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def make_truth_vcf(\",\n \"truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta,\",\n \"{delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta} > {delta_1}\\\", f\\\"show-snps -rlTHC\",\n \"import pymummer from cluster_vcf_records import vcf_record from varifier import utils\",\n \"fixes so the VCF has the correct REF and ALT\",\n \"an insertion. So need to # get the nucleotide before\",\n \"open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file)\",\n \"new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] )\",\n \"\\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\",\",\n \"by mummer. Need to do the appropriate # reverse (complement)\",\n \"vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start]\",\n \"so this is an insertion. So need to # get\",\n \"file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list in sorted(vcf_records.items()): for record\",\n \"if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name,\",\n \"sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta,\",\n \") ) else: raise Exception(\\\"Unknown variant type: \\\" + str(variant))\",\n \"\\\"\\\" commands = [ f\\\"nucmer --threads {threads} {maxmatch_opt} --delta {delta}\",\n \"file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch\",\n \"f_out: for snp_file in tmp_snp_files: with open(snp_file, \\\"rb\\\") as f_in:\",\n \"-1 {delta} > {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\", ]\",\n \"\\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\",\",\n \"for snp_file in tmp_snp_files: with open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in,\",\n \"run these commands: # # nucmer --maxmatch --delta out.delta ref.fasta\",\n \"# the docs inspecting that script, we need to run\",\n \"# get the nucleotide before the insertion as well. new_record\",\n \"variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif\",\n \"{outfile}\\\", ] for command in commands: logging.info(\\\"Start run command: \\\"\",\n \"# If the variant is reversed, it means that either\",\n \"= vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1]\",\n \"= pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta\",\n \"== query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ] ) if\",\n \"tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as f_out: for snp_file in tmp_snp_files:\",\n \"+ variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] ) )\",\n \"This is instead of just running show-snps, which runs several\",\n \"maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta}\",\n \"the # reference. We're making VCF records w.r.t. the #\",\n \"command: \\\" + command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command: \\\"\",\n \"new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\",\",\n \"ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [\",\n \"not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in\",\n \"command in commands: logging.info(\\\"Start run command: \\\" + command) subprocess.check_output(command,\",\n \"missing, compared to the # query. We're making VCF records\",\n \"utils # We only want the .snps file from the\",\n \"pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1),\",\n \"import utils # We only want the .snps file from\",\n \"ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base\",\n \"split_ref=False, threads=1, maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta,\",\n \"threads=1, maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file,\",\n \"before the deletion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [\",\n \"is an insertion. So need to # get the nucleotide\",\n \"snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta =\",\n \"import subprocess import pyfastaq import pymummer from cluster_vcf_records import vcf_record\",\n \"\\\"--maxmatch\\\" if maxmatch else \\\"\\\" commands = [ f\\\"nucmer --threads\",\n \"Writes a new VCF file unmerged records.\\\"\\\"\\\" vcf_records = {}\",\n \"well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1),\",\n \"to the # query. We're making VCF records w.r.t. the\",\n \"dnadiff script from MUMmer. From reading # the docs inspecting\",\n \"show-snps -rlTHC out.1delta > out.snps # # This is instead\",\n \"records w.r.t. the # query, so this is a deletion.\",\n \") else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq\",\n \"f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\", ] for command in commands:\",\n \"= pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\",\",\n \"before the insertion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [\",\n \"# nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1\",\n \"threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta)\",\n \"which runs several other commands # in addition to making\",\n \"vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", variant.qry_base, variant.ref_base,\",\n \"several other commands # in addition to making the snps\",\n \"_run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if\",\n \"tmp_snp_files: with open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out) if not\",\n \"print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list in sorted(vcf_records.items()): for record in\",\n \"these commands: # # nucmer --maxmatch --delta out.delta ref.fasta query.fasta\",\n \"snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch )\",\n \"[ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\",\",\n \"f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta,\",\n \"] ) if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = []\",\n \"script from MUMmer. From reading # the docs inspecting that\",\n \"+ variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\",\",\n \"the variant is reversed, it means that either the ref\",\n \"print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list in sorted(vcf_records.items()): for\",\n \"in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f)\",\n \"pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If\",\n \"VCF has the correct REF and ALT sequences if variant.reverse:\",\n \"reversed, it means that either the ref or query had\",\n \"else \\\"\\\" commands = [ f\\\"nucmer --threads {threads} {maxmatch_opt} --delta\",\n \"the nucleotide before the insertion as well. new_record = vcf_record.VcfRecord(\",\n \"variant is reversed, it means that either the ref or\",\n \"do the appropriate # reverse (complement) fixes so the VCF\",\n \"variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord(\",\n \"to run these commands: # # nucmer --maxmatch --delta out.delta\",\n \"tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as f: print(seq, file=f)\",\n \"{delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\" if maxmatch else \\\"\\\" commands\",\n \"unmerged records.\\\"\\\"\\\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs =\",\n \"-rlTHC {delta_1} > {outfile}\\\", ] for command in commands: logging.info(\\\"Start\",\n \"Exception(\\\"Unknown variant type: \\\" + str(variant)) assert ( new_record.REF ==\",\n \"\\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base,\",\n \"ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type\",\n \"_run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files =\",\n \"file from the dnadiff script from MUMmer. From reading #\",\n \"\\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) else: raise Exception(\\\"Unknown variant\",\n \"dnadiff. query_fasta = fasta file of query sequences. Writes a\",\n \"in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values():\",\n \"shell=True) maxmatch_opt = \\\"--maxmatch\\\" if maxmatch else \\\"\\\" commands =\",\n \"with open(tmp_fasta, \\\"w\\\") as f: print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\"\",\n \"str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1],\",\n \"variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants:\",\n \"that either the ref or query had to be #\",\n \"for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key,\",\n \"variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record\",\n \"- 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] ) )\",\n \"--delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta\",\n \"] for command in commands: logging.info(\\\"Start run command: \\\" +\",\n \"import logging import os import shutil import subprocess import pyfastaq\",\n \"_snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the .snps file made by dnadiff.\",\n \"the insertion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name,\",\n \"MUMmer. From reading # the docs inspecting that script, we\",\n \"= f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True)\",\n \"import attrgetter import logging import os import shutil import subprocess\",\n \"when aligned by mummer. Need to do the appropriate #\",\n \"pymummer from cluster_vcf_records import vcf_record from varifier import utils #\",\n \"REF and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base)\",\n \"delta-filter -1 out.delta > out.1delta # show-snps -rlTHC out.1delta >\",\n \"pyfastaq import pymummer from cluster_vcf_records import vcf_record from varifier import\",\n \"variant type: \\\" + str(variant)) assert ( new_record.REF == query_seqs[new_record.CHROM][\",\n \"sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base =\",\n \"\\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) else: raise\",\n \") if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record)\",\n \"pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta =\",\n \"# reverse complemented when aligned by mummer. Need to do\",\n \"# reverse (complement) fixes so the VCF has the correct\",\n \".snps file from the dnadiff script from MUMmer. From reading\",\n \"> out.snps # # This is instead of just running\",\n \") os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with open(outfile, \\\"wb\\\") as f_out: for snp_file\",\n \"= vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", variant.qry_base,\",\n \"if new_record.CHROM not in vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for\",\n \"the # query, so this is an insertion. So need\",\n \"to # get the nucleotide before the insertion as well.\",\n \"# # nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter\",\n \"def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1\",\n \"= f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta)\",\n \"need to run these commands: # # nucmer --maxmatch --delta\",\n \"\\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start + 1), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] +\",\n \"vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for\",\n \"\\\" + command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command: \\\" +\",\n \"new VCF file unmerged records.\\\"\\\"\\\" vcf_records = {} variants =\",\n \"variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ]\",\n \"query. We're making VCF records w.r.t. the # query, so\",\n \"# query, so this is a deletion. So need to\",\n \"\\\"\\\"\\\"Loads the .snps file made by dnadiff. query_fasta = fasta\",\n \"\\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) else: raise Exception(\\\"Unknown\",\n \"query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the\",\n \"\\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.INS: # The\",\n \"outfile): \\\"\\\"\\\"Loads the .snps file made by dnadiff. query_fasta =\",\n \"be # reverse complemented when aligned by mummer. Need to\",\n \"appropriate # reverse (complement) fixes so the VCF has the\",\n \"_run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file) with\",\n \"# in addition to making the snps file. def _run_dnadiff_one_split(ref_fasta,\",\n \"command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False,\",\n \"] ) ) elif variant.var_type == pymummer.variant.INS: # The ref\",\n \"nucmer --maxmatch --delta out.delta ref.fasta query.fasta # delta-filter -1 out.delta\",\n \"making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True):\",\n \"only want the .snps file from the dnadiff script from\",\n \"fasta file of query sequences. Writes a new VCF file\",\n \"1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) else:\",\n \"reference. We're making VCF records w.r.t. the # query, so\",\n \"the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta\",\n \"query_fasta, outfile, threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\"\",\n \"os import shutil import subprocess import pyfastaq import pymummer from\",\n \"vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record, file=f) def\",\n \"a deletion. So need to # get the nucleotide before\",\n \"shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile):\",\n \"for variant in variants: # If the variant is reversed,\",\n \"snp_file in tmp_snp_files: with open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out)\",\n \"inspecting that script, we need to run these commands: #\",\n \"the appropriate # reverse (complement) fixes so the VCF has\",\n \"import vcf_record from varifier import utils # We only want\",\n \"\\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.INS: #\",\n \"= ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join(\",\n \"= {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta) for variant\",\n \"new_record.POS : new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not\",\n \"ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type == pymummer.variant.SNP: new_record =\",\n \"variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start -\",\n \"( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF) ]\",\n \"logging.info(\\\"Start run command: \\\" + command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run\",\n \"in variants: # If the variant is reversed, it means\",\n \"elif variant.var_type == pymummer.variant.INS: # The ref has sequence missing,\",\n \"threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile,\",\n \"query_fasta = fasta file of query sequences. Writes a new\",\n \"+ command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command: \\\" + command)\",\n \"query had to be # reverse complemented when aligned by\",\n \"shutil import subprocess import pyfastaq import pymummer from cluster_vcf_records import\",\n \"vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start - 1] +\",\n \"need to # get the nucleotide before the deletion as\",\n \"varifier import utils # We only want the .snps file\",\n \"= pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if variant.var_type ==\",\n \"{ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta} > {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1}\",\n \"out.delta > out.1delta # show-snps -rlTHC out.1delta > out.snps #\",\n \"vcf_records: vcf_records[new_record.CHROM] = [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\"))\",\n \"is instead of just running show-snps, which runs several other\",\n \"= [] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile,\",\n \"- 1] + variant.qry_base, query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\",\",\n \"has sequence missing, compared to the # query. We're making\",\n \"sequence missing, compared to the # query. We're making VCF\",\n \"if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads the\",\n \"the # query, so this is a deletion. So need\",\n \"{delta} {delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\" if maxmatch else \\\"\\\"\",\n \"Need to do the appropriate # reverse (complement) fixes so\",\n \"new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\", query_seqs[variant.qry_name][variant.qry_start -\",\n \"> {outfile}\\\", ] for command in commands: logging.info(\\\"Start run command:\",\n \"query_seqs[variant.qry_name][variant.qry_start], query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ]\",\n \"] ) ) else: raise Exception(\\\"Unknown variant type: \\\" +\",\n \"in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False,\",\n \"records w.r.t. the # query, so this is an insertion.\",\n \"in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list in\",\n \"reading # the docs inspecting that script, we need to\",\n \"for record in vcf_list: print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta,\",\n \"_run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1, maxmatch=True): delta = f\\\"{outfile}.tmp.delta\\\" delta_1 =\",\n \"+ command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False,\",\n \"outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split(\",\n \"complemented when aligned by mummer. Need to do the appropriate\",\n \"threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not debug: os.unlink(snps_file)\",\n \"variant.var_type == pymummer.variant.INS: # The ref has sequence missing, compared\",\n \"From reading # the docs inspecting that script, we need\",\n \"docs inspecting that script, we need to run these commands:\",\n \"= \\\"--maxmatch\\\" if maxmatch else \\\"\\\" commands = [ f\\\"nucmer\",\n \"assert ( new_record.REF == query_seqs[new_record.CHROM][ new_record.POS : new_record.POS + len(new_record.REF)\",\n \"the nucleotide before the deletion as well. new_record = vcf_record.VcfRecord(\",\n \"pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base)\",\n \"ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile)\",\n \"command: \\\" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta, query_fasta,\",\n \"= fasta file of query sequences. Writes a new VCF\",\n \"query_fasta, outfile): \\\"\\\"\\\"Loads the .snps file made by dnadiff. query_fasta\",\n \"\\\"GT\\\", \\\"1/1\\\", ] ) ) else: raise Exception(\\\"Unknown variant type:\",\n \"vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\") as\",\n \"file=f) for key, vcf_list in sorted(vcf_records.items()): for record in vcf_list:\",\n \"operator import attrgetter import logging import os import shutil import\",\n \"else: raise Exception(\\\"Unknown variant type: \\\" + str(variant)) assert (\",\n \"query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query:\",\n \"run command: \\\" + command) os.unlink(delta) os.unlink(delta_1) def _run_dnadiff( ref_fasta,\",\n \"VCF records w.r.t. the # query, so this is an\",\n \"# query, so this is an insertion. So need to\",\n \"] ) ) elif variant.var_type == pymummer.variant.DEL: # The query\",\n \"maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta, query_fasta, outfile, threads=threads,\",\n \"query_fasta, outfile, threads=threads, maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader\",\n \"the ref or query had to be # reverse complemented\",\n \"query_seqs[variant.qry_name][variant.qry_start] + variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] )\",\n \"So need to # get the nucleotide before the deletion\",\n \"\\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def\",\n \"== pymummer.variant.DEL: # The query has sequence missing, compared to\",\n \"as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f)\",\n \"{delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} > {outfile}\\\", ] for command in\",\n \"commands: logging.info(\\\"Start run command: \\\" + command) subprocess.check_output(command, shell=True) logging.info(\\\"Finish\",\n \"f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file, query_fasta, outfile): \\\"\\\"\\\"Loads\",\n \"utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the variant is\",\n \"variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp() variant.qry_base = \\\"\\\".join(reversed(qry_seq.seq)) ref_seq\",\n \"else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in\",\n \"subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\" if maxmatch\",\n \"-f {delta} {delta_1}\\\", shell=True) maxmatch_opt = \\\"--maxmatch\\\" if maxmatch else\",\n \"maxmatch_opt = \\\"--maxmatch\\\" if maxmatch else \\\"\\\" commands = [\",\n \"running show-snps, which runs several other commands # in addition\",\n \"maxmatch=maxmatch ) else: tmp_snp_files = [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for\",\n \"== pymummer.variant.INS: # The ref has sequence missing, compared to\",\n \"split_query=False, debug=False, threads=1, maxmatch=True, ): if not split_query: _run_dnadiff_one_split( ref_fasta,\",\n \"records.\\\"\\\"\\\" vcf_records = {} variants = pymummer.snp_file.get_all_variants(snps_file) query_seqs = utils.file_to_dict_of_seqs(query_fasta)\",\n \"{query_fasta}\\\", f\\\"delta-filter -1 {delta} > {delta_1}\\\", f\\\"show-snps -rlTHC {delta_1} >\",\n \"We only want the .snps file from the dnadiff script\",\n \"# # This is instead of just running show-snps, which\",\n \"aligned by mummer. Need to do the appropriate # reverse\",\n \"the .snps file from the dnadiff script from MUMmer. From\",\n \"in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta,\",\n \"# The query has sequence missing, compared to the #\",\n \"def _run_dnadiff( ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ):\",\n \"variant.var_type == pymummer.variant.DEL: # The query has sequence missing, compared\",\n \"[] vcf_records[new_record.CHROM].append(new_record) for vcf_list in vcf_records.values(): vcf_list.sort(key=attrgetter(\\\"POS\\\")) with open(outfile, \\\"w\\\")\",\n \"correct REF and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\",\",\n \"the correct REF and ALT sequences if variant.reverse: qry_seq =\",\n \"show-snps, which runs several other commands # in addition to\",\n \"as f: print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta,\",\n \"as f_in: shutil.copyfileobj(f_in, f_out) if not debug: os.unlink(snp_file) def _snps_file_to_vcf(snps_file,\",\n \"nucleotide before the deletion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join(\",\n \"ref_fasta, query_fasta, outfile, split_query=False, debug=False, threads=1, maxmatch=True, ): if not\",\n \"the dnadiff script from MUMmer. From reading # the docs\",\n \"other commands # in addition to making the snps file.\",\n \"nucleotide before the insertion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join(\",\n \"= [] seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix\",\n \"query_seqs[variant.qry_name][variant.qry_start - 1], \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_DEL\\\", \\\"GT\\\", \\\"1/1\\\", ] )\",\n \"for key, vcf_list in sorted(vcf_records.items()): for record in vcf_list: print(record,\",\n \"out.1delta # show-snps -rlTHC out.1delta > out.snps # # This\",\n \"outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff(\",\n \"= utils.file_to_dict_of_seqs(query_fasta) for variant in variants: # If the variant\",\n \"from varifier import utils # We only want the .snps\",\n \"maxmatch else \\\"\\\" commands = [ f\\\"nucmer --threads {threads} {maxmatch_opt}\",\n \"(complement) fixes so the VCF has the correct REF and\",\n \"The ref has sequence missing, compared to the # query.\",\n \"out.delta ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta #\",\n \"seq_reader = pyfastaq.sequences.file_reader(query_fasta) for seq in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\"\",\n \"from cluster_vcf_records import vcf_record from varifier import utils # We\",\n \"f\\\"{outfile}.tmp.delta\\\" delta_1 = f\\\"{outfile}.tmp.1delta\\\" subprocess.check_output(f\\\"rm -f {delta} {delta_1}\\\", shell=True) maxmatch_opt\",\n \"get the nucleotide before the insertion as well. new_record =\",\n \"f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads, maxmatch=maxmatch ) os.unlink(tmp_fasta) tmp_snp_files.append(snp_file)\",\n \"+ 1), \\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\",\",\n \"want the .snps file from the dnadiff script from MUMmer.\",\n \"ref.fasta query.fasta # delta-filter -1 out.delta > out.1delta # show-snps\",\n \"as f_out: for snp_file in tmp_snp_files: with open(snp_file, \\\"rb\\\") as\",\n \"with open(outfile, \\\"wb\\\") as f_out: for snp_file in tmp_snp_files: with\",\n \"import pyfastaq import pymummer from cluster_vcf_records import vcf_record from varifier\",\n \"\\\".\\\", variant.qry_base, variant.ref_base, \\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] )\",\n \"So need to # get the nucleotide before the insertion\",\n \"deletion as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start),\",\n \"compared to the # reference. We're making VCF records w.r.t.\",\n \"reverse (complement) fixes so the VCF has the correct REF\",\n \"): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug,\",\n \"so this is a deletion. So need to # get\",\n \"\\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.DEL:\",\n \"query, so this is a deletion. So need to #\",\n \"seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for key, vcf_list\",\n \"subprocess.check_output(command, shell=True) logging.info(\\\"Finish run command: \\\" + command) os.unlink(delta) os.unlink(delta_1)\",\n \"commands: # # nucmer --maxmatch --delta out.delta ref.fasta query.fasta #\",\n \"-rlTHC out.1delta > out.snps # # This is instead of\",\n \"f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as f: print(seq, file=f) snp_file =\",\n \"out.1delta > out.snps # # This is instead of just\",\n \"print(record, file=f) def make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1,\",\n \"# show-snps -rlTHC out.1delta > out.snps # # This is\",\n \"# This is instead of just running show-snps, which runs\",\n \"\\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.DEL: # The\",\n \"this is an insertion. So need to # get the\",\n \"commands = [ f\\\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta}\",\n \"ref has sequence missing, compared to the # query. We're\",\n \"\\\"wb\\\") as f_out: for snp_file in tmp_snp_files: with open(snp_file, \\\"rb\\\")\",\n \"to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta, outfile, threads=1,\",\n \"ref or query had to be # reverse complemented when\",\n \") elif variant.var_type == pymummer.variant.INS: # The ref has sequence\",\n \"in tmp_snp_files: with open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out) if\",\n \"and ALT sequences if variant.reverse: qry_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.qry_base) qry_seq.revcomp()\",\n \"print(seq, file=f) snp_file = f\\\"{prefix}.snps\\\" _run_dnadiff_one_split( ref_fasta, tmp_fasta, snp_file, threads=threads,\",\n \"\\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type == pymummer.variant.DEL: #\",\n \"seq in seq_reader: prefix = f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with\",\n \"the .snps file made by dnadiff. query_fasta = fasta file\",\n \"with open(snp_file, \\\"rb\\\") as f_in: shutil.copyfileobj(f_in, f_out) if not debug:\",\n \"# We only want the .snps file from the dnadiff\",\n \"[ f\\\"nucmer --threads {threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter\",\n \"f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, )\",\n \"either the ref or query had to be # reverse\",\n \"file=f) for seq in query_seqs.values(): print(f\\\"##contig=<ID={seq.id},length={len(seq)}>\\\", file=f) print(\\\"#CHROM\\\\tPOS\\\\tID\\\\tREF\\\\tALT\\\\tQUAL\\\\tFILTER\\\\tINFO\\\\tFORMAT\\\\tsample\\\", file=f) for\",\n \"query.fasta # delta-filter -1 out.delta > out.1delta # show-snps -rlTHC\",\n \"it means that either the ref or query had to\",\n \"pymummer.variant.DEL: # The query has sequence missing, compared to the\",\n \"script, we need to run these commands: # # nucmer\",\n \"of just running show-snps, which runs several other commands #\",\n \"just running show-snps, which runs several other commands # in\",\n \"# reference. We're making VCF records w.r.t. the # query,\",\n \"\\\".\\\", \\\"SVTYPE=DNADIFF_INS\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type ==\",\n \"\\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq if\",\n \"as well. new_record = vcf_record.VcfRecord( \\\"\\\\t\\\".join( [ variant.qry_name, str(variant.qry_start), \\\".\\\",\",\n \"open(outfile, \\\"w\\\") as f: print(\\\"##fileformat=VCFv4.2\\\", file=f) for seq in query_seqs.values():\",\n \"maxmatch=True, ): snps_file = f\\\"{outfile}.tmp.snps\\\" _run_dnadiff( truth_fasta, ref_fasta, snps_file, split_query=split_ref,\",\n \"make_truth_vcf( ref_fasta, truth_fasta, outfile, debug=False, split_ref=False, threads=1, maxmatch=True, ): snps_file\",\n \"{threads} {maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta} >\",\n \"in addition to making the snps file. def _run_dnadiff_one_split(ref_fasta, query_fasta,\",\n \"= f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as f: print(seq, file=f) snp_file\",\n \"logging import os import shutil import subprocess import pyfastaq import\",\n \"w.r.t. the # query, so this is a deletion. So\",\n \"= f\\\"{outfile}.tmp.split.{len(tmp_snp_files)}\\\" tmp_fasta = f\\\"{prefix}.fasta\\\" with open(tmp_fasta, \\\"w\\\") as f:\",\n \"{maxmatch_opt} --delta {delta} {ref_fasta} {query_fasta}\\\", f\\\"delta-filter -1 {delta} > {delta_1}\\\",\",\n \"= \\\"\\\".join(reversed(qry_seq.seq)) ref_seq = pyfastaq.sequences.Fasta(\\\"x\\\", variant.ref_base) ref_seq.revcomp() variant.ref_base = ref_seq.seq\",\n \"split_query=split_ref, debug=debug, threads=threads, maxmatch=maxmatch, ) _snps_file_to_vcf(snps_file, ref_fasta, outfile) if not\",\n \"\\\".\\\", \\\".\\\", \\\"SVTYPE=DNADIFF_SNP\\\", \\\"GT\\\", \\\"1/1\\\", ] ) ) elif variant.var_type\"\n]","isConversational":false}}},{"rowIdx":697,"cells":{"ngram":{"kind":"list like","value":["-> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str","model for various reports\"\"\" @classmethod def key(cls: Type) -> str:","def to_table(self) -> List[str]: return [ 'Pull requests:', '--------------', 'Open","+ [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count:","old_count: int def to_table(self) -> List[str]: return [ 'Pull requests:',","return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True)","@dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel):","int closed_count: int old_count: int def to_table(self) -> List[str]: return","'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class","for various reports\"\"\" @classmethod def key(cls: Type) -> str: name","def to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open Closed","-> str: name = cls.__name__ return name[0].lower() + name[1:] def","Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int","class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self)","key(cls: Type) -> str: name = cls.__name__ return name[0].lower() +","(' ' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for","contributors: List[Contributor] def to_table(self) -> List[str]: return [ 'Most active","import asdict, dataclass from typing import Any, Dict, List, Type","class Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors:","int old_count: int def to_table(self) -> List[str]: return [ 'Pull","-> List[str]: return [ 'Issues:', '-------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}'","f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count:","@dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return","name[0].lower() + name[1:] def to_table(self) -> List[str]: raise NotImplementedError def","Type) -> str: name = cls.__name__ return name[0].lower() + name[1:]","+ (' ' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}'","asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True) class","def key(cls: Type) -> str: name = cls.__name__ return name[0].lower()","in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count:","+ 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True)","for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count:","List[str]: return [ 'Issues:', '-------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ]","cls.__name__ return name[0].lower() + name[1:] def to_table(self) -> List[str]: raise","from dataclasses import asdict, dataclass from typing import Any, Dict,","class StatsBaseModel: \"\"\"Base model for various reports\"\"\" @classmethod def key(cls:","int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]:","contributors:', '-------------------------', 'Name' + (' ' * 20) + 'Commits',","\"\"\"Base model for various reports\"\"\" @classmethod def key(cls: Type) ->","+ name[1:] def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self)","return name[0].lower() + name[1:] def to_table(self) -> List[str]: raise NotImplementedError","'Name' + (' ' * 20) + 'Commits', ] +","[f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int","reports\"\"\" @classmethod def key(cls: Type) -> str: name = cls.__name__","@dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def","'--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count:","int def to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open","Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count:","from typing import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel:","] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int","StatsBaseModel: \"\"\"Base model for various reports\"\"\" @classmethod def key(cls: Type)","various reports\"\"\" @classmethod def key(cls: Type) -> str: name =","PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self) ->","20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors]","IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self) ->","name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def","Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model for various reports\"\"\" @classmethod","to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name:","name[1:] def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) ->","Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int","str: name = cls.__name__ return name[0].lower() + name[1:] def to_table(self)","' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c","return [ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ]","@dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model for various reports\"\"\" @classmethod def","Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model for various","class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self)","int def to_table(self) -> List[str]: return [ 'Pull requests:', '--------------',","asdict, dataclass from typing import Any, Dict, List, Type @dataclass(frozen=True)","import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model","to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open Closed Old',","def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor:","Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor]","ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return [ 'Most","'-------------------------', 'Name' + (' ' * 20) + 'Commits', ]","dataclasses import asdict, dataclass from typing import Any, Dict, List,","name = cls.__name__ return name[0].lower() + name[1:] def to_table(self) ->","= cls.__name__ return name[0].lower() + name[1:] def to_table(self) -> List[str]:","old_count: int def to_table(self) -> List[str]: return [ 'Issues:', '-------',","[ 'Most active contributors:', '-------------------------', 'Name' + (' ' *","open_count: int closed_count: int old_count: int def to_table(self) -> List[str]:","NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class","List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self)","-> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return","to_table(self) -> List[str]: return [ 'Pull requests:', '--------------', 'Open Closed","-> List[str]: return [ 'Pull requests:', '--------------', 'Open Closed Old',","<gh_stars>0 from dataclasses import asdict, dataclass from typing import Any,","closed_count: int old_count: int def to_table(self) -> List[str]: return [","List, Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model for various reports\"\"\"","int old_count: int def to_table(self) -> List[str]: return [ 'Issues:',","self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int","active contributors:', '-------------------------', 'Name' + (' ' * 20) +","def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str,","@classmethod def key(cls: Type) -> str: name = cls.__name__ return","raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True)","requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel):","c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int","class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return [","return [ 'Most active contributors:', '-------------------------', 'Name' + (' '","List[str]: return [ 'Most active contributors:', '-------------------------', 'Name' + ('","commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) ->","dataclass from typing import Any, Dict, List, Type @dataclass(frozen=True) class","@dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def","str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self)","List[str]: return [ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}'","to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]:","-> List[str]: return [ 'Most active contributors:', '-------------------------', 'Name' +","to_table(self) -> List[str]: return [ 'Most active contributors:', '-------------------------', 'Name'","Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base model for","* 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in","typing import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \"\"\"Base","'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int","Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count:","'Most active contributors:', '-------------------------', 'Name' + (' ' * 20)","[ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True)","] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel):","List[Contributor] def to_table(self) -> List[str]: return [ 'Most active contributors:',","def to_table(self) -> List[str]: return [ 'Most active contributors:', '-------------------------',","'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class"],"string":"[\n \"-> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str\",\n \"model for various reports\\\"\\\"\\\" @classmethod def key(cls: Type) -> str:\",\n \"def to_table(self) -> List[str]: return [ 'Pull requests:', '--------------', 'Open\",\n \"+ [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count:\",\n \"old_count: int def to_table(self) -> List[str]: return [ 'Pull requests:',\",\n \"return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True)\",\n \"@dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel):\",\n \"int closed_count: int old_count: int def to_table(self) -> List[str]: return\",\n \"'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class\",\n \"for various reports\\\"\\\"\\\" @classmethod def key(cls: Type) -> str: name\",\n \"def to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open Closed\",\n \"-> str: name = cls.__name__ return name[0].lower() + name[1:] def\",\n \"Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int\",\n \"class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self)\",\n \"key(cls: Type) -> str: name = cls.__name__ return name[0].lower() +\",\n \"(' ' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for\",\n \"contributors: List[Contributor] def to_table(self) -> List[str]: return [ 'Most active\",\n \"import asdict, dataclass from typing import Any, Dict, List, Type\",\n \"class Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors:\",\n \"int old_count: int def to_table(self) -> List[str]: return [ 'Pull\",\n \"-> List[str]: return [ 'Issues:', '-------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}'\",\n \"f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count:\",\n \"@dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return\",\n \"name[0].lower() + name[1:] def to_table(self) -> List[str]: raise NotImplementedError def\",\n \"Type) -> str: name = cls.__name__ return name[0].lower() + name[1:]\",\n \"+ (' ' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}'\",\n \"asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count: int @dataclass(frozen=True) class\",\n \"def key(cls: Type) -> str: name = cls.__name__ return name[0].lower()\",\n \"in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count:\",\n \"+ 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True)\",\n \"for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count:\",\n \"List[str]: return [ 'Issues:', '-------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ]\",\n \"cls.__name__ return name[0].lower() + name[1:] def to_table(self) -> List[str]: raise\",\n \"from dataclasses import asdict, dataclass from typing import Any, Dict,\",\n \"class StatsBaseModel: \\\"\\\"\\\"Base model for various reports\\\"\\\"\\\" @classmethod def key(cls:\",\n \"int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]:\",\n \"contributors:', '-------------------------', 'Name' + (' ' * 20) + 'Commits',\",\n \"\\\"\\\"\\\"Base model for various reports\\\"\\\"\\\" @classmethod def key(cls: Type) ->\",\n \"+ name[1:] def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self)\",\n \"return name[0].lower() + name[1:] def to_table(self) -> List[str]: raise NotImplementedError\",\n \"'Name' + (' ' * 20) + 'Commits', ] +\",\n \"[f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int\",\n \"reports\\\"\\\"\\\" @classmethod def key(cls: Type) -> str: name = cls.__name__\",\n \"@dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def\",\n \"'--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count:\",\n \"int def to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open\",\n \"Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count:\",\n \"from typing import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel:\",\n \"] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int\",\n \"StatsBaseModel: \\\"\\\"\\\"Base model for various reports\\\"\\\"\\\" @classmethod def key(cls: Type)\",\n \"various reports\\\"\\\"\\\" @classmethod def key(cls: Type) -> str: name =\",\n \"PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self) ->\",\n \"20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors]\",\n \"IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self) ->\",\n \"name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def\",\n \"Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model for various reports\\\"\\\"\\\" @classmethod\",\n \"to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name:\",\n \"name[1:] def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) ->\",\n \"Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int\",\n \"str: name = cls.__name__ return name[0].lower() + name[1:] def to_table(self)\",\n \"' * 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c\",\n \"return [ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ]\",\n \"@dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model for various reports\\\"\\\"\\\" @classmethod def\",\n \"Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model for various\",\n \"class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def to_table(self)\",\n \"int def to_table(self) -> List[str]: return [ 'Pull requests:', '--------------',\",\n \"asdict, dataclass from typing import Any, Dict, List, Type @dataclass(frozen=True)\",\n \"import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model\",\n \"to_table(self) -> List[str]: return [ 'Issues:', '-------', 'Open Closed Old',\",\n \"def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor:\",\n \"Contributor: name: str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor]\",\n \"ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return [ 'Most\",\n \"'-------------------------', 'Name' + (' ' * 20) + 'Commits', ]\",\n \"dataclasses import asdict, dataclass from typing import Any, Dict, List,\",\n \"name = cls.__name__ return name[0].lower() + name[1:] def to_table(self) ->\",\n \"= cls.__name__ return name[0].lower() + name[1:] def to_table(self) -> List[str]:\",\n \"old_count: int def to_table(self) -> List[str]: return [ 'Issues:', '-------',\",\n \"[ 'Most active contributors:', '-------------------------', 'Name' + (' ' *\",\n \"open_count: int closed_count: int old_count: int def to_table(self) -> List[str]:\",\n \"NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class\",\n \"List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self)\",\n \"-> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return\",\n \"to_table(self) -> List[str]: return [ 'Pull requests:', '--------------', 'Open Closed\",\n \"-> List[str]: return [ 'Pull requests:', '--------------', 'Open Closed Old',\",\n \"<gh_stars>0 from dataclasses import asdict, dataclass from typing import Any,\",\n \"closed_count: int old_count: int def to_table(self) -> List[str]: return [\",\n \"List, Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model for various reports\\\"\\\"\\\"\",\n \"int old_count: int def to_table(self) -> List[str]: return [ 'Issues:',\",\n \"self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int old_count: int\",\n \"active contributors:', '-------------------------', 'Name' + (' ' * 20) +\",\n \"def to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str,\",\n \"@classmethod def key(cls: Type) -> str: name = cls.__name__ return\",\n \"raise NotImplementedError def to_dict(self) -> Dict[str, Any]: return asdict(self) @dataclass(frozen=True)\",\n \"requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel):\",\n \"c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel): open_count: int closed_count: int\",\n \"class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) -> List[str]: return [\",\n \"return [ 'Most active contributors:', '-------------------------', 'Name' + (' '\",\n \"List[str]: return [ 'Most active contributors:', '-------------------------', 'Name' + ('\",\n \"commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self) ->\",\n \"dataclass from typing import Any, Dict, List, Type @dataclass(frozen=True) class\",\n \"@dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int closed_count: int old_count: int def\",\n \"str commit_count: int @dataclass(frozen=True) class ContributorStats(StatsBaseModel): contributors: List[Contributor] def to_table(self)\",\n \"List[str]: return [ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}'\",\n \"to_table(self) -> List[str]: raise NotImplementedError def to_dict(self) -> Dict[str, Any]:\",\n \"-> List[str]: return [ 'Most active contributors:', '-------------------------', 'Name' +\",\n \"to_table(self) -> List[str]: return [ 'Most active contributors:', '-------------------------', 'Name'\",\n \"Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base model for\",\n \"* 20) + 'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in\",\n \"typing import Any, Dict, List, Type @dataclass(frozen=True) class StatsBaseModel: \\\"\\\"\\\"Base\",\n \"'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True) class IssueStats(StatsBaseModel): open_count: int\",\n \"Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str commit_count:\",\n \"'Most active contributors:', '-------------------------', 'Name' + (' ' * 20)\",\n \"[ 'Pull requests:', '--------------', 'Open Closed Old', f'{str(self.open_count).ljust(8)}{str(self.closed_count).ljust(8)}{str(self.old_count).ljust(8)}' ] @dataclass(frozen=True)\",\n \"] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class PullRequestStats(StatsBaseModel):\",\n \"List[Contributor] def to_table(self) -> List[str]: return [ 'Most active contributors:',\",\n \"def to_table(self) -> List[str]: return [ 'Most active contributors:', '-------------------------',\",\n \"'Commits', ] + [f'{c.name.ljust(24)}{c.commit_count}' for c in self.contributors] @dataclass(frozen=True) class\"\n]","isConversational":false}}},{"rowIdx":698,"cells":{"ngram":{"kind":"list like","value":["datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string","any((self.start, self.end)): return None query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)]","if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc)","= self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else:","DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\"start\")) @property def end(self):","start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if not end_date_str:","FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1)","datetime_string else: to_compare = parse(datetime_string) if not self.start and not","= datetime_string else: to_compare = parse(datetime_string) if not self.start and","@property def end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if not end_datetime: return","def query_params(self): if not any((self.start, self.end)): return None query_params =","False if self.end and (self.end < to_compare): return False return","= datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str:","dicts)) @property def query_params(self): if not any((self.start, self.end)): return None","0 and \\ value.second == 0 and \\ value.microsecond ==","return True return list(filter(in_range, dicts)) @property def query_params(self): if not","return False if self.start and (to_compare < self.start): return False","get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if not end_datetime:","= datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour == 0","if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime","not self.end: return False if self.start and (to_compare < self.start):","else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def","self.start and (to_compare < self.start): return False if self.end and","datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return","from __future__ import absolute_import import datetime from dateutil import parser",".queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime')","datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property","return value.hour == 0 and \\ value.minute == 0 and","datetime_data def _has_no_time_info(value): return value.hour == 0 and \\ value.minute","queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if","datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None return parse(start_date_str)","None return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None","== 0 and \\ value.minute == 0 and \\ value.second","return query_params def _do_django_query(self, queryset): query_params = self.query_params if query_params:","DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\"start\")) @property def","_has_no_time_info(value): return value.hour == 0 and \\ value.minute == 0","else: to_compare = parse(datetime_string) if not self.start and not self.end:","and (to_compare < self.start): return False if self.end and (self.end","return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not","import absolute_import import datetime from dateutil import parser import pytz","(to_compare < self.start): return False if self.end and (self.end <","if self.end and (self.end < to_compare): return False return True","not start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if not","end_date_str: return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def","def _has_no_time_info(value): return value.hour == 0 and \\ value.minute ==","False return True return list(filter(in_range, dicts)) @property def query_params(self): if","__future__ import absolute_import import datetime from dateutil import parser import","and (self.end < to_compare): return False return True return list(filter(in_range,","return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str):","query_params def _do_django_query(self, queryset): query_params = self.query_params if query_params: return","return get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if not","def get_end(end_date_str): if not end_date_str: return None return parse(end_date_str) def","value.hour == 0 and \\ value.minute == 0 and \\","self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare =","if not start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if","return list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start, self.end)):","query_params(self): if not any((self.start, self.end)): return None query_params = dict()","datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string) if not","< to_compare): return False return True return list(filter(in_range, dicts)) @property","= self.end return query_params def _do_django_query(self, queryset): query_params = self.query_params","def _do_django_query(self, queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params)","import datetime from dateutil import parser import pytz from .base","to_compare): return False return True return list(filter(in_range, dicts)) @property def","list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start, self.end)): return","return None query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start","parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo:","WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin,","end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts):","+ WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts): def","end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def","and not self.end: return False if self.start and (to_compare <","= get_end(self.filter_args.get(\"end\")) if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime","0 and \\ value.minute == 0 and \\ value.second ==","self.end)): return None query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] =","def start(self): return get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\"end\"))","ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string =","from dateutil import parser import pytz from .base import FieldFilter,","pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import","self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime =","and \\ value.minute == 0 and \\ value.second == 0","return None return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return","return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime =","_do_django_query(self, queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params) else:","import parser import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin","return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc)","dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime):","if not self.start and not self.end: return False if self.start","self.start): return False if self.end and (self.end < to_compare): return","min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not","queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc)","= dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end: query_params[\"{}__lte\".format(self.field_name)]","not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime +","self.end and (self.end < to_compare): return False return True return","def end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if not end_datetime: return None","if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value):","True return list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start,","get_end(self.filter_args.get(\"end\")) if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime =","None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND","on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string,","query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset): query_params =","= parse(datetime_string) if not self.start and not self.end: return False","not end_date_str: return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string))","def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data","@property def start(self): return get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime =","value.minute == 0 and \\ value.second == 0 and \\","datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour","def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare","DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND =","not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return","return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None return","def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data","end_datetime = get_end(self.filter_args.get(\"end\")) if not end_datetime: return None if _has_no_time_info(end_datetime):","if not any((self.start, self.end)): return None query_params = dict() if","not self.start and not self.end: return False if self.start and","parse(datetime_string) if not self.start and not self.end: return False if","return False return True return list(filter(in_range, dicts)) @property def query_params(self):","isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string) if",".base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY","self.start and not self.end: return False if self.start and (to_compare","if not end_date_str: return None return parse(end_date_str) def parse(datetime_string): return","ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def","if self.start and (to_compare < self.start): return False if self.end","= datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self):","False if self.start and (to_compare < self.start): return False if","datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return","= self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params def","self.end: return False if self.start and (to_compare < self.start): return","query_params = self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none()","= end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self,","query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end:","if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return","get_start(start_date_str): if not start_date_str: return None return parse(start_date_str) def get_end(end_date_str):","< self.start): return False if self.end and (self.end < to_compare):","return datetime_data def _has_no_time_info(value): return value.hour == 0 and \\","import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter","@property def query_params(self): if not any((self.start, self.end)): return None query_params","make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return","import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY =","not any((self.start, self.end)): return None query_params = dict() if self.start:","end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY","QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin,","= datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter):","absolute_import import datetime from dateutil import parser import pytz from","return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data):","self.end return query_params def _do_django_query(self, queryset): query_params = self.query_params if","WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum):","def get_start(start_date_str): if not start_date_str: return None return parse(start_date_str) def","@QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\"start\"))","datetime from dateutil import parser import pytz from .base import","dateutil import parser import pytz from .base import FieldFilter, DictFilterMixin,","datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour ==","DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND","None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if","datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour == 0 and","get_end(end_date_str): if not end_date_str: return None return parse(end_date_str) def parse(datetime_string):","from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1)","self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare","dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] =","None query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if","\\ value.minute == 0 and \\ value.second == 0 and","= self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime","(self.end < to_compare): return False return True return list(filter(in_range, dicts))","self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return","= datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None return","end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name)","if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string)","end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if not end_datetime: return None if","in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare =","== 0 and \\ value.second == 0 and \\ value.microsecond","FieldFilter): @property def start(self): return get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime","return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum,","queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params) else: return","return False if self.end and (self.end < to_compare): return False","import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class","if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset):","return None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY -","self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset): query_params","- ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string","from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter","max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None","and \\ value.second == 0 and \\ value.microsecond == 0","self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params def _do_django_query(self,","class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\"start\")) @property","to_compare = datetime_string else: to_compare = parse(datetime_string) if not self.start","parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data =","start(self): return get_start(self.filter_args.get(\"start\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\"end\")) if","if self.start: query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end","def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if","query_params[\"{}__gte\".format(self.field_name)] = self.start if self.end: query_params[\"{}__lte\".format(self.field_name)] = self.end return query_params","parser import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from","query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime","parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None return parse(end_date_str)","to_compare = parse(datetime_string) if not self.start and not self.end: return","_has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime","make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def"],"string":"[\n \"datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string\",\n \"any((self.start, self.end)): return None query_params = dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)]\",\n \"if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc)\",\n \"= self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else:\",\n \"DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self):\",\n \"start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if not end_date_str:\",\n \"FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1)\",\n \"datetime_string else: to_compare = parse(datetime_string) if not self.start and not\",\n \"= datetime_string else: to_compare = parse(datetime_string) if not self.start and\",\n \"@property def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return\",\n \"def query_params(self): if not any((self.start, self.end)): return None query_params =\",\n \"False if self.end and (self.end < to_compare): return False return\",\n \"= datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str:\",\n \"dicts)) @property def query_params(self): if not any((self.start, self.end)): return None\",\n \"0 and \\\\ value.second == 0 and \\\\ value.microsecond ==\",\n \"return True return list(filter(in_range, dicts)) @property def query_params(self): if not\",\n \"return False if self.start and (to_compare < self.start): return False\",\n \"get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime:\",\n \"= datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour == 0\",\n \"if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime\",\n \"not self.end: return False if self.start and (to_compare < self.start):\",\n \"else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def\",\n \"self.start and (to_compare < self.start): return False if self.end and\",\n \"datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return\",\n \"from __future__ import absolute_import import datetime from dateutil import parser\",\n \".queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime')\",\n \"datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property\",\n \"return value.hour == 0 and \\\\ value.minute == 0 and\",\n \"datetime_data def _has_no_time_info(value): return value.hour == 0 and \\\\ value.minute\",\n \"queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if\",\n \"datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None return parse(start_date_str)\",\n \"None return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None\",\n \"== 0 and \\\\ value.minute == 0 and \\\\ value.second\",\n \"return query_params def _do_django_query(self, queryset): query_params = self.query_params if query_params:\",\n \"DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def\",\n \"_has_no_time_info(value): return value.hour == 0 and \\\\ value.minute == 0\",\n \"else: to_compare = parse(datetime_string) if not self.start and not self.end:\",\n \"and (to_compare < self.start): return False if self.end and (self.end\",\n \"return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not\",\n \"import absolute_import import datetime from dateutil import parser import pytz\",\n \"(to_compare < self.start): return False if self.end and (self.end <\",\n \"if self.end and (self.end < to_compare): return False return True\",\n \"not start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if not\",\n \"end_date_str: return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def\",\n \"def _has_no_time_info(value): return value.hour == 0 and \\\\ value.minute ==\",\n \"False return True return list(filter(in_range, dicts)) @property def query_params(self): if\",\n \"__future__ import absolute_import import datetime from dateutil import parser import\",\n \"and (self.end < to_compare): return False return True return list(filter(in_range,\",\n \"return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str):\",\n \"query_params def _do_django_query(self, queryset): query_params = self.query_params if query_params: return\",\n \"return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not\",\n \"def get_end(end_date_str): if not end_date_str: return None return parse(end_date_str) def\",\n \"value.hour == 0 and \\\\ value.minute == 0 and \\\\\",\n \"self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare =\",\n \"if not start_date_str: return None return parse(start_date_str) def get_end(end_date_str): if\",\n \"return list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start, self.end)):\",\n \"query_params(self): if not any((self.start, self.end)): return None query_params = dict()\",\n \"datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string) if not\",\n \"< to_compare): return False return True return list(filter(in_range, dicts)) @property\",\n \"= self.end return query_params def _do_django_query(self, queryset): query_params = self.query_params\",\n \"def _do_django_query(self, queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params)\",\n \"import datetime from dateutil import parser import pytz from .base\",\n \"to_compare): return False return True return list(filter(in_range, dicts)) @property def\",\n \"list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start, self.end)): return\",\n \"return None query_params = dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start\",\n \"parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo:\",\n \"WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin,\",\n \"end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts):\",\n \"+ WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts): def\",\n \"end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def\",\n \"and not self.end: return False if self.start and (to_compare <\",\n \"= get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime\",\n \"0 and \\\\ value.minute == 0 and \\\\ value.second ==\",\n \"self.end)): return None query_params = dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] =\",\n \"def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\"))\",\n \"ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string =\",\n \"from dateutil import parser import pytz from .base import FieldFilter,\",\n \"pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import\",\n \"self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime =\",\n \"and \\\\ value.minute == 0 and \\\\ value.second == 0\",\n \"return None return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return\",\n \"return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime =\",\n \"_do_django_query(self, queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params) else:\",\n \"import parser import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin\",\n \"return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc)\",\n \"dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime):\",\n \"if not self.start and not self.end: return False if self.start\",\n \"self.start): return False if self.end and (self.end < to_compare): return\",\n \"min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not\",\n \"queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc)\",\n \"= dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)]\",\n \"not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime +\",\n \"self.end and (self.end < to_compare): return False return True return\",\n \"def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return None\",\n \"if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value):\",\n \"True return list(filter(in_range, dicts)) @property def query_params(self): if not any((self.start,\",\n \"get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return None if _has_no_time_info(end_datetime): end_datetime =\",\n \"None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND\",\n \"on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string,\",\n \"query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset): query_params =\",\n \"= parse(datetime_string) if not self.start and not self.end: return False\",\n \"not end_date_str: return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string))\",\n \"def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data\",\n \"@property def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime =\",\n \"value.minute == 0 and \\\\ value.second == 0 and \\\\\",\n \"datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour\",\n \"def in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare\",\n \"DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND =\",\n \"not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return\",\n \"return parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None return\",\n \"def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data\",\n \"end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return None if _has_no_time_info(end_datetime):\",\n \"if not any((self.start, self.end)): return None query_params = dict() if\",\n \"not self.start and not self.end: return False if self.start and\",\n \"parse(datetime_string) if not self.start and not self.end: return False if\",\n \"return False return True return list(filter(in_range, dicts)) @property def query_params(self):\",\n \"isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string) if\",\n \".base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY\",\n \"self.start and not self.end: return False if self.start and (to_compare\",\n \"if not end_date_str: return None return parse(end_date_str) def parse(datetime_string): return\",\n \"ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def\",\n \"if self.start and (to_compare < self.start): return False if self.end\",\n \"= datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self):\",\n \"False if self.start and (to_compare < self.start): return False if\",\n \"datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return\",\n \"= self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params def\",\n \"self.end: return False if self.start and (to_compare < self.start): return\",\n \"query_params = self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none()\",\n \"= end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self,\",\n \"query_params = dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end:\",\n \"if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return\",\n \"get_start(start_date_str): if not start_date_str: return None return parse(start_date_str) def get_end(end_date_str):\",\n \"< self.start): return False if self.end and (self.end < to_compare):\",\n \"return datetime_data def _has_no_time_info(value): return value.hour == 0 and \\\\\",\n \"import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter\",\n \"@property def query_params(self): if not any((self.start, self.end)): return None query_params\",\n \"make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return\",\n \"import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY =\",\n \"not any((self.start, self.end)): return None query_params = dict() if self.start:\",\n \"end_datetime: return None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY\",\n \"QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin,\",\n \"= datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter):\",\n \"absolute_import import datetime from dateutil import parser import pytz from\",\n \"return None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data):\",\n \"self.end return query_params def _do_django_query(self, queryset): query_params = self.query_params if\",\n \"WHOLE_DAY - ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum):\",\n \"def get_start(start_date_str): if not start_date_str: return None return parse(start_date_str) def\",\n \"@QueryFilter.register_type_condition('datetime') class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\\\"start\\\"))\",\n \"datetime from dateutil import parser import pytz from .base import\",\n \"dateutil import parser import pytz from .base import FieldFilter, DictFilterMixin,\",\n \"datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour ==\",\n \"DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND\",\n \"None return parse(end_date_str) def parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if\",\n \"datetime_data.replace(tzinfo=pytz.utc) return datetime_data def _has_no_time_info(value): return value.hour == 0 and\",\n \"get_end(end_date_str): if not end_date_str: return None return parse(end_date_str) def parse(datetime_string):\",\n \"from .queryfilter import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1)\",\n \"self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare\",\n \"dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] =\",\n \"None query_params = dict() if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if\",\n \"\\\\ value.minute == 0 and \\\\ value.second == 0 and\",\n \"= self.query_params if query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime\",\n \"(self.end < to_compare): return False return True return list(filter(in_range, dicts))\",\n \"self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return\",\n \"= datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None return\",\n \"end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name)\",\n \"if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string else: to_compare = parse(datetime_string)\",\n \"end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if not end_datetime: return None if\",\n \"in_range(datum): datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare =\",\n \"== 0 and \\\\ value.second == 0 and \\\\ value.microsecond\",\n \"FieldFilter): @property def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime\",\n \"return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum,\",\n \"queryset): query_params = self.query_params if query_params: return queryset.filter(**query_params) else: return\",\n \"return False if self.end and (self.end < to_compare): return False\",\n \"import QueryFilter WHOLE_DAY = datetime.timedelta(days=1) ONE_SECOND = datetime.timedelta(seconds=1) @QueryFilter.register_type_condition('datetime') class\",\n \"if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset):\",\n \"return None if _has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY -\",\n \"self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params def _do_django_query(self, queryset): query_params\",\n \"- ONE_SECOND return end_datetime def on_dicts(self, dicts): def in_range(datum): datetime_string\",\n \"from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from .queryfilter import QueryFilter\",\n \"max_datetime = datetime.datetime.max.replace(tzinfo=pytz.utc) def get_start(start_date_str): if not start_date_str: return None\",\n \"and \\\\ value.second == 0 and \\\\ value.microsecond == 0\",\n \"self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params def _do_django_query(self,\",\n \"class DatetimeRangeFilter(DjangoQueryFilterMixin, DictFilterMixin, FieldFilter): @property def start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property\",\n \"to_compare = datetime_string else: to_compare = parse(datetime_string) if not self.start\",\n \"parse(datetime_string): return make_time_aware(parser.parse(datetime_string)) def make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data =\",\n \"start(self): return get_start(self.filter_args.get(\\\"start\\\")) @property def end(self): end_datetime = get_end(self.filter_args.get(\\\"end\\\")) if\",\n \"if self.start: query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end\",\n \"def on_dicts(self, dicts): def in_range(datum): datetime_string = self.get(datum, self.field_name) if\",\n \"query_params[\\\"{}__gte\\\".format(self.field_name)] = self.start if self.end: query_params[\\\"{}__lte\\\".format(self.field_name)] = self.end return query_params\",\n \"parser import pytz from .base import FieldFilter, DictFilterMixin, DjangoQueryFilterMixin from\",\n \"query_params: return queryset.filter(**query_params) else: return queryset.none() min_datetime = datetime.datetime.min.replace(tzinfo=pytz.utc) max_datetime\",\n \"parse(start_date_str) def get_end(end_date_str): if not end_date_str: return None return parse(end_date_str)\",\n \"to_compare = parse(datetime_string) if not self.start and not self.end: return\",\n \"_has_no_time_info(end_datetime): end_datetime = end_datetime + WHOLE_DAY - ONE_SECOND return end_datetime\",\n \"make_time_aware(datetime_data): if not datetime_data.tzinfo: datetime_data = datetime_data.replace(tzinfo=pytz.utc) return datetime_data def\"\n]","isConversational":false}}},{"rowIdx":699,"cells":{"ngram":{"kind":"list like","value":["= reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy","diff * diff / beta, diff - 0.5 * beta)","0 diff = torch.abs(pred - target) loss = torch.where(diff <","torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] -","with reduction=\"sum\"') return loss def reduce_loss(loss, reduction): \"\"\"Reduce loss compute.","= partial(func, **kwargs) if kwargs else func map_results = map(pfunc,","= [] for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h,","# This program is distributed in the hope that it","use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy loss. :param desc: config","'sum') reduction = ( reduction_override if reduction_override else self.reduction) loss_cls","be used with reduction=\"sum\"') return loss def reduce_loss(loss, reduction): \"\"\"Reduce","[x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore","inds.numel() > 0: bin_labels[inds, labels[inds] - 1] = 1 if","= weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label,","beta > 0 assert pred.size() == target.size() and target.numel() >","= x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas =","x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets,","WITHOUT ANY WARRANTY; without even the implied warranty of #","the MIT License. # This program is distributed in the","= img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w =","loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta","\"\"\" loss = F.cross_entropy(pred, label, reduction='none') if weight is not","labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel()","self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module):","License for more details. \"\"\"Import all torch operators.\"\"\" import torch.nn.functional","License. # This program is distributed in the hope that","self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides","= x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights,","Ltd. All rights reserved. # This program is free software;","label and label weights \"\"\" bin_labels = labels.new_full((labels.size(0), label_channels), 0)","\"\"\"Cross Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init","weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss =","self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x):","Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3,","self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = []","import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget","multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def","return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight","num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice","= [] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i])","Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights reserved.","if weight is not None: loss = loss * weight","super(RpnLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score.\"\"\" cls_scores =","label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss","None num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device)","reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred,","beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross",":param avg_factor: avg factor :return: loss \"\"\" if weight is","function :param avg_factor: avg factor :return: loss \"\"\" if weight","warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.","weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce loss. :param loss: losses","if target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred,","anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)]","weight: weight :param avg_factor: avg factor :param reduction_override: reduce override","1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean',","losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,","avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \"\"\"Mask","for img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for i","if reduction == 'mean': loss = loss.sum() / avg_factor elif","= x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class","the # MIT License for more details. \"\"\"Import all torch","\"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg,","featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h /","= target_means or (.0, .0, .0, .0) self.target_stds = target_stds","1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1,","[8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0]","gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores]",":return: loss \"\"\" assert beta > 0 assert pred.size() ==","diff / beta, diff - 0.5 * beta) return loss","bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred,","inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1)","weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy losses. :param pred: predict result","= x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class","= binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function =","class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__()","torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1 loss. :param","3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return","is None: loss = reduce_loss(loss, reduction) else: if reduction ==","avg_factor elif reduction != 'none': raise ValueError('avg_factor can not be","override :return: loss \"\"\" reduction = ( reduction_override if reduction_override","'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler':","False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \"\"\"Create","valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling)","range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for","return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary","= [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas,","None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \"\"\"Create anchor.\"\"\"","beta :return: loss \"\"\" assert beta > 0 assert pred.size()","len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels): anchors =","*args, **kwargs): \"\"\"Multi apply. :param func: function :param args: args","3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets","bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def","self.out_channels = out_channels def forward(self, x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred,","= x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores,","2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples)","and label weights \"\"\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds","anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list = []","- 1] = 1 if label_weights is None: bin_label_weights =","avg_factor=None): \"\"\"Weight reduce loss. :param loss: losses :param weight: weight","\"\"\"Reduce loss compute. :param loss: losses :param reduction: reduce funtion","'mean' and avg_factor is None num_rois = pred.size()[0] inds =","else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr':","from functools import partial import numpy as np from six.moves","0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug':","def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means =","= CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \"\"\"Get x.\"\"\"","return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets,","cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score,","bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def","feat_h, feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h","\"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction","factor :param reduction_override: reduce override :return: loss \"\"\" reduction =","avg factor :return: loss \"\"\" loss = F.cross_entropy(pred, label, reduction='none')","= labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if","label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is","loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target,","[0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8, 16,","self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def","self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def forward(self,","else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores,","(None, 'none', 'mean', 'sum') reduction = ( reduction_override if reduction_override","loss = torch.where(diff < beta, 0.5 * diff * diff","= labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3,","= label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label,","== 'mean': loss = loss.sum() / avg_factor elif reduction !=","l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction","'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def","1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls,","bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights,","loss def reduce_loss(loss, reduction): \"\"\"Reduce loss compute. :param loss: losses","avg_factor=None): \"\"\"Mask cross entropy loss. :param pred: predict result :param","weight is not None: loss = loss * weight if","forward(self, x): \"\"\"Get cls score and bbox preds.\"\"\" cls_scores =","bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore","\"\"\"Get loss.\"\"\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score =","2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples)","loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator)","loss_weight=1.0): \"\"\"Init Cross Entropy loss. :param desc: config dict \"\"\"","target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0,","torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from","= F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction,","cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels","label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls =","[] for img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for","1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else 1",":param reduction_override: reduce override :return: loss \"\"\" reduction = (","weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none')","anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5,","not None: loss = loss * weight if avg_factor is","@NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss,","function :param args: args of function :return: result \"\"\" pfunc","bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross","return loss def reduce_loss(loss, reduction): \"\"\"Reduce loss compute. :param loss:","Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross","even the implied warranty of # MERCHANTABILITY or FITNESS FOR",":param target: target :param label: gt label :param reduction: reduce","32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None","smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction =","loss def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary labels. :param labels:","x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module):","predict :param target: target :param beta: beta :return: loss \"\"\"",":param pred: predict result :param label: gt label :param weight:","is not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred,","the implied warranty of # MERCHANTABILITY or FITNESS FOR A","for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w =","gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return","= num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name':","target :param label: gt label :param reduction: reduce function :param","it under the terms of the MIT License. # This",":param func: function :param args: args of function :return: result","else: if reduction == 'mean': loss = loss.sum() / avg_factor","loss \"\"\" if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label,","= reduction self.loss_weight = loss_weight def forward(self, pred, target, weight=None,","target: target :param weight: weight :param avg_factor: avg factor :param","anchors according to feature map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None,","reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy loss. :param desc: config dict","pfunc = partial(func, **kwargs) if kwargs else func map_results =","factor :return: loss \"\"\" if weight is not None: loss","avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute. :param pred: predict :param target:","import partial import numpy as np from six.moves import map,","= torch.where(diff < beta, 0.5 * diff * diff /","sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg","reduce loss. :param loss: losses :param weight: weight :param reduction:","can redistribute it and/or modify # it under the terms","CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \"\"\"Get x.\"\"\" (cls_score,","or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls","= [multi_level_anchors for _ in range(num_imgs)] valid_flag_list = [] for","anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs):","gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \"\"\"Multi apply. :param func: function","weight: weight :param reduction: reduce function :param avg_factor: avg factor","gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \"\"\"Multi apply. :param","self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function","* weight if avg_factor is None: loss = reduce_loss(loss, reduction)","beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss",":return: loss \"\"\" loss = F.cross_entropy(pred, label, reduction='none') if weight","target) loss = torch.where(diff < beta, 0.5 * diff *","num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0, .0,","__init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score.\"\"\" cls_scores","\"\"\"Binary cross entropy loss. :param pred: predict result :param label:","bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn","gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator)","img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for i in","bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return","= cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels,","import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator","reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight,","reserved. # This program is free software; you can redistribute","cross entropy loss. :param pred: predict result :param target: target","@NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False,","avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\"","result :param target: target :param label: gt label :param reduction:","weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross entropy loss. :param pred: predict","factor :return: loss \"\"\" loss = F.cross_entropy(pred, label, reduction='none') if","= label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls","labels[inds] - 1] = 1 if label_weights is None: bin_label_weights","'mean': loss = loss.sum() / avg_factor elif reduction != 'none':","else func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator)","reduction: reduce funtion :return: loss \"\"\" reduction_function = F._Reduction.get_enum(reduction) if","loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction,",":return: binary label and label weights \"\"\" bin_labels = labels.new_full((labels.size(0),","= x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss","1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg","x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x","reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score,","weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def","4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3,","_ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w","and bbox preds.\"\"\" cls_scores = x[0] bbox_preds = x[1] return","x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list,","self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights =","reduction == 'mean' and avg_factor is None num_rois = pred.size()[0]","label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score,",":param pred: predict :param target: target :param beta: beta :return:","'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border':","label channels :return: binary label and label weights \"\"\" bin_labels","if weight is not None: weight = weight.float() loss =","cross entropy loss. :param pred: predict result :param label: gt","img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg).","reduction='none') if weight is not None: weight = weight.float() loss","bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels =","return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def __init__(self,","return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def","import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes","None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels)","valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list,","= Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou':","class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__() def","NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from","= bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0,","avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred","**kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override in (None, 'none', 'mean', 'sum')","return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor",":param reduction: reduce function :param avg_factor: avg factor :return: loss","\"\"\"Rpn loss input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x):","= bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets,","device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def",":return: loss \"\"\" if pred.dim() != label.dim(): label, weight =","# it under the terms of the MIT License. #","\"\"\" assert beta > 0 assert pred.size() == target.size() and","reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce loss. :param","result \"\"\" pfunc = partial(func, **kwargs) if kwargs else func","anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16,","img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w","-*- coding: utf-8 -*- # Copyright (C) 2020. Huawei Technologies","or [4, 8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides)","self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function =","bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross","'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp,","featmap in cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes:","def forward(self, x): \"\"\"Get cls score and bbox preds.\"\"\" cls_scores","None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights","loss \"\"\" reduction = ( reduction_override if reduction_override else self.reduction)","weight :param avg_factor: avg factor :param reduction_override: reduce override :return:","__init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means","out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas,","for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list","predict :param target: target :param weight: weight :param avg_factor: avg","- 0.5 * beta) return loss def cross_entropy(pred, label, weight=None,","\"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes","This program is free software; you can redistribute it and/or","* diff / beta, diff - 0.5 * beta) return","reduction): \"\"\"Reduce loss compute. :param loss: losses :param reduction: reduce","labels :param label_weights: label weights :param label_channels: label channels :return:","beta self.reduction = reduction self.loss_weight = loss_weight def forward(self, pred,","rights reserved. # This program is free software; you can","return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy","self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight","gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors according to","num_total_samples): \"\"\"Get loss.\"\"\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score","> 0 diff = torch.abs(pred - target) loss = torch.where(diff","self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]),","= None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels,","# -*- coding: utf-8 -*- # Copyright (C) 2020. Huawei","0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name':","= use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight =","= len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels): anchors","0: return loss elif reduction_function == 1: return loss.mean() elif","\"\"\"Cross entropy losses. :param pred: predict result :param label: gt","if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self,","weight if avg_factor is None: loss = reduce_loss(loss, reduction) else:","\"\"\" pfunc = partial(func, **kwargs) if kwargs else func map_results","CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0):","labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls,","loss.\"\"\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0,","beta=1.0): \"\"\"Smooth l1 loss. :param pred: predict :param target: target","= torch.abs(pred - target) loss = torch.where(diff < beta, 0.5","tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def __init__(self,","func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class","self.anchor_strides = anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes","16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is","self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg","else self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight *","target_means or (.0, .0, .0, .0) self.target_stds = target_stds or","from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator","**kwargs) if kwargs else func map_results = map(pfunc, *args) return","label_weights, label_channels): \"\"\"Expand binary labels. :param labels: labels :param label_weights:","used with reduction=\"sum\"') return loss def reduce_loss(loss, reduction): \"\"\"Reduce loss","if reduction_override else self.reduction) if target.numel() > 0: loss_bbox =","pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean',","x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds,","avg_factor: avg factor :return: loss \"\"\" if weight is not","from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from functools","(C) 2020. Huawei Technologies Co., Ltd. All rights reserved. #","more details. \"\"\"Import all torch operators.\"\"\" import torch.nn.functional as F","(1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else","bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg,","numpy as np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss","in range(num_imgs)] valid_flag_list = [] for img_id, img_meta in enumerate(img_metas):",":param args: args of function :return: result \"\"\" pfunc =","and/or modify # it under the terms of the MIT","@NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean',","assert beta > 0 assert pred.size() == target.size() and target.numel()","avg factor :return: loss \"\"\" assert reduction == 'mean' and","from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target","0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0:","labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2,","anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales","binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy","modify # it under the terms of the MIT License.","reduction='mean', avg_factor=None): \"\"\"Mask cross entropy loss. :param pred: predict result","0 assert pred.size() == target.size() and target.numel() > 0 diff","cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\"",":param avg_factor: avg factor :return: loss \"\"\" assert reduction ==","reduction='mean', avg_factor=None): \"\"\"Cross entropy losses. :param pred: predict result :param","self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list","None: loss = reduce_loss(loss, reduction) else: if reduction == 'mean':","anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes =","X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x #","reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary","use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight = loss_weight","anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios))","loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def __init__(self, beta=1.0,","feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h,","self.use_mask = use_mask self.reduction = reduction self.loss_weight = loss_weight if","target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if","sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list,","= self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class","'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub':","class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean',","featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators = []","= min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h,","that it will be useful, # but WITHOUT ANY WARRANTY;","\"\"\"Forward compute.\"\"\" assert reduction_override in (None, 'none', 'mean', 'sum') reduction","reduction: reduce function :param avg_factor: avg factor :return: loss \"\"\"","reduction self.loss_weight = loss_weight def forward(self, pred, target, weight=None, avg_factor=None,","\"\"\"Import all torch operators.\"\"\" import torch.nn.functional as F import torch.nn","vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import","!= 'none': raise ValueError('avg_factor can not be used with reduction=\"sum\"')","= self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _ =","mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score, label, weight,","entropy losses. :param pred: predict result :param label: gt label","def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary labels. :param labels: labels","self).__init__() def forward(self, x): \"\"\"Get cls score and bbox preds.\"\"\"","self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors","reduction_override else self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight","== target.size() and target.numel() > 0 diff = torch.abs(pred -","target :param weight: weight :param avg_factor: avg factor :param reduction_override:","RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls","def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross entropy loss.","weights :param label_channels: label channels :return: binary label and label","zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\"","reduction = ( reduction_override if reduction_override else self.reduction) loss_cls =",":param labels: labels :param label_weights: label weights :param label_channels: label","self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs)","Anchors(nn.Module): \"\"\"Get anchors according to feature map sizes.\"\"\" def __init__(self,",">= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] - 1]","be useful, # but WITHOUT ANY WARRANTY; without even the",":param desc: config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid","/ anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags","= cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs):","Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox","avg_factor=None): \"\"\"Binary cross entropy loss. :param pred: predict result :param","weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute. :param pred: predict :param","reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta =","1.0) self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg =","= x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators","beta, 0.5 * diff * diff / beta, diff -","x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos,","h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)),","reduce function :param avg_factor: avg factor :return: loss \"\"\" if","loss = loss * weight if avg_factor is None: loss","partial(func, **kwargs) if kwargs else func map_results = map(pfunc, *args)","- target) loss = torch.where(diff < beta, 0.5 * diff","= self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return","reduce funtion :return: loss \"\"\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function","reduction != 'none': raise ValueError('avg_factor can not be used with","x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda()","anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios =","False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling","import Config from functools import partial import numpy as np","assert reduction_override in (None, 'none', 'mean', 'sum') reduction = (","cls score.\"\"\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes =","\"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction","in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h,","= min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)),","is not None: loss = loss * weight if avg_factor","label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels)","or FITNESS FOR A PARTICULAR PURPOSE. See the # MIT","CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init","reduction_override if reduction_override else self.reduction) if target.numel() > 0: loss_bbox","@NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__()","loss. :param pred: predict :param target: target :param beta: beta","'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling =","desc: config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask","1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8, 16, 32,","label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors according to feature","range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w,","loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction self.loss_weight","func: function :param args: args of function :return: result \"\"\"","0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256,","loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def __init__(self, use_sigmoid=False,","in enumerate(img_metas): multi_level_flags = [] for i in range(num_levels): anchor_stride","multi_level_flags = [] for i in range(num_levels): anchor_stride = self.anchor_strides[i]","AnchorGenerator from vega.core.common.config import Config from functools import partial import","labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1,","32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides =","valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \"\"\"Multi","label_channels: label channels :return: binary label and label weights \"\"\"","loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy losses.","loss elif reduction_function == 1: return loss.mean() elif reduction_function ==","1] = 1 if label_weights is None: bin_label_weights = None","is free software; you can redistribute it and/or modify #","def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \"\"\"Mask cross entropy loss.","super(RpnClsLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score and bbox","elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def","losses :param weight: weight :param reduction: reduce function :param avg_factor:","= _expand_binary_labels(label, weight, pred.size(-1)) if weight is not None: weight","bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred,","label, reduction='mean', avg_factor=None): \"\"\"Mask cross entropy loss. :param pred: predict","weights \"\"\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels","self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors =","cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy losses. :param pred:",".0, .0, .0) self.target_stds = target_stds or (1.0, 1.0, 1.0,","( reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight *","funtion :return: loss \"\"\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function ==","torch.nn as nn import torch from vega.search_space.networks.network_factory import NetworkFactory from","binary label and label weights \"\"\" bin_labels = labels.new_full((labels.size(0), label_channels),","use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy loss. :param desc:","coding: utf-8 -*- # Copyright (C) 2020. Huawei Technologies Co.,","# but WITHOUT ANY WARRANTY; without even the implied warranty","label, reduction='none') if weight is not None: weight = weight.float()","torch.nn.functional as F import torch.nn as nn import torch from","bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross entropy","target.size() and target.numel() > 0 diff = torch.abs(pred - target)","*args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\"","This program is distributed in the hope that it will","4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox","implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR","weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override in (None,","avg_factor=None): \"\"\"Cross entropy losses. :param pred: predict result :param label:","\"\"\"Forward compute. :param pred: predict :param target: target :param weight:","cls score and bbox preds.\"\"\" cls_scores = x[0] bbox_preds =","avg_factor is None num_rois = pred.size()[0] inds = torch.arange(0, num_rois,","under the terms of the MIT License. # This program","anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels","loss. :param loss: losses :param weight: weight :param reduction: reduce","is None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0),","factor :return: loss \"\"\" if pred.dim() != label.dim(): label, weight","reduce function :param avg_factor: avg factor :return: loss \"\"\" assert","(cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) =","bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox =","def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score.\"\"\"","= CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self,","else: self.loss_function = cross_entropy def forward(self, cls_score, label, weight, avg_factor,","None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor)","to feature map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None):","label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy losses. :param pred: predict","it and/or modify # it under the terms of the","label_weights is None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1,","beta: beta :return: loss \"\"\" assert beta > 0 assert","vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from functools import","= use_mask self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid:","ValueError('avg_factor can not be used with reduction=\"sum\"') return loss def","__init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales =","= featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h","i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i]","it will be useful, # but WITHOUT ANY WARRANTY; without","cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales,","reduction = ( reduction_override if reduction_override else self.reduction) if target.numel()","if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1))","label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):","= loss.sum() / avg_factor elif reduction != 'none': raise ValueError('avg_factor","label: gt label :param weight: weight :param reduction: reduce function","feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return","@NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors according to feature map sizes.\"\"\"","bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox",":param label_weights: label weights :param label_channels: label channels :return: binary","F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce","terms of the MIT License. # This program is distributed","if avg_factor is None: loss = reduce_loss(loss, reduction) else: if","free software; you can redistribute it and/or modify # it","= loss * weight if avg_factor is None: loss =","gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self,","= anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios or","return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args,","gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return","[] for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w","avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred,","/ beta, diff - 0.5 * beta) return loss def","return loss elif reduction_function == 1: return loss.mean() elif reduction_function","FOR A PARTICULAR PURPOSE. See the # MIT License for","operators.\"\"\" import torch.nn.functional as F import torch.nn as nn import","1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets =","Class Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss()","bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def __init__(self): super(RpnLossInput,","avg_factor: avg factor :param reduction_override: reduce override :return: loss \"\"\"","hope that it will be useful, # but WITHOUT ANY","weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels):","binary labels. :param labels: labels :param label_weights: label weights :param","label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels = labels.reshape(-1) label_weights","1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] - 1] =","= pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None,","Co., Ltd. All rights reserved. # This program is free","map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module):","program is distributed in the hope that it will be","super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels =","if reduction_function == 0: return loss elif reduction_function == 1:",":param avg_factor: avg factor :return: loss \"\"\" loss = F.cross_entropy(pred,","img_meta in enumerate(img_metas): multi_level_flags = [] for i in range(num_levels):","target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or","4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox =","super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction =","details. \"\"\"Import all torch operators.\"\"\" import torch.nn.functional as F import","self).__init__() def forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas,","losses. :param pred: predict result :param label: gt label :param","0: bin_labels[inds, labels[inds] - 1] = 1 if label_weights is","bin_labels[inds, labels[inds] - 1] = 1 if label_weights is None:","reduction) else: if reduction == 'mean': loss = loss.sum() /","= 1 if label_weights is None: bin_label_weights = None else:","from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config","= sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list,","**kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target,","* self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator)","class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False,","FITNESS FOR A PARTICULAR PURPOSE. See the # MIT License","self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get","diff - 0.5 * beta) return loss def cross_entropy(pred, label,","anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes = list(","cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward","bbox preds.\"\"\" cls_scores = x[0] bbox_preds = x[1] return cls_scores,","'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1,","assert pred.size() == target.size() and target.numel() > 0 diff =","channels :return: binary label and label weights \"\"\" bin_labels =","predict result :param target: target :param label: gt label :param","= [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs","torch.where(diff < beta, 0.5 * diff * diff / beta,",".0, .0) self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0)","PURPOSE. See the # MIT License for more details. \"\"\"Import",":param pred: predict :param target: target :param weight: weight :param","compute. :param pred: predict :param target: target :param weight: weight","Config from functools import partial import numpy as np from","target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute. :param pred: predict","def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales","will be useful, # but WITHOUT ANY WARRANTY; without even","losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples):","labels: labels :param label_weights: label weights :param label_channels: label channels","self).__init__() def forward(self, x): \"\"\"Get cls score.\"\"\" cls_scores = x[2][0]","pred: predict result :param label: gt label :param weight: weight","self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction = reduction","if label_weights is None: bin_label_weights = None else: bin_label_weights =","else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def","'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x):","not None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction,","x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']]","target :param beta: beta :return: loss \"\"\" assert beta >","= anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios","in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels =","avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\"","pred.size(-1)) if weight is not None: weight = weight.float() loss",":param pred: predict result :param target: target :param label: gt","MIT License for more details. \"\"\"Import all torch operators.\"\"\" import","All rights reserved. # This program is free software; you","reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override in (None, 'none', 'mean',","self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \"\"\"Get","pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else:","args: args of function :return: result \"\"\" pfunc = partial(func,","loss \"\"\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return","sampling=True): self.target_means = target_means or (.0, .0, .0, .0) self.target_stds","2.0] self.anchor_strides = anchor_strides or [4, 8, 16, 32, 64]","dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None]","2020. Huawei Technologies Co., Ltd. All rights reserved. # This","self).__init__() def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes,","losses :param reduction: reduce funtion :return: loss \"\"\" reduction_function =","'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction':","\"\"\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >=","* diff * diff / beta, diff - 0.5 *","\"\"\" if weight is not None: loss = loss *","can not be used with reduction=\"sum\"') return loss def reduce_loss(loss,","as nn import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils","function :param avg_factor: avg factor :return: loss \"\"\" assert reduction","weight, pred.size(-1)) if weight is not None: weight = weight.float()","self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape']","= beta self.reduction = reduction self.loss_weight = loss_weight def forward(self,","function :param avg_factor: avg factor :return: loss \"\"\" loss =","num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore,","use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0, .0, .0,","<reponame>Huawei-Ascend/modelzoo<filename>built-in/TensorFlow/Research/cv/image_classification/Cars_for_TensorFlow/automl/vega/search_space/networks/pytorch/operator/rpn.py # -*- coding: utf-8 -*- # Copyright (C) 2020.","self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score,","elif reduction != 'none': raise ValueError('avg_factor can not be used","smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox","diff = torch.abs(pred - target) loss = torch.where(diff < beta,","bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze()","or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8,","score and bbox preds.\"\"\" cls_scores = x[0] bbox_preds = x[1]","vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import","import AnchorGenerator from vega.core.common.config import Config from functools import partial","anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors,","img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None,","from vega.core.common.config import Config from functools import partial import numpy","/ anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags)","program is free software; you can redistribute it and/or modify","loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\"","* smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return","= pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice =","bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels = labels.reshape(-1) label_weights =","# Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights","\"\"\" if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight,","cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas","16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides","according to feature map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None,","img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap in","F import torch.nn as nn import torch from vega.search_space.networks.network_factory import","pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss,","> 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight,","labels. :param labels: labels :param label_weights: label weights :param label_channels:","redistribute it and/or modify # it under the terms of","{'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr':","target: target :param beta: beta :return: loss \"\"\" assert beta","reduction='mean', avg_factor=None): \"\"\"Weight reduce loss. :param loss: losses :param weight:","label :param reduction: reduce function :param avg_factor: avg factor :return:","anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _","num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels,","PARTICULAR PURPOSE. See the # MIT License for more details.","@weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1 loss. :param pred:","self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor,","reduction_function == 0: return loss elif reduction_function == 1: return","reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif","from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator)","\"\"\"Get cls score and bbox preds.\"\"\" cls_scores = x[0] bbox_preds","self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner':","self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes)","beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__() self.beta","self.loss_weight = loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None,","loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs)","bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred,","= reduce_loss(loss, reduction) else: if reduction == 'mean': loss =","loss: losses :param weight: weight :param reduction: reduce function :param","x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module):","num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights,","forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore =","loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss,","CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x):","losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,","num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred,","num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target,","label, weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override in","bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for","Technologies Co., Ltd. All rights reserved. # This program is","self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32]","factor :return: loss \"\"\" assert reduction == 'mean' and avg_factor","ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY","loss. :param pred: predict result :param label: gt label :param","= torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return","bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x","0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta,","in (None, 'none', 'mean', 'sum') reduction = ( reduction_override if","avg_factor: avg factor :return: loss \"\"\" loss = F.cross_entropy(pred, label,","[multi_level_anchors for _ in range(num_imgs)] valid_flag_list = [] for img_id,","gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None, target_stds=None,","-*- # Copyright (C) 2020. Huawei Technologies Co., Ltd. All","all torch operators.\"\"\" import torch.nn.functional as F import torch.nn as","in cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base,","i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list =","multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls,","bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1,","label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module):","kwargs else func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results)))","loss input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \"\"\"Get","self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else:","the hope that it will be useful, # but WITHOUT","reduction_override in (None, 'none', 'mean', 'sum') reduction = ( reduction_override","\"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x","label_weights: label weights :param label_channels: label channels :return: binary label","min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w)","label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4)","label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox","x.\"\"\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples)","use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7,","loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth","result :param label: gt label :param weight: weight :param reduction:","def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss,","None: loss = loss * weight if avg_factor is None:","out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels","RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self,","gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list,","self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss(",":param beta: beta :return: loss \"\"\" assert beta > 0","0.5 * diff * diff / beta, diff - 0.5","_expand_binary_labels(label, weight, pred.size(-1)) if weight is not None: weight =","img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \"\"\"Multi apply. :param func:","== 0: return loss elif reduction_function == 1: return loss.mean()","class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self,","valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func,","loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \"\"\"Mask cross entropy","@NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None, target_stds=None, num_classes=2,","elif reduction_function == 1: return loss.mean() elif reduction_function == 2:","See the # MIT License for more details. \"\"\"Import all","the terms of the MIT License. # This program is","class Anchors(nn.Module): \"\"\"Get anchors according to feature map sizes.\"\"\" def","\"\"\"Init Cross Entropy loss. :param desc: config dict \"\"\" super(CustomCrossEntropyLoss,","Cross Entropy loss. :param desc: config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__()","= weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss",":param loss: losses :param reduction: reduce funtion :return: loss \"\"\"","vega.core.common.config import Config from functools import partial import numpy as","def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1 loss. :param pred: predict","cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:]","if reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label,","RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x):","vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self):","valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w),","anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags)","= self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor,","flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list,","img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class","cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def __init__(self):","original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list,","anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8,","loss = loss.sum() / avg_factor elif reduction != 'none': raise","loss \"\"\" if weight is not None: loss = loss","**kwargs): \"\"\"Forward compute. :param pred: predict :param target: target :param","(valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas,","def reduce_loss(loss, reduction): \"\"\"Reduce loss compute. :param loss: losses :param","reduction='mean', avg_factor=None): \"\"\"Binary cross entropy loss. :param pred: predict result","def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce loss. :param loss:","weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is not None:","= F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif reduction_function","def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\"","list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__()","anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors","__init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score and","sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors according to feature map","feature map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg","forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute. :param","__init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss()","num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner',","reduce override :return: loss \"\"\" reduction = ( reduction_override if","= ( reduction_override if reduction_override else self.reduction) if target.numel() >","if inds.numel() > 0: bin_labels[inds, labels[inds] - 1] = 1","= multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return","reduce_loss(loss, reduction) else: if reduction == 'mean': loss = loss.sum()","cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox",":param label_channels: label channels :return: binary label and label weights","weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def","distributed in the hope that it will be useful, #","input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \"\"\"Get cls","avg factor :param reduction_override: reduce override :return: loss \"\"\" reduction","0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5,","or (.0, .0, .0, .0) self.target_stds = target_stds or (1.0,",":param target: target :param beta: beta :return: loss \"\"\" assert","pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if","[] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors)","def forward(self, x): \"\"\"Get cls score.\"\"\" cls_scores = x[2][0] bbox_preds","def forward(self, x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels, label_weights, bbox_targets,","weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand","= x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list,","'none', 'mean', 'sum') reduction = ( reduction_override if reduction_override else","pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds,","avg_factor: avg factor :return: loss \"\"\" assert reduction == 'mean'","out_channels def forward(self, x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels, label_weights,","loss = reduce_loss(loss, reduction) else: if reduction == 'mean': loss","loss: losses :param reduction: reduce funtion :return: loss \"\"\" reduction_function","weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth","# out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes,","self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class","label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is not","x): \"\"\"Get cls score.\"\"\" cls_scores = x[2][0] bbox_preds = x[2][1]","of the MIT License. # This program is distributed in","'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5},","= F.cross_entropy(pred, label, reduction='none') if weight is not None: weight","map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg =","loss \"\"\" loss = F.cross_entropy(pred, label, reduction='none') if weight is","reduce_loss(loss, reduction): \"\"\"Reduce loss compute. :param loss: losses :param reduction:","\"\"\"Expand binary labels. :param labels: labels :param label_weights: label weights","== 1: return loss.mean() elif reduction_function == 2: return loss.sum()","**kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def","= weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss",":return: loss \"\"\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0:","pred.size() == target.size() and target.numel() > 0 diff = torch.abs(pred","x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\"","cls_scores = x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator)","self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def","\"\"\"Get cls score.\"\"\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes","useful, # but WITHOUT ANY WARRANTY; without even the implied","import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config","w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h)",":return: loss \"\"\" if weight is not None: loss =","reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None):","x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore =","@NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__() def","you can redistribute it and/or modify # it under the","min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w))","label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means,","input.\"\"\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \"\"\"Get cls","len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for i in","avg factor :return: loss \"\"\" if weight is not None:","'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0,","gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator)","def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get",":param avg_factor: avg factor :return: loss \"\"\" if pred.dim() !=","return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0):","multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list =",".0) self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels","loss compute. :param loss: losses :param reduction: reduce funtion :return:","multi_level_anchors = [] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i],","pred: predict :param target: target :param beta: beta :return: loss","label :param weight: weight :param reduction: reduce function :param avg_factor:","inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds,","target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return","__init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\" super(CustomSmoothL1Loss, self).__init__()","label weights \"\"\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds =","nn import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import","256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight':","= [] for img_id, img_meta in enumerate(img_metas): multi_level_flags = []","is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \"\"\"Create",":param loss: losses :param weight: weight :param reduction: reduce function","_expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary labels. :param labels: labels :param","self.loss_function = cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None,","weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce loss. :param loss: losses :param","= anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes =","self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels","is not None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight,","\"\"\"Smooth L1 Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth","forward(self, x): \"\"\"Get cls score.\"\"\" cls_scores = x[2][0] bbox_preds =","weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean',",":param weight: weight :param avg_factor: avg factor :param reduction_override: reduce","\"\"\"Weight reduce loss. :param loss: losses :param weight: weight :param","F.cross_entropy(pred, label, reduction='none') if weight is not None: weight =","multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \"\"\"Multi apply.","bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2,","__init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy loss. :param",":param label: gt label :param weight: weight :param reduction: reduce","= list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors,","return loss def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary labels. :param","AnchorTargetOp(nn.Module): \"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None,","for _ in range(num_imgs)] valid_flag_list = [] for img_id, img_meta","return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \"\"\"Mask cross","A PARTICULAR PURPOSE. See the # MIT License for more","= [featmap.size()[-2:] for featmap in cls_scores] anchor_generators = [] for","six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class","utf-8 -*- # Copyright (C) 2020. Huawei Technologies Co., Ltd.","assert reduction == 'mean' and avg_factor is None num_rois =","preds.\"\"\" cls_scores = x[0] bbox_preds = x[1] return cls_scores, bbox_preds","\"\"\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return loss","function :param avg_factor: avg factor :return: loss \"\"\" if pred.dim()","= ( reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight","\"\"\"Smooth l1 loss. :param pred: predict :param target: target :param","mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \"\"\"Mask cross entropy loss. :param","gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap","feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes,","beta, diff - 0.5 * beta) return loss def cross_entropy(pred,","label_channels): \"\"\"Expand binary labels. :param labels: labels :param label_weights: label","torch.abs(pred - target) loss = torch.where(diff < beta, 0.5 *","0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1,","\"\"\"Multi apply. :param func: function :param args: args of function","label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):","labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self,","import numpy as np from six.moves import map, zip from","weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return","= out_channels def forward(self, x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels,","dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask","\"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \"\"\"Get","'mean', 'sum') reduction = ( reduction_override if reduction_override else self.reduction)","l1 loss. :param pred: predict :param target: target :param beta:","( reduction_override if reduction_override else self.reduction) if target.numel() > 0:","self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module):","\"\"\" reduction = ( reduction_override if reduction_override else self.reduction) if","forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert","binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross entropy loss. :param","is None num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long,","super(Anchors, self).__init__() def forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes,","label: gt label :param reduction: reduce function :param avg_factor: avg","valid_flag_list = [] for img_id, img_meta in enumerate(img_metas): multi_level_flags =","target: target :param label: gt label :param reduction: reduce function","F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif reduction_function ==","= len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for i","= target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes","self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\"","bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds,","loss * weight if avg_factor is None: loss = reduce_loss(loss,","if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr':","loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels,","def multi_apply(func, *args, **kwargs): \"\"\"Multi apply. :param func: function :param","of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See","'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler',","forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore","vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import","partial import numpy as np from six.moves import map, zip","zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class Loss.\"\"\" def __init__(self, out_channels=2):","reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def","return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1 loss.","0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__()","return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def","and target.numel() > 0 diff = torch.abs(pred - target) loss","anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list,",":param weight: weight :param reduction: reduce function :param avg_factor: avg","bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss,","anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags =","else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction,","from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def","AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels,","# This program is free software; you can redistribute it","super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction self.loss_weight =","but WITHOUT ANY WARRANTY; without even the implied warranty of","**kwargs): \"\"\"Multi apply. :param func: function :param args: args of","of function :return: result \"\"\" pfunc = partial(func, **kwargs) if","reduction_override=None, **kwargs): \"\"\"Forward compute. :param pred: predict :param target: target","compute.\"\"\" assert reduction_override in (None, 'none', 'mean', 'sum') reduction =","software; you can redistribute it and/or modify # it under","for featmap in cls_scores] anchor_generators = [] for anchor_base in","use_mask self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function","= bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4)","= loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs):","and avg_factor is None num_rois = pred.size()[0] inds = torch.arange(0,","reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif","'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals':","smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1 loss. :param pred: predict :param","self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds,",":return: result \"\"\" pfunc = partial(func, **kwargs) if kwargs else","self.target_means = target_means or (.0, .0, .0, .0) self.target_stds =","Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1 loss.\"\"\"","predict result :param label: gt label :param weight: weight :param","map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn","import torch.nn.functional as F import torch.nn as nn import torch","pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute. :param pred:","np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss","anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4,","\"\"\"Anchor Target.\"\"\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True):","MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the #","def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox =","self.beta = beta self.reduction = reduction self.loss_weight = loss_weight def","forward(self, x): \"\"\"Get x.\"\"\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,","reduction_function == 1: return loss.mean() elif reduction_function == 2: return","loss.sum() / avg_factor elif reduction != 'none': raise ValueError('avg_factor can","loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4)","def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \"\"\"Get cls score","= anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in","feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h =","weight is not None: weight = weight.float() loss = weight_reduce_loss(loss,","avg factor :return: loss \"\"\" if pred.dim() != label.dim(): label,","losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples)","target.numel() > 0 diff = torch.abs(pred - target) loss =","import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput,","(.0, .0, .0, .0) self.target_stds = target_stds or (1.0, 1.0,","= anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or","bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred =","Entropy loss. :param desc: config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid","args of function :return: result \"\"\" pfunc = partial(func, **kwargs)","!= label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight","num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for","label weights :param label_channels: label channels :return: binary label and","map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn Class","'none': raise ValueError('avg_factor can not be used with reduction=\"sum\"') return","functools import partial import numpy as np from six.moves import","= torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds]","[4, 8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if","enumerate(img_metas): multi_level_flags = [] for i in range(num_levels): anchor_stride =","raise ValueError('avg_factor can not be used with reduction=\"sum\"') return loss","< beta, 0.5 * diff * diff / beta, diff","score.\"\"\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda()","num_levels = len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels):","AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from","if kwargs else func map_results = map(pfunc, *args) return tuple(map(list,","[featmap.size()[-2:] for featmap in cls_scores] anchor_generators = [] for anchor_base","[] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs =","pred: predict result :param target: target :param label: gt label","loss \"\"\" assert beta > 0 assert pred.size() == target.size()","= x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore =","return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None,","self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1,","function :return: result \"\"\" pfunc = partial(func, **kwargs) if kwargs","reduction=\"sum\"') return loss def reduce_loss(loss, reduction): \"\"\"Reduce loss compute. :param","self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels =","labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1,","None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight,","return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \"\"\"Rpn loss input.\"\"\" def","\"\"\"Get anchors according to feature map sizes.\"\"\" def __init__(self, anchor_base_sizes_cfg=None,","self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios","_ in range(num_imgs)] valid_flag_list = [] for img_id, img_meta in","import torch.nn as nn import torch from vega.search_space.networks.network_factory import NetworkFactory","1 if label_weights is None: bin_label_weights = None else: bin_label_weights","NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from","self.reduction = reduction self.loss_weight = loss_weight def forward(self, pred, target,","cfg=None, sampling=True): self.target_means = target_means or (.0, .0, .0, .0)","labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels = labels.reshape(-1)","* beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):","label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None,","\"\"\"Rpn Class Loss.\"\"\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls =","= self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights",":return: loss \"\"\" reduction = ( reduction_override if reduction_override else",":param label: gt label :param reduction: reduce function :param avg_factor:","for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas)",":param reduction: reduce funtion :return: loss \"\"\" reduction_function = F._Reduction.get_enum(reduction)","= loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function","def forward(self, x): \"\"\"Create anchor.\"\"\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore","super(AnchorTargetOp, self).__init__() def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors,","loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function =","compute. :param loss: losses :param reduction: reduce funtion :return: loss","x): \"\"\"Get cls score and bbox preds.\"\"\" cls_scores = x[0]","loss = F.cross_entropy(pred, label, reduction='none') if weight is not None:","tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num':","def forward(self, x): \"\"\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\"\"\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas,","not be used with reduction=\"sum\"') return loss def reduce_loss(loss, reduction):","MIT License. # This program is distributed in the hope","= map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \"\"\"Rpn","reduction_override: reduce override :return: loss \"\"\" reduction = ( reduction_override","weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda()","bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy","Target.\"\"\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means","avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override in (None, 'none',","num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score,","gt label :param reduction: reduce function :param avg_factor: avg factor","weight is not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits(","anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _","'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False})","def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward compute.","0.5 * beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean',","-1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling","in the hope that it will be useful, # but","entropy loss. :param pred: predict result :param label: gt label","reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1",":param target: target :param weight: weight :param avg_factor: avg factor",":return: loss \"\"\" assert reduction == 'mean' and avg_factor is","target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \"\"\"Weight reduce loss.","Huawei Technologies Co., Ltd. All rights reserved. # This program","class CustomSmoothL1Loss(nn.Module): \"\"\"Smooth L1 Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0):","else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \"\"\"Smooth l1","self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls","= mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score, label,","{'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False},","in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors","reduction == 'mean': loss = loss.sum() / avg_factor elif reduction","weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \"\"\"Rpn input.\"\"\" def __init__(self): super(RpnClsLossInput, self).__init__()","L1 Loss.\"\"\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \"\"\"Init smooth l1","loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \"\"\"Forward","loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \"\"\"Cross Entropy Loss.\"\"\" def __init__(self,","valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore","anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or","reduce function :param avg_factor: avg factor :return: loss \"\"\" loss","or [8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0,","entropy loss. :param pred: predict result :param target: target :param",":param avg_factor: avg factor :param reduction_override: reduce override :return: loss","valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w /","== 'mean' and avg_factor is None num_rois = pred.size()[0] inds","gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \"\"\"Get anchors according","label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() >","not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(),","as F import torch.nn as nn import torch from vega.search_space.networks.network_factory","WARRANTY; without even the implied warranty of # MERCHANTABILITY or","bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels,","self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask:","Loss.\"\"\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy","/ avg_factor elif reduction != 'none': raise ValueError('avg_factor can not","range(num_imgs)] valid_flag_list = [] for img_id, img_meta in enumerate(img_metas): multi_level_flags","= weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights,","# MIT License for more details. \"\"\"Import all torch operators.\"\"\"","x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators =","pred: predict :param target: target :param weight: weight :param avg_factor:","def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \"\"\"Cross entropy losses. :param","label, weight=None, reduction='mean', avg_factor=None): \"\"\"Binary cross entropy loss. :param pred:","loss \"\"\" assert reduction == 'mean' and avg_factor is None","cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights,","-1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self,","img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes,","loss. :param desc: config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid =","if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy","avg_factor: avg factor :return: loss \"\"\" if pred.dim() != label.dim():","pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return","target, label, reduction='mean', avg_factor=None): \"\"\"Mask cross entropy loss. :param pred:","> 0: bin_labels[inds, labels[inds] - 1] = 1 if label_weights","target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target,","weight :param reduction: reduce function :param avg_factor: avg factor :return:","loss. :param pred: predict result :param target: target :param label:","64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else","target, beta=1.0): \"\"\"Smooth l1 loss. :param pred: predict :param target:","\"\"\" assert reduction == 'mean' and avg_factor is None num_rois","is distributed in the hope that it will be useful,","F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor)","avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \"\"\"Expand binary labels.","avg_factor is None: loss = reduce_loss(loss, reduction) else: if reduction","8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg","import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module):","for more details. \"\"\"Import all torch operators.\"\"\" import torch.nn.functional as","bbox_weights, num_total_samples): \"\"\"Get loss.\"\"\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1)","def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \"\"\"Init Cross Entropy loss.","as np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import","cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \"\"\"Forward compute.\"\"\" assert reduction_override","self).__init__() self.beta = beta self.reduction = reduction self.loss_weight = loss_weight","= anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list,","gt label :param weight: weight :param reduction: reduce function :param","config dict \"\"\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask =","> 0 assert pred.size() == target.size() and target.numel() > 0","\"\"\"Mask cross entropy loss. :param pred: predict result :param target:","# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the","multi_apply(func, *args, **kwargs): \"\"\"Multi apply. :param func: function :param args:","torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice,","torch operators.\"\"\" import torch.nn.functional as F import torch.nn as nn","without even the implied warranty of # MERCHANTABILITY or FITNESS","apply. :param func: function :param args: args of function :return:"],"string":"[\n \"= reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy\",\n \"diff * diff / beta, diff - 0.5 * beta)\",\n \"0 diff = torch.abs(pred - target) loss = torch.where(diff <\",\n \"torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] -\",\n \"with reduction=\\\"sum\\\"') return loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute.\",\n \"= partial(func, **kwargs) if kwargs else func map_results = map(pfunc,\",\n \"= [] for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h,\",\n \"# This program is distributed in the hope that it\",\n \"use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss. :param desc: config\",\n \"'sum') reduction = ( reduction_override if reduction_override else self.reduction) loss_cls\",\n \"be used with reduction=\\\"sum\\\"') return loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce\",\n \"[x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore\",\n \"inds.numel() > 0: bin_labels[inds, labels[inds] - 1] = 1 if\",\n \"= weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label,\",\n \"beta > 0 assert pred.size() == target.size() and target.numel() >\",\n \"= x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas =\",\n \"x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets,\",\n \"WITHOUT ANY WARRANTY; without even the implied warranty of #\",\n \"the MIT License. # This program is distributed in the\",\n \"= img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w =\",\n \"loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta\",\n \"\\\"\\\"\\\" loss = F.cross_entropy(pred, label, reduction='none') if weight is not\",\n \"labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel()\",\n \"self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module):\",\n \"License for more details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import torch.nn.functional\",\n \"License. # This program is distributed in the hope that\",\n \"self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides\",\n \"= x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights,\",\n \"Ltd. All rights reserved. # This program is free software;\",\n \"label and label weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels), 0)\",\n \"\\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init\",\n \"weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss =\",\n \"self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x):\",\n \"Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3,\",\n \"self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = []\",\n \"import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget\",\n \"multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def\",\n \"return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight\",\n \"num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice\",\n \"= [] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i])\",\n \"Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights reserved.\",\n \"if weight is not None: loss = loss * weight\",\n \"super(RpnLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores =\",\n \"label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss\",\n \"None num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device)\",\n \"reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred,\",\n \"beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross\",\n \":param avg_factor: avg factor :return: loss \\\"\\\"\\\" if weight is\",\n \"function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" if weight\",\n \"warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\",\n \"weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param loss: losses\",\n \"if target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred,\",\n \"anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)]\",\n \"weight: weight :param avg_factor: avg factor :param reduction_override: reduce override\",\n \"1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean',\",\n \"losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,\",\n \"avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask\",\n \"for img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for i\",\n \"if reduction == 'mean': loss = loss.sum() / avg_factor elif\",\n \"= x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class\",\n \"the # MIT License for more details. \\\"\\\"\\\"Import all torch\",\n \"\\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg,\",\n \"featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h /\",\n \"= target_means or (.0, .0, .0, .0) self.target_stds = target_stds\",\n \"1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1,\",\n \"[8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0]\",\n \"gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores]\",\n \":return: loss \\\"\\\"\\\" assert beta > 0 assert pred.size() ==\",\n \"diff / beta, diff - 0.5 * beta) return loss\",\n \"bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred,\",\n \"inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1)\",\n \"weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param pred: predict result\",\n \"= x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class\",\n \"= binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function =\",\n \"class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__()\",\n \"torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param\",\n \"3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return\",\n \"is None: loss = reduce_loss(loss, reduction) else: if reduction ==\",\n \"avg_factor elif reduction != 'none': raise ValueError('avg_factor can not be\",\n \"override :return: loss \\\"\\\"\\\" reduction = ( reduction_override if reduction_override\",\n \"'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler':\",\n \"False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \\\"\\\"\\\"Create\",\n \"valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling)\",\n \"range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for\",\n \"return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary\",\n \"= [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas,\",\n \"None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\"\",\n \"beta :return: loss \\\"\\\"\\\" assert beta > 0 assert pred.size()\",\n \"len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels): anchors =\",\n \"*args, **kwargs): \\\"\\\"\\\"Multi apply. :param func: function :param args: args\",\n \"3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets\",\n \"bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def\",\n \"self.out_channels = out_channels def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred,\",\n \"= x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores,\",\n \"2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples)\",\n \"and label weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds\",\n \"anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list = []\",\n \"- 1] = 1 if label_weights is None: bin_label_weights =\",\n \"avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param loss: losses :param weight: weight\",\n \"\\\"\\\"\\\"Reduce loss compute. :param loss: losses :param reduction: reduce funtion\",\n \"'mean' and avg_factor is None num_rois = pred.size()[0] inds =\",\n \"else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr':\",\n \"from functools import partial import numpy as np from six.moves\",\n \"0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug':\",\n \"def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means =\",\n \"= CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\"\",\n \"return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets,\",\n \"cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score,\",\n \"bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def\",\n \"feat_h, feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h\",\n \"\\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction\",\n \"factor :param reduction_override: reduce override :return: loss \\\"\\\"\\\" reduction =\",\n \"avg factor :return: loss \\\"\\\"\\\" loss = F.cross_entropy(pred, label, reduction='none')\",\n \"= labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if\",\n \"label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is\",\n \"loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target,\",\n \"[0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8, 16,\",\n \"self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def\",\n \"self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def forward(self,\",\n \"else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores,\",\n \"(None, 'none', 'mean', 'sum') reduction = ( reduction_override if reduction_override\",\n \"loss = torch.where(diff < beta, 0.5 * diff * diff\",\n \"= labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3,\",\n \"= label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label,\",\n \"== 'mean': loss = loss.sum() / avg_factor elif reduction !=\",\n \"l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction\",\n \"'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def\",\n \"1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls,\",\n \"bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights,\",\n \"loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute. :param loss: losses\",\n \"avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss. :param pred: predict result :param\",\n \"weight is not None: loss = loss * weight if\",\n \"forward(self, x): \\\"\\\"\\\"Get cls score and bbox preds.\\\"\\\"\\\" cls_scores =\",\n \"bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore\",\n \"\\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score =\",\n \"2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples)\",\n \"loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator)\",\n \"loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss. :param desc: config dict \\\"\\\"\\\"\",\n \"target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0,\",\n \"torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from\",\n \"= F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction,\",\n \"cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels\",\n \"label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls =\",\n \"[] for img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for\",\n \"1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else 1\",\n \":param reduction_override: reduce override :return: loss \\\"\\\"\\\" reduction = (\",\n \"weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none')\",\n \"anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5,\",\n \"not None: loss = loss * weight if avg_factor is\",\n \"@NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss,\",\n \"function :param args: args of function :return: result \\\"\\\"\\\" pfunc\",\n \"bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross\",\n \"return loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute. :param loss:\",\n \"Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross\",\n \"even the implied warranty of # MERCHANTABILITY or FITNESS FOR\",\n \":param target: target :param label: gt label :param reduction: reduce\",\n \"32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None\",\n \"smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction =\",\n \"loss def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary labels. :param labels:\",\n \"x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module):\",\n \"predict :param target: target :param beta: beta :return: loss \\\"\\\"\\\"\",\n \":param pred: predict result :param label: gt label :param weight:\",\n \"is not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred,\",\n \"the implied warranty of # MERCHANTABILITY or FITNESS FOR A\",\n \"for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w =\",\n \"gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return\",\n \"= num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name':\",\n \"target :param label: gt label :param reduction: reduce function :param\",\n \"it under the terms of the MIT License. # This\",\n \":param func: function :param args: args of function :return: result\",\n \"else: if reduction == 'mean': loss = loss.sum() / avg_factor\",\n \"loss \\\"\\\"\\\" if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label,\",\n \"= reduction self.loss_weight = loss_weight def forward(self, pred, target, weight=None,\",\n \"target: target :param weight: weight :param avg_factor: avg factor :param\",\n \"anchors according to feature map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None,\",\n \"reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss. :param desc: config dict\",\n \"pfunc = partial(func, **kwargs) if kwargs else func map_results =\",\n \"factor :return: loss \\\"\\\"\\\" if weight is not None: loss\",\n \"avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param pred: predict :param target:\",\n \"import partial import numpy as np from six.moves import map,\",\n \"= torch.where(diff < beta, 0.5 * diff * diff /\",\n \"sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg\",\n \"reduce loss. :param loss: losses :param weight: weight :param reduction:\",\n \"can redistribute it and/or modify # it under the terms\",\n \"CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score,\",\n \"or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls\",\n \"= [multi_level_anchors for _ in range(num_imgs)] valid_flag_list = [] for\",\n \"anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs):\",\n \"gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply. :param func: function\",\n \"weight: weight :param reduction: reduce function :param avg_factor: avg factor\",\n \"gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply. :param\",\n \"self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function\",\n \"* weight if avg_factor is None: loss = reduce_loss(loss, reduction)\",\n \"beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss\",\n \":return: loss \\\"\\\"\\\" loss = F.cross_entropy(pred, label, reduction='none') if weight\",\n \"target) loss = torch.where(diff < beta, 0.5 * diff *\",\n \"num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0, .0,\",\n \"__init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores\",\n \"\\\"\\\"\\\"Binary cross entropy loss. :param pred: predict result :param label:\",\n \"bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn\",\n \"gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator)\",\n \"img_id, img_meta in enumerate(img_metas): multi_level_flags = [] for i in\",\n \"bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return\",\n \"= cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels,\",\n \"import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator\",\n \"reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight,\",\n \"reserved. # This program is free software; you can redistribute\",\n \"cross entropy loss. :param pred: predict result :param target: target\",\n \"@NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False,\",\n \"avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\"\",\n \"result :param target: target :param label: gt label :param reduction:\",\n \"weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss. :param pred: predict\",\n \"factor :return: loss \\\"\\\"\\\" loss = F.cross_entropy(pred, label, reduction='none') if\",\n \"= label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls\",\n \"labels[inds] - 1] = 1 if label_weights is None: bin_label_weights\",\n \"'mean': loss = loss.sum() / avg_factor elif reduction != 'none':\",\n \"else func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator)\",\n \"reduction: reduce funtion :return: loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if\",\n \"loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction,\",\n \":return: binary label and label weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0),\",\n \"= x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss\",\n \"1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg\",\n \"x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x\",\n \"reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score,\",\n \"weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def\",\n \"4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3,\",\n \"_ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w\",\n \"and bbox preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds = x[1] return\",\n \"x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list,\",\n \"self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights =\",\n \"reduction == 'mean' and avg_factor is None num_rois = pred.size()[0]\",\n \"label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score,\",\n \":param pred: predict :param target: target :param beta: beta :return:\",\n \"'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border':\",\n \"label channels :return: binary label and label weights \\\"\\\"\\\" bin_labels\",\n \"if weight is not None: weight = weight.float() loss =\",\n \"cross entropy loss. :param pred: predict result :param label: gt\",\n \"img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg).\",\n \"reduction='none') if weight is not None: weight = weight.float() loss\",\n \"bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels =\",\n \"return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self,\",\n \"return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def\",\n \"import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes\",\n \"None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels)\",\n \"valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list,\",\n \"= Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou':\",\n \"class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__() def\",\n \"NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from\",\n \"= bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0,\",\n \"avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred\",\n \"**kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in (None, 'none', 'mean', 'sum')\",\n \"return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor\",\n \":param reduction: reduce function :param avg_factor: avg factor :return: loss\",\n \"\\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x):\",\n \"= bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets,\",\n \"device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def\",\n \":return: loss \\\"\\\"\\\" if pred.dim() != label.dim(): label, weight =\",\n \"# it under the terms of the MIT License. #\",\n \"\\\"\\\"\\\" assert beta > 0 assert pred.size() == target.size() and\",\n \"reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param\",\n \"result \\\"\\\"\\\" pfunc = partial(func, **kwargs) if kwargs else func\",\n \"anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16,\",\n \"img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w\",\n \"-*- coding: utf-8 -*- # Copyright (C) 2020. Huawei Technologies\",\n \"or [4, 8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides)\",\n \"self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function =\",\n \"bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross\",\n \"'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp,\",\n \"featmap in cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes:\",\n \"def forward(self, x): \\\"\\\"\\\"Get cls score and bbox preds.\\\"\\\"\\\" cls_scores\",\n \"None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights\",\n \"loss \\\"\\\"\\\" reduction = ( reduction_override if reduction_override else self.reduction)\",\n \"weight :param avg_factor: avg factor :param reduction_override: reduce override :return:\",\n \"__init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means\",\n \"out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas,\",\n \"for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list\",\n \"predict :param target: target :param weight: weight :param avg_factor: avg\",\n \"- 0.5 * beta) return loss def cross_entropy(pred, label, weight=None,\",\n \"\\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes\",\n \"This program is free software; you can redistribute it and/or\",\n \"* diff / beta, diff - 0.5 * beta) return\",\n \"reduction): \\\"\\\"\\\"Reduce loss compute. :param loss: losses :param reduction: reduce\",\n \"labels :param label_weights: label weights :param label_channels: label channels :return:\",\n \"beta self.reduction = reduction self.loss_weight = loss_weight def forward(self, pred,\",\n \"rights reserved. # This program is free software; you can\",\n \"return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy\",\n \"self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight\",\n \"gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to\",\n \"num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score\",\n \"> 0 diff = torch.abs(pred - target) loss = torch.where(diff\",\n \"self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]),\",\n \"= None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels,\",\n \"# -*- coding: utf-8 -*- # Copyright (C) 2020. Huawei\",\n \"0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name':\",\n \"= use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight =\",\n \"= len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels): anchors\",\n \"0: return loss elif reduction_function == 1: return loss.mean() elif\",\n \"\\\"\\\"\\\"Cross entropy losses. :param pred: predict result :param label: gt\",\n \"if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self,\",\n \"weight if avg_factor is None: loss = reduce_loss(loss, reduction) else:\",\n \"\\\"\\\"\\\" pfunc = partial(func, **kwargs) if kwargs else func map_results\",\n \"CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0):\",\n \"labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls,\",\n \"loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0,\",\n \"beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param pred: predict :param target: target\",\n \"= torch.abs(pred - target) loss = torch.where(diff < beta, 0.5\",\n \"tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self,\",\n \"func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class\",\n \"self.anchor_strides = anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes\",\n \"16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is\",\n \"self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg\",\n \"else self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight *\",\n \"target_means or (.0, .0, .0, .0) self.target_stds = target_stds or\",\n \"from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator\",\n \"**kwargs) if kwargs else func map_results = map(pfunc, *args) return\",\n \"label_weights, label_channels): \\\"\\\"\\\"Expand binary labels. :param labels: labels :param label_weights:\",\n \"used with reduction=\\\"sum\\\"') return loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss\",\n \"if reduction_override else self.reduction) if target.numel() > 0: loss_bbox =\",\n \"pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean',\",\n \"x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds,\",\n \"avg_factor: avg factor :return: loss \\\"\\\"\\\" if weight is not\",\n \"from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from functools\",\n \"(C) 2020. Huawei Technologies Co., Ltd. All rights reserved. #\",\n \"more details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import torch.nn.functional as F\",\n \"(1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if use_sigmoid_cls else\",\n \"bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg,\",\n \"numpy as np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss\",\n \"in range(num_imgs)] valid_flag_list = [] for img_id, img_meta in enumerate(img_metas):\",\n \":param args: args of function :return: result \\\"\\\"\\\" pfunc =\",\n \"and/or modify # it under the terms of the MIT\",\n \"@NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean',\",\n \"assert beta > 0 assert pred.size() == target.size() and target.numel()\",\n \"avg factor :return: loss \\\"\\\"\\\" assert reduction == 'mean' and\",\n \"from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target\",\n \"0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0:\",\n \"labels = labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2,\",\n \"anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales\",\n \"binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy\",\n \"modify # it under the terms of the MIT License.\",\n \"reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss. :param pred: predict result\",\n \"0 assert pred.size() == target.size() and target.numel() > 0 diff\",\n \"cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\"\",\n \":param avg_factor: avg factor :return: loss \\\"\\\"\\\" assert reduction ==\",\n \"reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param pred: predict result :param\",\n \"self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list\",\n \"None: loss = reduce_loss(loss, reduction) else: if reduction == 'mean':\",\n \"anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes =\",\n \"X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x #\",\n \"reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary\",\n \"use_sigmoid self.use_mask = use_mask self.reduction = reduction self.loss_weight = loss_weight\",\n \"anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios))\",\n \"loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0,\",\n \"feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h,\",\n \"self.use_mask = use_mask self.reduction = reduction self.loss_weight = loss_weight if\",\n \"target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes if\",\n \"sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list,\",\n \"= self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class\",\n \"'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub':\",\n \"class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean',\",\n \"featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators = []\",\n \"= min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h,\",\n \"that it will be useful, # but WITHOUT ANY WARRANTY;\",\n \"\\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in (None, 'none', 'mean', 'sum') reduction\",\n \"reduction: reduce function :param avg_factor: avg factor :return: loss \\\"\\\"\\\"\",\n \"reduction self.loss_weight = loss_weight def forward(self, pred, target, weight=None, avg_factor=None,\",\n \"\\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import torch.nn.functional as F import torch.nn\",\n \"vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import\",\n \"!= 'none': raise ValueError('avg_factor can not be used with reduction=\\\"sum\\\"')\",\n \"= self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _ =\",\n \"mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score, label, weight,\",\n \"entropy losses. :param pred: predict result :param label: gt label\",\n \"def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary labels. :param labels: labels\",\n \"self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score and bbox preds.\\\"\\\"\\\"\",\n \"self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors\",\n \"reduction_override else self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight\",\n \"== target.size() and target.numel() > 0 diff = torch.abs(pred -\",\n \"target :param weight: weight :param avg_factor: avg factor :param reduction_override:\",\n \"RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls\",\n \"def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss.\",\n \"weights :param label_channels: label channels :return: binary label and label\",\n \"zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\"\",\n \"reduction = ( reduction_override if reduction_override else self.reduction) loss_cls =\",\n \":param labels: labels :param label_weights: label weights :param label_channels: label\",\n \"self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs)\",\n \"Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to feature map sizes.\\\"\\\"\\\" def __init__(self,\",\n \">= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] - 1]\",\n \"be useful, # but WITHOUT ANY WARRANTY; without even the\",\n \":param desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid\",\n \"/ anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags\",\n \"= cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs):\",\n \"Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox\",\n \"avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss. :param pred: predict result :param\",\n \"weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param pred: predict :param\",\n \"reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta =\",\n \"1.0) self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg =\",\n \"= x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators\",\n \"beta, 0.5 * diff * diff / beta, diff -\",\n \"x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos,\",\n \"h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)),\",\n \"reduce function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" if\",\n \"loss = loss * weight if avg_factor is None: loss\",\n \"partial(func, **kwargs) if kwargs else func map_results = map(pfunc, *args)\",\n \"- target) loss = torch.where(diff < beta, 0.5 * diff\",\n \"= self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return\",\n \"reduce funtion :return: loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function\",\n \"reduction != 'none': raise ValueError('avg_factor can not be used with\",\n \"x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda()\",\n \"anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios =\",\n \"False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling\",\n \"import Config from functools import partial import numpy as np\",\n \"assert reduction_override in (None, 'none', 'mean', 'sum') reduction = (\",\n \"cls score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes =\",\n \"\\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction\",\n \"in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h,\",\n \"= min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)),\",\n \"is not None: loss = loss * weight if avg_factor\",\n \"label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels)\",\n \"or FITNESS FOR A PARTICULAR PURPOSE. See the # MIT\",\n \"CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init\",\n \"reduction_override if reduction_override else self.reduction) if target.numel() > 0: loss_bbox\",\n \"@NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__()\",\n \"loss. :param pred: predict :param target: target :param beta: beta\",\n \"'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling =\",\n \"desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask\",\n \"1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8, 16, 32,\",\n \"label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to feature\",\n \"range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w,\",\n \"loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction self.loss_weight\",\n \"func: function :param args: args of function :return: result \\\"\\\"\\\"\",\n \"0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256,\",\n \"loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False,\",\n \"in enumerate(img_metas): multi_level_flags = [] for i in range(num_levels): anchor_stride\",\n \"multi_level_flags = [] for i in range(num_levels): anchor_stride = self.anchor_strides[i]\",\n \"AnchorGenerator from vega.core.common.config import Config from functools import partial import\",\n \"labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1,\",\n \"32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides =\",\n \"valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi\",\n \"label_channels: label channels :return: binary label and label weights \\\"\\\"\\\"\",\n \"loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses.\",\n \"loss elif reduction_function == 1: return loss.mean() elif reduction_function ==\",\n \"1] = 1 if label_weights is None: bin_label_weights = None\",\n \"is free software; you can redistribute it and/or modify #\",\n \"def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss.\",\n \"super(RpnClsLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score and bbox\",\n \"elif self.use_mask: self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def\",\n \"losses :param weight: weight :param reduction: reduce function :param avg_factor:\",\n \"= _expand_binary_labels(label, weight, pred.size(-1)) if weight is not None: weight\",\n \"bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox = self.loss_bbox(bbox_pred,\",\n \"label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss. :param pred: predict\",\n \"weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels\",\n \"self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors =\",\n \"cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param pred:\",\n \".0, .0, .0) self.target_stds = target_stds or (1.0, 1.0, 1.0,\",\n \"( reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight *\",\n \"funtion :return: loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function ==\",\n \"torch.nn as nn import torch from vega.search_space.networks.network_factory import NetworkFactory from\",\n \"binary label and label weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels),\",\n \"use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss. :param desc:\",\n \"coding: utf-8 -*- # Copyright (C) 2020. Huawei Technologies Co.,\",\n \"# but WITHOUT ANY WARRANTY; without even the implied warranty\",\n \"label, reduction='none') if weight is not None: weight = weight.float()\",\n \"torch.nn.functional as F import torch.nn as nn import torch from\",\n \"bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy\",\n \"target.size() and target.numel() > 0 diff = torch.abs(pred - target)\",\n \"*args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\"\",\n \"This program is distributed in the hope that it will\",\n \"4) loss_bbox = self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox\",\n \"implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR\",\n \"weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in (None,\",\n \"avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param pred: predict result :param label:\",\n \"\\\"\\\"\\\"Forward compute. :param pred: predict :param target: target :param weight:\",\n \"cls score and bbox preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds =\",\n \"avg_factor is None num_rois = pred.size()[0] inds = torch.arange(0, num_rois,\",\n \"under the terms of the MIT License. # This program\",\n \"anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels\",\n \"loss. :param loss: losses :param weight: weight :param reduction: reduce\",\n \"is None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0),\",\n \"factor :return: loss \\\"\\\"\\\" if pred.dim() != label.dim(): label, weight\",\n \"reduce function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" assert\",\n \"(cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) =\",\n \"bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox =\",\n \"def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\"\",\n \"= CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self,\",\n \"else: self.loss_function = cross_entropy def forward(self, cls_score, label, weight, avg_factor,\",\n \"None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor)\",\n \"to feature map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None):\",\n \"label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param pred: predict\",\n \"it and/or modify # it under the terms of the\",\n \"label_weights is None: bin_label_weights = None else: bin_label_weights = label_weights.view(-1,\",\n \"beta: beta :return: loss \\\"\\\"\\\" assert beta > 0 assert\",\n \"vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from functools import\",\n \"= use_mask self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid:\",\n \"ValueError('avg_factor can not be used with reduction=\\\"sum\\\"') return loss def\",\n \"__init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales =\",\n \"= featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h\",\n \"i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i]\",\n \"it will be useful, # but WITHOUT ANY WARRANTY; without\",\n \"cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales,\",\n \"reduction = ( reduction_override if reduction_override else self.reduction) if target.numel()\",\n \"if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1))\",\n \"label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):\",\n \"= loss.sum() / avg_factor elif reduction != 'none': raise ValueError('avg_factor\",\n \"label: gt label :param weight: weight :param reduction: reduce function\",\n \"feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return\",\n \"@NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to feature map sizes.\\\"\\\"\\\"\",\n \"bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox\",\n \":param label_weights: label weights :param label_channels: label channels :return: binary\",\n \"F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce\",\n \"terms of the MIT License. # This program is distributed\",\n \"if avg_factor is None: loss = reduce_loss(loss, reduction) else: if\",\n \"free software; you can redistribute it and/or modify # it\",\n \"= loss * weight if avg_factor is None: loss =\",\n \"gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self,\",\n \"= anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios or\",\n \"return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args,\",\n \"gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return\",\n \"[] for i in range(num_levels): anchor_stride = self.anchor_strides[i] feat_h, feat_w\",\n \"avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred,\",\n \"/ beta, diff - 0.5 * beta) return loss def\",\n \"return loss elif reduction_function == 1: return loss.mean() elif reduction_function\",\n \"FOR A PARTICULAR PURPOSE. See the # MIT License for\",\n \"operators.\\\"\\\"\\\" import torch.nn.functional as F import torch.nn as nn import\",\n \"1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets =\",\n \"Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss()\",\n \"bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput,\",\n \"avg_factor: avg factor :param reduction_override: reduce override :return: loss \\\"\\\"\\\"\",\n \"hope that it will be useful, # but WITHOUT ANY\",\n \"weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels):\",\n \"binary labels. :param labels: labels :param label_weights: label weights :param\",\n \"label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights\",\n \"1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] - 1] =\",\n \"= pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None,\",\n \"Co., Ltd. All rights reserved. # This program is free\",\n \"map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module):\",\n \"program is distributed in the hope that it will be\",\n \"super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels =\",\n \"if reduction_function == 0: return loss elif reduction_function == 1:\",\n \":param avg_factor: avg factor :return: loss \\\"\\\"\\\" loss = F.cross_entropy(pred,\",\n \"img_meta in enumerate(img_metas): multi_level_flags = [] for i in range(num_levels):\",\n \"target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or\",\n \"4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4) loss_bbox =\",\n \"super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction =\",\n \"details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import torch.nn.functional as F import\",\n \"self).__init__() def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas,\",\n \"losses. :param pred: predict result :param label: gt label :param\",\n \"0: bin_labels[inds, labels[inds] - 1] = 1 if label_weights is\",\n \"bin_labels[inds, labels[inds] - 1] = 1 if label_weights is None:\",\n \"reduction) else: if reduction == 'mean': loss = loss.sum() /\",\n \"= 1 if label_weights is None: bin_label_weights = None else:\",\n \"from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config\",\n \"= sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list,\",\n \"**kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target,\",\n \"* self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator)\",\n \"class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False,\",\n \"FITNESS FOR A PARTICULAR PURPOSE. See the # MIT License\",\n \"self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get\",\n \"diff - 0.5 * beta) return loss def cross_entropy(pred, label,\",\n \"anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes = list(\",\n \"cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward\",\n \"bbox preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds = x[1] return cls_scores,\",\n \"'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1,\",\n \"assert pred.size() == target.size() and target.numel() > 0 diff =\",\n \"channels :return: binary label and label weights \\\"\\\"\\\" bin_labels =\",\n \"predict result :param target: target :param label: gt label :param\",\n \"= [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs\",\n \"torch.where(diff < beta, 0.5 * diff * diff / beta,\",\n \".0, .0) self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0)\",\n \"PURPOSE. See the # MIT License for more details. \\\"\\\"\\\"Import\",\n \":param pred: predict :param target: target :param weight: weight :param\",\n \"compute. :param pred: predict :param target: target :param weight: weight\",\n \"Config from functools import partial import numpy as np from\",\n \"target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param pred: predict\",\n \"def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales\",\n \"will be useful, # but WITHOUT ANY WARRANTY; without even\",\n \"losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples):\",\n \"labels: labels :param label_weights: label weights :param label_channels: label channels\",\n \"self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores = x[2][0]\",\n \"pred: predict result :param label: gt label :param weight: weight\",\n \"self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask self.reduction = reduction\",\n \"if label_weights is None: bin_label_weights = None else: bin_label_weights =\",\n \"else: bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def\",\n \"'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x):\",\n \"not None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction,\",\n \"x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']]\",\n \"target :param beta: beta :return: loss \\\"\\\"\\\" assert beta >\",\n \"= anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios\",\n \"in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels =\",\n \"avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\"\",\n \"pred.size(-1)) if weight is not None: weight = weight.float() loss\",\n \":param pred: predict result :param target: target :param label: gt\",\n \"MIT License for more details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import\",\n \"All rights reserved. # This program is free software; you\",\n \"reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in (None, 'none', 'mean',\",\n \"self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x): \\\"\\\"\\\"Get\",\n \"pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else:\",\n \"args: args of function :return: result \\\"\\\"\\\" pfunc = partial(func,\",\n \"loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return\",\n \"sampling=True): self.target_means = target_means or (.0, .0, .0, .0) self.target_stds\",\n \"2.0] self.anchor_strides = anchor_strides or [4, 8, 16, 32, 64]\",\n \"dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None]\",\n \"2020. Huawei Technologies Co., Ltd. All rights reserved. # This\",\n \"self).__init__() def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes,\",\n \"losses :param reduction: reduce funtion :return: loss \\\"\\\"\\\" reduction_function =\",\n \"'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction':\",\n \"\\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >=\",\n \"* diff * diff / beta, diff - 0.5 *\",\n \"\\\"\\\"\\\" if weight is not None: loss = loss *\",\n \"can not be used with reduction=\\\"sum\\\"') return loss def reduce_loss(loss,\",\n \"as nn import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils\",\n \"function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" assert reduction\",\n \"weight, pred.size(-1)) if weight is not None: weight = weight.float()\",\n \"self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape']\",\n \"= beta self.reduction = reduction self.loss_weight = loss_weight def forward(self,\",\n \"function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" loss =\",\n \"num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore,\",\n \"use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means = target_means or (.0, .0, .0,\",\n \"<reponame>Huawei-Ascend/modelzoo<filename>built-in/TensorFlow/Research/cv/image_classification/Cars_for_TensorFlow/automl/vega/search_space/networks/pytorch/operator/rpn.py # -*- coding: utf-8 -*- # Copyright (C) 2020.\",\n \"self.loss_function = mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score,\",\n \"elif reduction != 'none': raise ValueError('avg_factor can not be used\",\n \"smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox\",\n \"diff = torch.abs(pred - target) loss = torch.where(diff < beta,\",\n \"bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze()\",\n \"or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4, 8,\",\n \"score and bbox preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds = x[1]\",\n \"vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import\",\n \"import AnchorGenerator from vega.core.common.config import Config from functools import partial\",\n \"anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors,\",\n \"img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None,\",\n \"from vega.core.common.config import Config from functools import partial import numpy\",\n \"/ anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags)\",\n \"program is free software; you can redistribute it and/or modify\",\n \"loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\"\",\n \"* smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return\",\n \"= pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice =\",\n \"bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights =\",\n \"# Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights\",\n \"\\\"\\\"\\\" if pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight,\",\n \"cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas\",\n \"16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides\",\n \"according to feature map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None,\",\n \"img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap in\",\n \"F import torch.nn as nn import torch from vega.search_space.networks.network_factory import\",\n \"pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss,\",\n \"> 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight,\",\n \"labels. :param labels: labels :param label_weights: label weights :param label_channels:\",\n \"redistribute it and/or modify # it under the terms of\",\n \"{'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr':\",\n \"target: target :param beta: beta :return: loss \\\"\\\"\\\" assert beta\",\n \"reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param loss: losses :param weight:\",\n \"label :param reduction: reduce function :param avg_factor: avg factor :return:\",\n \"anchor_stride = self.anchor_strides[i] feat_h, feat_w = featmap_sizes[i] h, w, _\",\n \"num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels,\",\n \"PARTICULAR PURPOSE. See the # MIT License for more details.\",\n \"@weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param pred:\",\n \"self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor,\",\n \"reduction_function == 0: return loss elif reduction_function == 1: return\",\n \"reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif\",\n \"from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator)\",\n \"\\\"\\\"\\\"Get cls score and bbox preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds\",\n \"self.label_channels = num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner':\",\n \"self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes)\",\n \"beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__() self.beta\",\n \"self.loss_weight = loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None,\",\n \"loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs)\",\n \"bin_label_weights = label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred,\",\n \"= reduce_loss(loss, reduction) else: if reduction == 'mean': loss =\",\n \"loss: losses :param weight: weight :param reduction: reduce function :param\",\n \"x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module):\",\n \"num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights,\",\n \"forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore =\",\n \"loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss,\",\n \"CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels def forward(self, x):\",\n \"losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,\",\n \"num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score, bbox_pred,\",\n \"num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target,\",\n \"label, weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in\",\n \"bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for\",\n \"Technologies Co., Ltd. All rights reserved. # This program is\",\n \"self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8, 16, 32]\",\n \"factor :return: loss \\\"\\\"\\\" assert reduction == 'mean' and avg_factor\",\n \"ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY\",\n \"loss. :param pred: predict result :param label: gt label :param\",\n \"= torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return\",\n \"bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x\",\n \"0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta,\",\n \"in (None, 'none', 'mean', 'sum') reduction = ( reduction_override if\",\n \"avg_factor: avg factor :return: loss \\\"\\\"\\\" loss = F.cross_entropy(pred, label,\",\n \"[multi_level_anchors for _ in range(num_imgs)] valid_flag_list = [] for img_id,\",\n \"gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None,\",\n \"-*- # Copyright (C) 2020. Huawei Technologies Co., Ltd. All\",\n \"all torch operators.\\\"\\\"\\\" import torch.nn.functional as F import torch.nn as\",\n \"in cls_scores] anchor_generators = [] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base,\",\n \"i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list =\",\n \"multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls,\",\n \"bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1,\",\n \"label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module):\",\n \"kwargs else func map_results = map(pfunc, *args) return tuple(map(list, zip(*map_results)))\",\n \"loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get\",\n \"self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy else:\",\n \"the hope that it will be useful, # but WITHOUT\",\n \"reduction_override in (None, 'none', 'mean', 'sum') reduction = ( reduction_override\",\n \"\\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x\",\n \"label_weights: label weights :param label_channels: label channels :return: binary label\",\n \"min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w)\",\n \"label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4)\",\n \"label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox\",\n \"x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_pos, num_total_neg, num_total_samples)\",\n \"use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7,\",\n \"loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth\",\n \"result :param label: gt label :param weight: weight :param reduction:\",\n \"def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss,\",\n \"None: loss = loss * weight if avg_factor is None:\",\n \"out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels\",\n \"RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self,\",\n \"gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list,\",\n \"self.reduction) if target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss(\",\n \":param beta: beta :return: loss \\\"\\\"\\\" assert beta > 0\",\n \"0.5 * diff * diff / beta, diff - 0.5\",\n \"_expand_binary_labels(label, weight, pred.size(-1)) if weight is not None: weight =\",\n \"img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply. :param func:\",\n \"== 0: return loss elif reduction_function == 1: return loss.mean()\",\n \"class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self,\",\n \"valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func,\",\n \"loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy\",\n \"@NetworkFactory.register(NetTypes.Operator) class AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None, num_classes=2,\",\n \"elif reduction_function == 1: return loss.mean() elif reduction_function == 2:\",\n \"See the # MIT License for more details. \\\"\\\"\\\"Import all\",\n \"the terms of the MIT License. # This program is\",\n \"class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to feature map sizes.\\\"\\\"\\\" def\",\n \"\\\"\\\"\\\"Init Cross Entropy loss. :param desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss,\",\n \"Cross Entropy loss. :param desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__()\",\n \"= weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss\",\n \":param loss: losses :param reduction: reduce funtion :return: loss \\\"\\\"\\\"\",\n \"vega.core.common.config import Config from functools import partial import numpy as\",\n \"def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param pred: predict\",\n \"cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:]\",\n \"if reduction_override else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label,\",\n \"RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x):\",\n \"vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self):\",\n \"valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w),\",\n \"anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags)\",\n \"= self.loss_weight * smooth_l1_loss( pred, target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor,\",\n \"flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list,\",\n \"img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class\",\n \"cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def __init__(self):\",\n \"original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list, label_weights_list, bbox_targets_list,\",\n \"anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or [8,\",\n \"loss = loss.sum() / avg_factor elif reduction != 'none': raise\",\n \"loss \\\"\\\"\\\" if weight is not None: loss = loss\",\n \"**kwargs): \\\"\\\"\\\"Forward compute. :param pred: predict :param target: target :param\",\n \"(valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas,\",\n \"def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute. :param loss: losses :param\",\n \"reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss. :param pred: predict result\",\n \"def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param loss:\",\n \"weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is not None:\",\n \"= F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif reduction_function\",\n \"def forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\"\",\n \"list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__()\",\n \"anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors\",\n \"__init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score and\",\n \"sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according to feature map\",\n \"feature map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg\",\n \"forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param\",\n \"__init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss()\",\n \"num_classes if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner',\",\n \"reduce override :return: loss \\\"\\\"\\\" reduction = ( reduction_override if\",\n \"= ( reduction_override if reduction_override else self.reduction) if target.numel() >\",\n \"if inds.numel() > 0: bin_labels[inds, labels[inds] - 1] = 1\",\n \"= multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return\",\n \"reduce_loss(loss, reduction) else: if reduction == 'mean': loss = loss.sum()\",\n \"cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox\",\n \":param label_channels: label channels :return: binary label and label weights\",\n \"weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def\",\n \"distributed in the hope that it will be useful, #\",\n \"input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls\",\n \"avg factor :param reduction_override: reduce override :return: loss \\\"\\\"\\\" reduction\",\n \"0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5,\",\n \"or (.0, .0, .0, .0) self.target_stds = target_stds or (1.0,\",\n \":param target: target :param beta: beta :return: loss \\\"\\\"\\\" assert\",\n \"pred.dim() != label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if\",\n \"[] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors)\",\n \"def forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds\",\n \"def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights, bbox_targets,\",\n \"weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand\",\n \"= x # out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list,\",\n \"'none', 'mean', 'sum') reduction = ( reduction_override if reduction_override else\",\n \"pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds,\",\n \"avg_factor: avg factor :return: loss \\\"\\\"\\\" assert reduction == 'mean'\",\n \"out_channels def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights,\",\n \"loss = reduce_loss(loss, reduction) else: if reduction == 'mean': loss\",\n \"loss: losses :param reduction: reduce funtion :return: loss \\\"\\\"\\\" reduction_function\",\n \"weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth\",\n \"# out=(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes,\",\n \"self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class\",\n \"label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight is not\",\n \"x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds = x[2][1]\",\n \"of the MIT License. # This program is distributed in\",\n \"'MaxIoUAllNegAssigner', 'pos_iou_thr': 0.7, 'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5},\",\n \"= F.cross_entropy(pred, label, reduction='none') if weight is not None: weight\",\n \"map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None, anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg =\",\n \"loss \\\"\\\"\\\" loss = F.cross_entropy(pred, label, reduction='none') if weight is\",\n \"reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute. :param loss: losses :param reduction:\",\n \"\\\"\\\"\\\"Expand binary labels. :param labels: labels :param label_weights: label weights\",\n \"== 1: return loss.mean() elif reduction_function == 2: return loss.sum()\",\n \"**kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def\",\n \"= weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss\",\n \":return: loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0:\",\n \"pred.size() == target.size() and target.numel() > 0 diff = torch.abs(pred\",\n \"x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\"\",\n \"cls_scores = x[0] bbox_preds = x[1] return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator)\",\n \"self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def\",\n \"\\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes\",\n \"useful, # but WITHOUT ANY WARRANTY; without even the implied\",\n \"import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config\",\n \"w, _ = img_meta['pad_shape'] valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h)\",\n \":return: loss \\\"\\\"\\\" if weight is not None: loss =\",\n \"reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None):\",\n \"x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore =\",\n \"@NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__() def\",\n \"you can redistribute it and/or modify # it under the\",\n \"min(int(np.ceil(w / anchor_stride)), feat_w) flags = anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w))\",\n \"label_weights_list, bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means,\",\n \"input.\\\"\\\"\\\" def __init__(self): super(RpnLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls\",\n \"len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for i in\",\n \"avg factor :return: loss \\\"\\\"\\\" if weight is not None:\",\n \"'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0,\",\n \"gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore @NetworkFactory.register(NetTypes.Operator)\",\n \"def loss(self, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get\",\n \":param avg_factor: avg factor :return: loss \\\"\\\"\\\" if pred.dim() !=\",\n \"return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0):\",\n \"multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in range(num_imgs)] valid_flag_list =\",\n \".0) self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels\",\n \"loss compute. :param loss: losses :param reduction: reduce funtion :return:\",\n \"multi_level_anchors = [] for i in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i],\",\n \"pred: predict :param target: target :param beta: beta :return: loss\",\n \"label :param weight: weight :param reduction: reduce function :param avg_factor:\",\n \"inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds,\",\n \"target, weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return\",\n \"__init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\" super(CustomSmoothL1Loss, self).__init__()\",\n \"label weights \\\"\\\"\\\" bin_labels = labels.new_full((labels.size(0), label_channels), 0) inds =\",\n \"nn import torch from vega.search_space.networks.network_factory import NetworkFactory from vega.search_space.networks.net_utils import\",\n \"256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight':\",\n \"= [] for img_id, img_meta in enumerate(img_metas): multi_level_flags = []\",\n \"is None else self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \\\"\\\"\\\"Create\",\n \":param loss: losses :param weight: weight :param reduction: reduce function\",\n \"_expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary labels. :param labels: labels :param\",\n \"self.loss_function = cross_entropy def forward(self, cls_score, label, weight, avg_factor, reduction_override=None,\",\n \"weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss. :param loss: losses :param\",\n \"= anchor_strides or [4, 8, 16, 32, 64] self.anchor_base_sizes =\",\n \"self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox = CustomSmoothL1Loss() self.out_channels = out_channels\",\n \"is not None: weight = weight.float() loss = weight_reduce_loss(loss, weight=weight,\",\n \"\\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth\",\n \"forward(self, x): \\\"\\\"\\\"Get cls score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds =\",\n \"weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return loss def mask_cross_entropy(pred, target, label, reduction='mean',\",\n \":param weight: weight :param avg_factor: avg factor :param reduction_override: reduce\",\n \"\\\"\\\"\\\"Weight reduce loss. :param loss: losses :param weight: weight :param\",\n \"F.cross_entropy(pred, label, reduction='none') if weight is not None: weight =\",\n \"multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply.\",\n \"bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2,\",\n \"__init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss. :param\",\n \":param label: gt label :param weight: weight :param reduction: reduce\",\n \"= list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else self.anchor_base_sizes_cfg super(Anchors,\",\n \"return loss def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary labels. :param\",\n \"AnchorTargetOp(nn.Module): \\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None,\",\n \"for _ in range(num_imgs)] valid_flag_list = [] for img_id, img_meta\",\n \"return loss def mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross\",\n \"A PARTICULAR PURPOSE. See the # MIT License for more\",\n \"= [featmap.size()[-2:] for featmap in cls_scores] anchor_generators = [] for\",\n \"six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class\",\n \"utf-8 -*- # Copyright (C) 2020. Huawei Technologies Co., Ltd.\",\n \"assert reduction == 'mean' and avg_factor is None num_rois =\",\n \"preds.\\\"\\\"\\\" cls_scores = x[0] bbox_preds = x[1] return cls_scores, bbox_preds\",\n \"\\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return loss\",\n \"function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" if pred.dim()\",\n \"= ( reduction_override if reduction_override else self.reduction) loss_cls = self.loss_weight\",\n \"\\\"\\\"\\\"Smooth l1 loss. :param pred: predict :param target: target :param\",\n \"mask_cross_entropy(pred, target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss. :param\",\n \"gt_bboxes, img_metas, gt_bboxes_ignore = x featmap_sizes = [featmap.size()[-2:] for featmap\",\n \"feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes,\",\n \"beta, diff - 0.5 * beta) return loss def cross_entropy(pred,\",\n \"label_channels): \\\"\\\"\\\"Expand binary labels. :param labels: labels :param label_weights: label\",\n \"torch.abs(pred - target) loss = torch.where(diff < beta, 0.5 *\",\n \"0.5}, 'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1,\",\n \"\\\"\\\"\\\"Multi apply. :param func: function :param args: args of function\",\n \"label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice, target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):\",\n \"labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self,\",\n \"import numpy as np from six.moves import map, zip from\",\n \"weight = weight.float() loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return\",\n \"= out_channels def forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels,\",\n \"dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask = use_mask\",\n \"\\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get\",\n \"'mean', 'sum') reduction = ( reduction_override if reduction_override else self.reduction)\",\n \"l1 loss. :param pred: predict :param target: target :param beta:\",\n \"( reduction_override if reduction_override else self.reduction) if target.numel() > 0:\",\n \"self.loss_bbox(bbox_pred, bbox_targets, bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module):\",\n \"\\\"\\\"\\\" reduction = ( reduction_override if reduction_override else self.reduction) if\",\n \"forward(self, cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert\",\n \"binary_cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss. :param\",\n \"is None num_rois = pred.size()[0] inds = torch.arange(0, num_rois, dtype=torch.long,\",\n \"super(Anchors, self).__init__() def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes,\",\n \"label: gt label :param reduction: reduce function :param avg_factor: avg\",\n \"valid_flag_list = [] for img_id, img_meta in enumerate(img_metas): multi_level_flags =\",\n \"target: target :param label: gt label :param reduction: reduce function\",\n \"F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif reduction_function ==\",\n \"= len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for i\",\n \"= target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels = num_classes\",\n \"self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\"\",\n \"bbox_targets_list, bbox_weights_list, num_total_pos,num_total_neg). return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds,\",\n \"loss * weight if avg_factor is None: loss = reduce_loss(loss,\",\n \"if use_sigmoid_cls else 1 self.cfg = Config({'assigner': {'name': 'MaxIoUAllNegAssigner', 'pos_iou_thr':\",\n \"loss = weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels,\",\n \"def multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply. :param func: function :param\",\n \"of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See\",\n \"'neg_iou_thr': tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler',\",\n \"forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore\",\n \"vega.search_space.networks.net_utils import NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import\",\n \"partial import numpy as np from six.moves import map, zip\",\n \"zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2):\",\n \"reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def\",\n \"return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss.\",\n \"0, 'pos_weight': -1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__()\",\n \"return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def\",\n \"and target.numel() > 0 diff = torch.abs(pred - target) loss\",\n \"anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas, gt_bboxes_ignore = x # out=(labels_list,\",\n \":param weight: weight :param reduction: reduce function :param avg_factor: avg\",\n \"bbox_weights, num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss,\",\n \"anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w / anchor_stride)), feat_w) flags =\",\n \"else self.reduction) loss_cls = self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction,\",\n \"from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def\",\n \"AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels,\",\n \"# This program is free software; you can redistribute it\",\n \"super(CustomSmoothL1Loss, self).__init__() self.beta = beta self.reduction = reduction self.loss_weight =\",\n \"but WITHOUT ANY WARRANTY; without even the implied warranty of\",\n \"**kwargs): \\\"\\\"\\\"Multi apply. :param func: function :param args: args of\",\n \"of function :return: result \\\"\\\"\\\" pfunc = partial(func, **kwargs) if\",\n \"reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param pred: predict :param target: target\",\n \"compute.\\\"\\\"\\\" assert reduction_override in (None, 'none', 'mean', 'sum') reduction =\",\n \"software; you can redistribute it and/or modify # it under\",\n \"for featmap in cls_scores] anchor_generators = [] for anchor_base in\",\n \"use_mask self.reduction = reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function\",\n \"= bbox_weights.reshape(-1, 4) bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4)\",\n \"= loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs):\",\n \"and avg_factor is None num_rois = pred.size()[0] inds = torch.arange(0,\",\n \"reduction_function = F._Reduction.get_enum(reduction) if reduction_function == 0: return loss elif\",\n \"'sampler': {'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals':\",\n \"smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param pred: predict :param\",\n \"self.anchor_base_sizes_cfg super(Anchors, self).__init__() def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds,\",\n \":return: result \\\"\\\"\\\" pfunc = partial(func, **kwargs) if kwargs else\",\n \"self.target_means = target_means or (.0, .0, .0, .0) self.target_stds =\",\n \"Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1 loss.\\\"\\\"\\\"\",\n \"predict result :param label: gt label :param weight: weight :param\",\n \"map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn\",\n \"import torch.nn.functional as F import torch.nn as nn import torch\",\n \"pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute. :param pred:\",\n \"np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss\",\n \"anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or [4,\",\n \"\\\"\\\"\\\"Anchor Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True):\",\n \"MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the #\",\n \"def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls = CustomCrossEntropyLoss() self.loss_bbox =\",\n \"self.beta = beta self.reduction = reduction self.loss_weight = loss_weight def\",\n \"forward(self, x): \\\"\\\"\\\"Get x.\\\"\\\"\\\" (cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights,\",\n \"reduction_function == 1: return loss.mean() elif reduction_function == 2: return\",\n \"loss.sum() / avg_factor elif reduction != 'none': raise ValueError('avg_factor can\",\n \"loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4)\",\n \"def __init__(self): super(RpnClsLossInput, self).__init__() def forward(self, x): \\\"\\\"\\\"Get cls score\",\n \"= anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _ in\",\n \"feat_w = featmap_sizes[i] h, w, _ = img_meta['pad_shape'] valid_feat_h =\",\n \"weight is not None: weight = weight.float() loss = weight_reduce_loss(loss,\",\n \"avg factor :return: loss \\\"\\\"\\\" if pred.dim() != label.dim(): label,\",\n \"losses_bbox = multi_apply(self.loss, cls_score, bbox_pred, labels, label_weights, bbox_targets, bbox_weights, num_total_samples=num_total_samples)\",\n \"target.numel() > 0 diff = torch.abs(pred - target) loss =\",\n \"import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput,\",\n \"(.0, .0, .0, .0) self.target_stds = target_stds or (1.0, 1.0,\",\n \"= anchor_ratios or [0.5, 1.0, 2.0] self.anchor_strides = anchor_strides or\",\n \"bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights = bbox_weights.reshape(-1, 4) bbox_pred =\",\n \"Entropy loss. :param desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid\",\n \"args of function :return: result \\\"\\\"\\\" pfunc = partial(func, **kwargs)\",\n \"!= label.dim(): label, weight = _expand_binary_labels(label, weight, pred.size(-1)) if weight\",\n \"num_imgs = len(img_metas) num_levels = len(featmap_sizes) multi_level_anchors = [] for\",\n \"label weights :param label_channels: label channels :return: binary label and\",\n \"map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn Class\",\n \"'none': raise ValueError('avg_factor can not be used with reduction=\\\"sum\\\"') return\",\n \"functools import partial import numpy as np from six.moves import\",\n \"= torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds]\",\n \"[4, 8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if\",\n \"enumerate(img_metas): multi_level_flags = [] for i in range(num_levels): anchor_stride =\",\n \"raise ValueError('avg_factor can not be used with reduction=\\\"sum\\\"') return loss\",\n \"< beta, 0.5 * diff * diff / beta, diff\",\n \"score.\\\"\\\"\\\" cls_scores = x[2][0] bbox_preds = x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda()\",\n \"num_levels = len(featmap_sizes) multi_level_anchors = [] for i in range(num_levels):\",\n \"AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from vega.core.common.config import Config from\",\n \"if kwargs else func map_results = map(pfunc, *args) return tuple(map(list,\",\n \"[featmap.size()[-2:] for featmap in cls_scores] anchor_generators = [] for anchor_base\",\n \"[] for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs =\",\n \"pred: predict result :param target: target :param label: gt label\",\n \"loss \\\"\\\"\\\" assert beta > 0 assert pred.size() == target.size()\",\n \"= x[2][1] gt_bboxes = x[0]['gt_bboxes'].cuda() img_metas = [x[0]['img_meta']] gt_bboxes_ignore =\",\n \"return AnchorTarget(anchor_list, valid_flag_list, gt_bboxes, img_metas, self.target_means, self.target_stds, self.cfg, gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None,\",\n \"self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1,\",\n \"function :return: result \\\"\\\"\\\" pfunc = partial(func, **kwargs) if kwargs\",\n \"reduction=\\\"sum\\\"') return loss def reduce_loss(loss, reduction): \\\"\\\"\\\"Reduce loss compute. :param\",\n \"self.target_stds = target_stds or (1.0, 1.0, 1.0, 1.0) self.label_channels =\",\n \"labels.reshape(-1) label_weights = label_weights.reshape(-1) cls_score = cls_score.permute(0, 2, 3, 1).reshape(-1,\",\n \"None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(), weight,\",\n \"return cls_scores, bbox_preds @NetworkFactory.register(NetTypes.Operator) class RpnLossInput(nn.Module): \\\"\\\"\\\"Rpn loss input.\\\"\\\"\\\" def\",\n \"\\\"\\\"\\\"Get anchors according to feature map sizes.\\\"\\\"\\\" def __init__(self, anchor_base_sizes_cfg=None,\",\n \"self.anchor_scales = anchor_scales or [8, 16, 32] self.anchor_ratios = anchor_ratios\",\n \"_ in range(num_imgs)] valid_flag_list = [] for img_id, img_meta in\",\n \"import torch.nn as nn import torch from vega.search_space.networks.network_factory import NetworkFactory\",\n \"1 if label_weights is None: bin_label_weights = None else: bin_label_weights\",\n \"NetTypes from vega.search_space.networks.pytorch.utils.anchor_utils.anchor_target import AnchorTarget from vega.search_space.networks.pytorch.utils.bbox_utils.anchor_generator import AnchorGenerator from\",\n \"self.reduction = reduction self.loss_weight = loss_weight def forward(self, pred, target,\",\n \"cfg=None, sampling=True): self.target_means = target_means or (.0, .0, .0, .0)\",\n \"labels, label_weights, bbox_targets, bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1)\",\n \"* beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):\",\n \"label_weights.view(-1, 1).expand(label_weights.size(0), label_channels) return bin_labels, bin_label_weights def binary_cross_entropy(pred, label, weight=None,\",\n \"\\\"\\\"\\\"Rpn Class Loss.\\\"\\\"\\\" def __init__(self, out_channels=2): super(RpnClsLoss, self).__init__() self.loss_cls =\",\n \"= self.loss_cls(cls_score, labels, label_weights, avg_factor=num_total_samples) bbox_targets = bbox_targets.reshape(-1, 4) bbox_weights\",\n \":return: loss \\\"\\\"\\\" reduction = ( reduction_override if reduction_override else\",\n \":param label: gt label :param reduction: reduce function :param avg_factor:\",\n \"for anchor_base in self.anchor_base_sizes: anchor_generators.append(AnchorGenerator(anchor_base, self.anchor_scales, self.anchor_ratios)) num_imgs = len(img_metas)\",\n \":param reduction: reduce funtion :return: loss \\\"\\\"\\\" reduction_function = F._Reduction.get_enum(reduction)\",\n \"= loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function\",\n \"def forward(self, x): \\\"\\\"\\\"Create anchor.\\\"\\\"\\\" cls_scores, bbox_preds, gt_bboxes, img_metas, gt_bboxes_ignore\",\n \"super(AnchorTargetOp, self).__init__() def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors,\",\n \"loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function =\",\n \"compute. :param loss: losses :param reduction: reduce funtion :return: loss\",\n \"x): \\\"\\\"\\\"Get cls score and bbox preds.\\\"\\\"\\\" cls_scores = x[0]\",\n \"loss = F.cross_entropy(pred, label, reduction='none') if weight is not None:\",\n \"tuple([-1, 0.3]), 'min_pos_iou': 0.3, 'ignore_iof_thr': 0.5}, 'sampler': {'name': 'RandomSampler', 'num':\",\n \"def forward(self, x): \\\"\\\"\\\"Create X=(anchor_list,valid_flag_list,gt_bboxes,img_metas,).\\\"\\\"\\\" anchor_list, valid_flag_list, original_anchors, gt_bboxes, img_metas,\",\n \"not be used with reduction=\\\"sum\\\"') return loss def reduce_loss(loss, reduction):\",\n \"MIT License. # This program is distributed in the hope\",\n \"= map(pfunc, *args) return tuple(map(list, zip(*map_results))) @NetworkFactory.register(NetTypes.Operator) class RpnClsLoss(nn.Module): \\\"\\\"\\\"Rpn\",\n \"reduction_override: reduce override :return: loss \\\"\\\"\\\" reduction = ( reduction_override\",\n \"weight, beta=self.beta, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_bbox else: return torch.FloatTensor([0.0]).cuda()\",\n \"bbox_weights, avg_factor=num_total_samples) return loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy\",\n \"Target.\\\"\\\"\\\" def __init__(self, target_means=None, target_stds=None, num_classes=2, use_sigmoid_cls=False, cfg=None, sampling=True): self.target_means\",\n \"avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override in (None, 'none',\",\n \"num_total_pos, num_total_neg, num_total_samples) = x losses_cls, losses_bbox = multi_apply(self.loss, cls_score,\",\n \"gt label :param reduction: reduce function :param avg_factor: avg factor\",\n \"weight is not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits(\",\n \"anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors for _\",\n \"'neg_pos_ub': -1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False})\",\n \"def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\",\n \"0.5 * beta) return loss def cross_entropy(pred, label, weight=None, reduction='mean',\",\n \"-1, 'add_gt_as_proposals': False}, 'allowed_border': 0, 'pos_weight': -1, 'debug': False}) self.sampling\",\n \"in the hope that it will be useful, # but\",\n \"entropy loss. :param pred: predict result :param label: gt label\",\n \"reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls @NetworkFactory.register(NetTypes.Operator) class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1\",\n \":param target: target :param weight: weight :param avg_factor: avg factor\",\n \":return: loss \\\"\\\"\\\" assert reduction == 'mean' and avg_factor is\",\n \"target, reduction='mean')[None] def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Weight reduce loss.\",\n \"Huawei Technologies Co., Ltd. All rights reserved. # This program\",\n \"class CustomSmoothL1Loss(nn.Module): \\\"\\\"\\\"Smooth L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0):\",\n \"else: return torch.FloatTensor([0.0]).cuda() @weighted_loss def smooth_l1_loss(pred, target, beta=1.0): \\\"\\\"\\\"Smooth l1\",\n \"self.loss_weight * self.loss_function(cls_score, label, weight, reduction=reduction, avg_factor=avg_factor, **kwargs) return loss_cls\",\n \"= mask_cross_entropy else: self.loss_function = cross_entropy def forward(self, cls_score, label,\",\n \"{'name': 'RandomSampler', 'num': 256, 'pos_fraction': 0.5, 'neg_pos_ub': -1, 'add_gt_as_proposals': False},\",\n \"in range(num_levels): anchors = anchor_generators[i].grid_anchors(featmap_sizes[i], self.anchor_strides[i]) multi_level_anchors.append(anchors) anchor_list = [multi_level_anchors\",\n \"reduction == 'mean': loss = loss.sum() / avg_factor elif reduction\",\n \"weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module): \\\"\\\"\\\"Rpn input.\\\"\\\"\\\" def __init__(self): super(RpnClsLossInput, self).__init__()\",\n \"L1 Loss.\\\"\\\"\\\" def __init__(self, beta=1.0, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init smooth l1\",\n \"loss_weight def forward(self, pred, target, weight=None, avg_factor=None, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward\",\n \"loss_cls, loss_bbox @NetworkFactory.register(NetTypes.Operator) class CustomCrossEntropyLoss(nn.Module): \\\"\\\"\\\"Cross Entropy Loss.\\\"\\\"\\\" def __init__(self,\",\n \"valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list, multi_level_anchors, gt_bboxes, img_metas, gt_bboxes_ignore\",\n \"anchor_scales=None, anchor_ratios=None, anchor_strides=None): self.anchor_base_sizes_cfg = anchor_base_sizes_cfg self.anchor_scales = anchor_scales or\",\n \"reduce function :param avg_factor: avg factor :return: loss \\\"\\\"\\\" loss\",\n \"or [8, 16, 32] self.anchor_ratios = anchor_ratios or [0.5, 1.0,\",\n \"entropy loss. :param pred: predict result :param target: target :param\",\n \":param avg_factor: avg factor :param reduction_override: reduce override :return: loss\",\n \"valid_feat_h = min(int(np.ceil(h / anchor_stride)), feat_h) valid_feat_w = min(int(np.ceil(w /\",\n \"== 'mean' and avg_factor is None num_rois = pred.size()[0] inds\",\n \"gt_bboxes_ignore_list=gt_bboxes_ignore, gt_labels_list=None, label_channels=self.label_channels, sampling=self.sampling) @NetworkFactory.register(NetTypes.Operator) class Anchors(nn.Module): \\\"\\\"\\\"Get anchors according\",\n \"label_channels), 0) inds = torch.nonzero(labels >= 1).squeeze() if inds.numel() >\",\n \"not None: weight = weight.float() loss = F.binary_cross_entropy_with_logits( pred, label.float(),\",\n \"as F import torch.nn as nn import torch from vega.search_space.networks.network_factory\",\n \"WARRANTY; without even the implied warranty of # MERCHANTABILITY or\",\n \"bbox_weights, num_total_samples=num_total_samples) return losses_cls, losses_bbox def loss(self, cls_score, bbox_pred, labels,\",\n \"self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask:\",\n \"Loss.\\\"\\\"\\\" def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy\",\n \"/ avg_factor elif reduction != 'none': raise ValueError('avg_factor can not\",\n \"range(num_imgs)] valid_flag_list = [] for img_id, img_meta in enumerate(img_metas): multi_level_flags\",\n \"= weight_reduce_loss(loss, weight=weight, reduction=reduction, avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights,\",\n \"# MIT License for more details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\"\",\n \"x featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores] anchor_generators =\",\n \"pred: predict :param target: target :param weight: weight :param avg_factor:\",\n \"def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Cross entropy losses. :param\",\n \"label, weight=None, reduction='mean', avg_factor=None): \\\"\\\"\\\"Binary cross entropy loss. :param pred:\",\n \"loss \\\"\\\"\\\" assert reduction == 'mean' and avg_factor is None\",\n \"cls_score.permute(0, 2, 3, 1).reshape(-1, self.out_channels) loss_cls = self.loss_cls(cls_score, labels, label_weights,\",\n \"-1, 'debug': False}) self.sampling = sampling super(AnchorTargetOp, self).__init__() def forward(self,\",\n \"img_metas = [x[0]['img_meta']] gt_bboxes_ignore = x[0]['gt_bboxes_ignore'].cuda() return cls_scores, bbox_preds, gt_bboxes,\",\n \"loss. :param desc: config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid =\",\n \"if self.use_sigmoid: self.loss_function = binary_cross_entropy elif self.use_mask: self.loss_function = mask_cross_entropy\",\n \"avg_factor: avg factor :return: loss \\\"\\\"\\\" if pred.dim() != label.dim():\",\n \"pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor) return\",\n \"target, label, reduction='mean', avg_factor=None): \\\"\\\"\\\"Mask cross entropy loss. :param pred:\",\n \"> 0: bin_labels[inds, labels[inds] - 1] = 1 if label_weights\",\n \"target.numel() > 0: loss_bbox = self.loss_weight * smooth_l1_loss( pred, target,\",\n \"weight :param reduction: reduce function :param avg_factor: avg factor :return:\",\n \"loss. :param pred: predict result :param target: target :param label:\",\n \"64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg is None else\",\n \"target, beta=1.0): \\\"\\\"\\\"Smooth l1 loss. :param pred: predict :param target:\",\n \"\\\"\\\"\\\" assert reduction == 'mean' and avg_factor is None num_rois\",\n \"is distributed in the hope that it will be useful,\",\n \"F.binary_cross_entropy_with_logits( pred, label.float(), weight, reduction='none') loss = weight_reduce_loss(loss, reduction=reduction, avg_factor=avg_factor)\",\n \"avg_factor=avg_factor) return loss def _expand_binary_labels(labels, label_weights, label_channels): \\\"\\\"\\\"Expand binary labels.\",\n \"avg_factor is None: loss = reduce_loss(loss, reduction) else: if reduction\",\n \"8, 16, 32, 64] self.anchor_base_sizes = list( self.anchor_strides) if self.anchor_base_sizes_cfg\",\n \"import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import weighted_loss @NetworkFactory.register(NetTypes.Operator) class RpnClsLossInput(nn.Module):\",\n \"for more details. \\\"\\\"\\\"Import all torch operators.\\\"\\\"\\\" import torch.nn.functional as\",\n \"bbox_weights, num_total_samples): \\\"\\\"\\\"Get loss.\\\"\\\"\\\" labels = labels.reshape(-1) label_weights = label_weights.reshape(-1)\",\n \"def __init__(self, use_sigmoid=False, use_mask=False, reduction='mean', loss_weight=1.0): \\\"\\\"\\\"Init Cross Entropy loss.\",\n \"as np from six.moves import map, zip from vega.search_space.networks.pytorch.losses.reduce_loss import\",\n \"cls_score, label, weight, avg_factor, reduction_override=None, **kwargs): \\\"\\\"\\\"Forward compute.\\\"\\\"\\\" assert reduction_override\",\n \"self).__init__() self.beta = beta self.reduction = reduction self.loss_weight = loss_weight\",\n \"= anchor_generators[i].valid_flags((feat_h, feat_w), (valid_feat_h, valid_feat_w)) multi_level_flags.append(flags) valid_flag_list.append(multi_level_flags) return anchor_list, valid_flag_list,\",\n \"gt label :param weight: weight :param reduction: reduce function :param\",\n \"config dict \\\"\\\"\\\" super(CustomCrossEntropyLoss, self).__init__() self.use_sigmoid = use_sigmoid self.use_mask =\",\n \"> 0 assert pred.size() == target.size() and target.numel() > 0\",\n \"\\\"\\\"\\\"Mask cross entropy loss. :param pred: predict result :param target:\",\n \"# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\",\n \"multi_apply(func, *args, **kwargs): \\\"\\\"\\\"Multi apply. :param func: function :param args:\",\n \"torch.arange(0, num_rois, dtype=torch.long, device=pred.device) pred_slice = pred[inds, label].squeeze(1) return F.binary_cross_entropy_with_logits(pred_slice,\",\n \"torch operators.\\\"\\\"\\\" import torch.nn.functional as F import torch.nn as nn\",\n \"without even the implied warranty of # MERCHANTABILITY or FITNESS\",\n \"apply. :param func: function :param args: args of function :return:\"\n]","isConversational":false}}}],"truncated":false,"partial":false},"paginationData":{"pageIndex":6,"numItemsPerPage":100,"numTotalItems":12866649,"offset":600,"length":100}},"jwt":"eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc3NTY4NjcxMiwic3ViIjoiL2RhdGFzZXRzL3RlcnJ5eXovc3RhcmNvZGVyZGF0YV9uZ3JhbV8xMCIsImV4cCI6MTc3NTY5MDMxMiwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.AhiolOvJNXeSjvUbdWqF7T0ozxGAasD5tt2qeWP1MNr0EJJ77xzUgy93U2jkztJ3TxHR7OWJse86-U_WMyCOBA","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":12866649,"numBytesParquet":126831957853}]},"discussionsStats":{"closed":0,"open":0,"total":0},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[],"user":[]}}">
ngram listlengths 0 82k |
|---|
[
"# Make sure to only set the hash after the",
"central_atom_indices = sorted( map_to_atom_index[task.central_bond[i]] for i in (0, 1) )",
"None: return task_id.decode() task_id = compute.delay(task_json=task.json()).id redis_connection.hset(\"qcgenerator:types\", task_id, task.type) #",
"hits. canonical_molec... |
[
":param str state_start: Verb to describe operation start :param str",
"to execute :param str state_start: Verb to describe operation start",
"seq_snap.add(*build_instance_sequence(instance, OP_SS_D)) if OP_SS_C in instance.node.operations: seq_snap.add(*build_instance_sequence(instance, OP_SS_C)) # Init",
":... |
[
"# Wait until sending file while not self._sending_file: pass #",
"with parameters\") def threadfunc(worker_num, fpath): self._workers_active += 1 try: recver_socket",
"return printerror(\"Couldn't access file due to permission error\") except timeout:",
"path of the storage dir for received files. If storage... |
[
"= is_training self.use_labels = use_labels self.vocab_size = vocab_size self.hidden_size =",
"to next_input_ids next_tokens = ids_tensor((self.batch_size, 3), config.vocab_size) next_attn_mask = tf.cast(ids_tensor((self.batch_size,",
"is a problem with the signature check. # Test passes",
"passes for TFLEDMo... |
[
"x: np.sin(np.pi*x)), ] enabled_act_functions = available_act_functions def get_measurements(self, weights, x,",
"scores in x. Returns: softmax - softmax normalized in dim",
"y_raw y_pred = np.argmax(y_raw, axis=-1) y_prob = softmax(y_raw, axis=-1) if",
"np.abs(x)), ('cos', lambda x: np.cos(np.pi*x)), ('sin '... |
[
"expected_data == imported_data def assert_many_records_match( expected: Sequence[TrackedModel], imported: Sequence[TrackedModel], ignore=frozenset(),",
"from django.urls import reverse from freezegun import freeze_time from lxml",
"def assert_many_records_match( expected: Sequence[TrackedModel], imported: Sequ... |
[
"# 发送HTML邮件 # msg = MIMEText('<html><body><h1>Hello</h1>' + # '<p>send by",
"'plain', 'utf-8') # 发送HTML邮件 # msg = MIMEText('<html><body><h1>Hello</h1>' + #",
"把附件的内容读进来: mime.set_payload(f.read()) # 用Base64编码: encoders.encode_base64(mime) # 添加到MIMEMultipart: msg.attach(mime) msg['From'] =",
"'</body></html>',... |
[
"queue system installed. queue_resources = QueueWorkerResources( number_of_threads=1, number_of_gpus=1, preferred_gpu_toolkit=QueueWorkerResources.GPUToolkit.CUDA, per_thread_memory_limit=5",
"a HPC resources with the LSF queue system installed. queue_resources",
"\"module load cuda/10.1\"] calculation_backend ... |
[
"repeat runtime test for _ in range(self.n_repeat_trials): observation = IDMAgents(",
"iteration.time_point.time_s, ) observation.update_observation(iteration, next_iteration, buffer) profiler.stop() if self.display_results: logger.info(profiler.output_text(unicode=True, color=True))",
"next_iteration = Simulat... |
[
"tf.train.Checkpoint(model=agent) manager = tf.train.CheckpointManager(ckpt, load_path, max_to_keep=None) ckpt.restore(manager.latest_checkpoint) print(\"Restoring from {}\".format(manager.latest_checkpoint))",
"in range(nb_rollout_steps): # Predict next action. action, q, _, _",
"rank == 0 and logdir: if hasat... |
[
"from PySide2.QtWidgets import * from PySide2.QtPrintSupport import QPrinter, QPrintDialog from",
"def print_statement(self): printer = QPrinter() printer.setOutputFileName(os.path.join( os.environ.get('HOME'), '%s_%s.pdf' % (self._transaction.created_at.strftime('%Y%m%d'),",
"from PySide2.QtCore import * from ... |
[
"equal to other. \"\"\" # Check that the other is",
"optional Which columns to read, with 0 being the first.",
"isinstance(ret, YTArray): if inp.units.same_dimensions_as(ret.units): ret.in_units(inp.units) return ret def sanitize_units_add(this_object, other_object,",
"expm1, log1p, sqrt, square, \\ reciproca... |
[
"= serializers.SerializerMethodField() class Meta: model = Post fields = [",
"image class PostListSerializer(serializers.ModelSerializer): url = post_detail_url user = serializers.SerializerMethodField() class",
"'html', ] def get_html(self, obj): return obj.get_markdown() def get_user(self, obj):",
"except: ... |
[
"grid_number=1) self.oHelper.SetButton('Concluir') self.oHelper.SetButton('Fechar') self.oHelper.AssertTrue() @classmethod def tearDownClass(inst): inst.oHelper.TearDown() if __name__",
"import unittest class GTPA036E(unittest.TestCase): @classmethod def setUpClass(inst): inst.oHelper = Webapp()",
"\"T1\", \"D ... |
[
"DjangoFilterBackend from django.http import HttpResponse,FileResponse,JsonResponse from rest_framework_jwt.authentication import JSONWebTokenAuthentication from",
"from rest_framework.generics import ListAPIView from rest_framework.filters import SearchFilter, OrderingFilter from",
"HttpResponse,FileResponse,J... |
[
"def __init__(self, *args, **kwargs): super(HeartbeatCollector, self).__init__(*args, **kwargs) self.hostname = self.get_hostname()",
"self.config: self.connection_timeout = int(self.config['netuitive_connection_timeout']) self.api = netuitive.Client(url=self.config['netuitive_url'], api_key=self.config['netuitiv... |
[
"[exp2_Delta.min(), exp2_Delta.median(), exp2_Delta.std()]) stat2_Delta.index = [\"Best\", \"Median\", \"Std.\"] # find",
"# print(\"{}.Hypervolume:\".format(market), # best2_Hypervolume[0], best2_Hypervolume[1]) # TODO: check error print(\"{}\\n----------------------------------------------\".format(market))",
... |
[
"np.sqrt(cov(x, x, w) * cov(y, y, w)) print('Completed: Computed weighted",
"int(lat.shape[0]),int(lon.shape[0]))) varyravel = np.reshape(vary.copy(), (int(vary.shape[0]*12.),levsq, int(lat.shape[0]),int(lon.shape[0]))) varx_dj = np.empty((int(varx.shape[0]-1),levsq, int(lat.shape[0]),int(lon.shape[0])))",
"for... |
[
"\"C\") as [vi, vj]: C[vi, vj] = B[vi, vj] +",
"vi + 1, vj], (1, 1)) C[vi, vj] = Bb[0,",
"\"C\" def test_compute_inline_under_loop(): sch = tir.Schedule(elementwise_under_loop, debug_mode=True) block_b = sch.get_block(\"B\")",
"128], \"C\") as [vi, vj]: D[vi, vj] = B[vi, vj]",
"D = tir.match_buffer(d, (128,... |
[
"request, redirect import subprocess import tempfile import json import yaml",
"\"%sjobs/%i\" % (request.url_root, self.jobid), \"log\": \"%sjobs/%i/log\" % (request.url_root, self.jobid), \"run\":",
"= self.proc.communicate(self.inputobj) if self.proc.returncode == 0: outobj = yaml.load(self.stdoutdata, Loader... |
[
"2: result = result << 1 = 0000_0000_0000_0000_0000_0000_0000_0100. n&1 =",
"side if (i < 31) { // CATCH: for last",
"binary as 0000_0000_0000_0000_0000_0000_0000_1101, the \"_\" is for readability), calling reverseBits(13)",
"1 therefore result = result + 1 = = 0000_0000_0000_0000_0000_0000_0000_1011",
"00... |
[
"the last in the module, exactly as formatted: __version__ =",
"be the last in the module, exactly as formatted: __version__",
"*must* be the last in the module, exactly as formatted:",
"information.\"\"\" # The following line *must* be the last in",
"\"\"\"Version information.\"\"\" # The following line *m... |
[
"PdfFileWriter import re range_pattern = re.compile(r'(\\d+)(\\.\\.|-)(\\d+)') comma_pattern = re.compile('\\d+(,\\d+)*') def",
"\"usage examle:\\npython3 selective_merge_pdf.py file1.pdf 1-3 file2.pdf 3,4,10 file1.pdf 50\" assert(len(argv)",
"groups = range_pattern.search(pages_str) if groups: start = int(grou... |
[
"detected_vps, image_dims): \"\"\"Measures error in direction from center of detected",
"VPs do not count against the score. Based on log",
"/ height def vp_direction_error(ground_truth_vps, detected_vps, image_dims): \"\"\"Measures error in direction",
"in the location of detected vanishing points. \"Missed\... |
[
"replace=False) args.task_id = int(task_list[proc_id()]) _robots = [\"IIWA\", \"Jaco\", \"Kinova3\", \"Panda\"]",
"type=int, default=-1) parser.add_argument('--task-id', type=int, default=-1) parser.add_argument('--hid', type=int, default=256) parser.add_argument('--l', type=int,",
"default=0.02) parser.add_arg... |
[
"f(): yield 1 g = f() res = g.next() assert",
"object\") def test_throw_finishes_generator(self): def f(): yield 1 g = f()",
"def f(): try: yield 1 v = (yield 2) except:",
"set() def test_explicit_stop_iteration_unpackiterable(self): def f(): yield 1 raise StopIteration assert",
"assert d['f'].x == 42 def t... |
[
"with the same environment of VM' ) subparser.add_argument( '--disk-size', dest='disk_size',",
"use latest hypervisor settings', ) subparser = subparsers.add_parser( 'delete', description=vm_delete.__doc__,",
"any(isinstance(a, _SubParsersAction) for a in self._actions): return super(IGVMArgumentParser, self).f... |
[
"dataset_path = os.path.join(base_path, dataset_name) processor = PROCESSORS[dataset_name.lower()]() train_dataset = processor.get_train_examples(dataset_path)",
"18025 assert len(test_dataset) == 4928 assert test_dataset[0].text_a == \" |",
"os.path import abspath, join root_dir = d(d(d(abspath(__file__)))) sy... |
[
"syntax/format \", RESOURCE_NOT_AVAILABLE : \"Not available\", RESOURCE_ALLOCATE_FAILURE : \"Allocate failure\",",
"0x1004 OPTION_NOT_SUPPORTED = 0x1005 OPTION_UNKNOWN_ERROR = 0x1FFF PARAMETER_INVALID = 0x2000",
"0x4000 RESOURCE_ALLOCATE_FAILURE = 0x4001 RESOURCE_FULL = 0x4002 RESOURCE_OVERFLOW = 0x4003",
"se... |
[
"random class SUN397EncodableDataset(Dataset): \"\"\"SUN397 encodable dataset class\"\"\" def __init__(self, train=True):",
"transforms.Resize((224, 224)), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) ])",
"'data/SUN397/train/*/*.jpg' if train else 'data/SUN... |
[
"All rights reserved. # See LICENSE.txt for complete terms. from",
"= super(Location, self).to_dict() if self._XSI_TYPE: d[\"xsi:type\"] = self._XSI_TYPE return d",
"= common_binding.LocationType _namespace = 'http://cybox.mitre.org/common-2' _XSI_TYPE = None # overridden",
"fields.IdrefField(\"idref\") name ... |
[
"pandas as pd import pysam import matplotlib matplotlib.use(\"agg\") import matplotlib.pyplot",
"not None: softclips.append([clip_right, False, is_first_read]) if i == 10000000: break",
"max(softclips) plt.plot([m, m], [0, 1.0], \":k\") plt.text(m, 1, \"max={}\".format(m), horizontalalignment=\"right\",",
"fu... |
[
"import django.utils.timezone class Migration(migrations.Migration): dependencies = [ ('rss_feeder_api', '0002_feed_subtitle'), ]",
"), migrations.AlterModelOptions( name='feed', options={'ordering': ('-updated_at',), 'verbose_name': 'Feed', 'verbose_name_plural': 'Feeds'}, ),",
"] operations = [ migrations.Alt... |
[
"v2 has BN-ReLU before Pooling x = BatchNormalization()(x) x =",
"res_block == 0: # first layer and first stage activation",
"y = resnet_layer(inputs=y, num_filters=num_filters_in, conv_first=False) y = resnet_layer(inputs=y, num_filters=num_filters_out, kernel_size=1,",
"dimensions activation (string): activ... |
[
"'isSpouseOf'] RELS_TO_INFER = ['hasGrandParent', 'isGrandParentOf', 'hasGreatGrandParent', 'isGreatGrandParentOf', 'isUncleOf', 'hasUncle', 'isGreatUncleOf',",
"{} graph['nodes'] = [] graph['edges'] = [] nodes = {}",
"graph['nodes'] = [] graph['edges'] = [] nodes = {} q",
"?obj . VALUES ?testPred { fhkb:isWi... |
[
"\"kermit\", \"iss\": \"test-issuer\"}) self.assertEqual(channel.result[\"code\"], b\"200\", channel.result) self.assertEqual(channel.json_body[\"user_id\"], \"@kermit:test\") # An",
"\"\"\"An unknown IdP should cause a 400\"\"\" channel = self.make_request(",
"ApplicationService.NS_ALIASES: [], }, ) self.anoth... |
[
"in range(5): publisher.publish(request) sleep(0.5) rclpy.shutdown() print(f'Delivery request submitted to {args.robot_type}')",
"= args.dropoff for _ in range(5): publisher.publish(request) sleep(0.5) rclpy.shutdown() print(f'Delivery",
"dispenser - parameters: [request_guid: xxx, target_guid:cssdbot, transpor... |
[
"log.warning( f\"{route.endpoint} Has exceeded it's ratelimit ({lock.limit})! Reset in {lock.delta}",
"logging.getLogger(logger_name) class GlobalLock: \"\"\"Manages the global ratelimit\"\"\" def __init__(self) ->",
"\"\"\" if bucket_hash := self._endpoints.get(route.rl_bucket): # we have seen this",
"import... |
[
"= environ.get('SECRET_KEY') USER = environ.get('DB_USER') PASSWORD = environ.get('DB_PASSWORD') DB_NAME =",
"False CSRF_ENABLED = True SECRET_KEY = environ.get('SECRET_KEY') USER = environ.get('DB_USER')",
"= environ.get('DB_HOST') SQLALCHEMY_DATABASE_URI = f\"postgresql://{USER}:{PASSWORD}@{HOST}/{DB_NAME}\" ... |
[
"the mnemonic seed must begin with this SEED_PREFIX = '01'",
"return SEED_PREFIX_SW elif seed_type == '2fa': return SEED_PREFIX_2FA elif seed_type",
"<filename>electrum/version.py ELECTRUM_VERSION = '4.1.5-radc' # version of the client package",
"'4.1.5-radc' # version of the client package APK_VERSION = '4.1... |
[
"torch import numpy as np import cv2 from lib.utils.visualize_utils import",
"= p.forward(feat_dim) print(p1) def test_filp(cfg, tb_writer): cfg = copy.deepcopy(cfg) cfg['feature_maps']",
"0') def _setup(self, cfg): num_feat = len(self._steps) for item in",
"map size\"\"\" assert self._prior_cfg is not {} ret... |
[
"s = str( self ) if IECore.FileSequence.fileNameValidator().match( s ) :",
"isinstance( exampleValue, basestring ) : combiner = mostCommon if combiner",
"child ) ) else : nonLeafPaths.append( child ) sequences =",
"values ) / len( values ) def mostCommon( values )",
"path ) : result = self.__basePathSeed.co... |
[
"django.db import migrations class Migration(migrations.Migration): dependencies = [ ('reo', '0117_financialmodel_generator_fuel_escalation_pct'),",
"= [ ('reo', '0117_financialmodel_generator_fuel_escalation_pct'), ('reo', '0120_auto_20210927_2046'), ('reo', '0121_auto_20211012_0305') ] operations",
"from djan... |
[
"\"by%sIsolationMVArun2v1DBnewDMwLT\") tau.idCI3hit = id3(tau, \"by%sCombinedIsolationDeltaBetaCorr3Hits\") tau.idAntiMu = tau.tauID(\"againstMuonLoose3\") + tau.tauID(\"againstMuonTight3\")",
"alltaus: tau.associatedVertex = event.goodVertices[0] if len(event.goodVertices)>0 else event.vertices[0] tau.lepVeto ="... |
[
"e - (b // a) * d, d) def modInvert(a,b):",
"open(\"plaintext.txt\", \"r\") plain1 = plain1File.read().split(\"\\n\")[0] plain2 = plain2File.read().split(\"\\n\")[0] if plain1",
"plain2: print(\"Dosyalar Özdeştir..\") else: print(\"Dosyalar özdeş değildir..\") n = int(input(\"Oluşturulmak",
"işlemi yapılamz. ... |
[
"as predecessors print networkx.descendants(self.networkx, e) # almost as child_to_parents def",
"# write_to_dot def get_subgraph(self, envos=None): \"\"\"Given a list of ENVO",
"* https://github.com/biolab/orange-bio * https://bitbucket.org/biolab/orange-bioinformatics To install: $ pip install Orange-Bioinfor... |
[
"Possible values include: \"undefined\", \"false\", \"true\". :type result: str or",
"to specify whether the fallback route is enabled. :type is_enabled:",
"__init__( self, **kwargs ): super(ExportDevicesRequest, self).__init__(**kwargs) self.export_blob_container_uri = kwargs['export_blob_container_uri'] self.... |
[
"('aliases', ['alias1.yaml']), ('triggers', ['trigger1.yaml', 'cron1.yaml']), ('rules', ['rule1.yaml']), ('triggertypes', ['triggertype1.yaml']), ('executions',",
"'path': '/v1/ruleenforcements/%s' % (enforcement_model.id), 'method': 'GET' }, # Action Executions",
"{ 'path': '/v1/packs/index/health', 'method': ... |
[
"assert len(r.json()[\"data\"][\"tensor\"][\"values\"]) == 1 r = rest_request_ambassador(\"myabtest\",None,API_AMBASSADOR) res = r.json()",
"myabtest --imageA seldonio/mock_classifier:1.0 --imageB seldonio/mock_classifier:1.0 --oauthKey=oauth-key --oauthSecret=oauth-secret && ks apply",
"k8s_utils import * def ... |
[
"# proxy module from __future__ import absolute_import from envisage.safeweakref import",
"<reponame>enthought/etsproxy # proxy module from __future__ import absolute_import from envisage.safeweakref",
"proxy module from __future__ import absolute_import from envisage.safeweakref import *"
] |
[
"models.ManyToManyField('Genre') class Author(models.Model): name = models.CharField(max_length=50) nationality = models.ForeignKey('Nation', null=True)",
"name = models.CharField(max_length=50) nationality = models.ForeignKey('Nation', null=True) class Genre(models.Model): name",
"= models.CharField(max_length... |
[
"- keep_prob) * width**2 / dropblock_size**2 / (width - dropblock_size",
"keepdims=True) else: ksize = [1, 1, dropblock_size, dropblock_size] block_pattern =",
"C]) if data_format == 'channels_last' else tf.reshape(cam, [N, C, height,",
"code from tensorpack.tfutils.tower import get_current_tower_context from... |
[
"tree. # from rlstructures import logging from rlstructures.env_wrappers import GymEnv,",
"import GymEnv, GymEnvInf from rlstructures.tools import weight_init import torch.nn as",
"envs.append(e) return GymEnvInf(envs, seed) def create_agent(model, n_actions=1): return RecurrentAgent(model=model, n_actions=n_ac... |
[
"alt import pydeck as pdk import os import glob from",
"wc = WordCloud(background_color=\"white\", max_words=100, width=2000, height=800, colormap='tab20') wc.generate_from_frequencies(worte) return st.image(wc.to_array())",
"gesamt: {relations.replace(',','.')}\") def systematik(): #Ranking der meistverwendete... |
[
"processed. class NoParkingAvailableError(NoOverlapSpatialJoinError): def __init__(self, message): self.message = message class",
"class NoOverlapSpatialJoinError(Exception): def __init__(self, message): self.message = message # msg:",
"overlapping polygons representing geographic features # please how shapefil... |
[
"port = 42 s_soket = socket(AF_INET, SOCK_DGRAM) s_soket.bind((ip, port)) print(\"\\nSunucu",
"istemciAdres = s_soket.recvfrom(4096) dosyaIcerigi = s_soket.recvfrom(4096) if(os.path.exists(dosyaIsmi.decode(\"utf-8\")) == True): s_soket.sendto(bytes(\"aynisi",
"True): dosya = open(dosyaIsmi.decode(\"utf-8\"), \"... |
[
"custom_order = [ \"method\", \"url\", \"params\", \"headers\", \"cookies\", \"data\", \"json\",",
"Dict]) -> Dict: \"\"\" convert extract list(v2) to dict(v3) Args:",
"deprecated argument: --failfast\") args.pop(args.index(\"--failfast\")) # convert --report-file to --html if",
"args: logger.warning( f\"gene... |
[
"can click-drag a rectangular region to zoom. If you use",
"scatter plot\" bg_color=\"lightgray\" #=============================================================================== # # Demo class that is",
"bg_color=\"lightgray\" #=============================================================================== # #... |
[
"# # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law",
"= 'ANY' SELENIUM_SERVER = None SELENIUM_PORT = 4444 BROWSER_WIDTH =",
"# # Licensed under the Apache License, Version 2.0 (the",
"configuration module of webstr selenium tests. This module provides configuration",
"comp... |
[
"__call__(self, population): chroms_to_mutate = random.sample( population, round(self.chrom_mut_chance * len(population))) for",
"gen_mut_chance): self.chrom_mut_chance = chrom_mut_chance self.gen_mut_chance = gen_mut_chance def __call__(self, population):",
"chrom in chroms_to_mutate: genes_to_mutate = random.... |
[
"for i in range(len(chars)): if (i + 1) % 20",
"* [0] # For each lowercase letter in the list,",
"displayCounts(counts) def createList(): \"\"\"Create a list of characters.\"\"\" # Create",
"def createList(): \"\"\"Create a list of characters.\"\"\" # Create an",
"counts = 26 * [0] # For each lowercase lett... |
[
"you can use the ``Config`` API:: from ddtrace import config,",
"To configure the Vertica integration globally you can use the",
"# override the service and tracer to be used Pin.override(conn,",
"missing_modules: from .patch import patch, unpatch __all__ = [patch, unpatch]",
"use the ``patch`` function. No... |
[
"/>', ] def __init__(self, data): datadriven.DataDrivenTestCase.__init__(self, data) self.data = data",
"OpenIDServiceEndpoint.fromHTML('http://unused.url/', self.data) expected = [] self.failUnlessEqual(expected, actual) def pyUnitTests(): return",
"self.data) expected = [] self.failUnlessEqual(expected, actua... |
[
"TEMPLATE_LOADERS = ( ('django.template.loaders.cached.Loader', ( 'django.template.loaders.filesystem.Loader', 'django.template.loaders.app_directories.Loader', )), ) EMAIL_BACKEND",
"= get_env_var('EMAIL_HOST_PASSWORD') EMAIL_PORT = 587 EMAIL_USE_TLS = True DEFAULT_FROM_EMAIL =",
") EMAIL_BACKEND = 'django.cor... |
[
"# df[column_name].max()) if df[column_name].min() < 0 and df[column_name].max() <= 0:",
"df.columns states = [] g.attr('node', shape='ellipse') for column_name in column_names:",
"def generate_react_flow_chart_from_df(df): column_names = df.columns nodes = {} # Elipses",
"edges.append([state, name1, 'one_way... |
[
"os.path.exists( filename ) : old_month -= 1 # 調査用に+1してあるので、実際の値は、これにold_monthに-1したものとなる。 break",
": return None #df = pd.DataFrame({ # 'start': [None, None],",
"os import json import datetime import pandas as pd from",
"break labels.append( m_year + \"/\" + m_month ) fileNameList.append( filename",
"base.C... |
[
"n, m: (int(end.split(\".\")[n]) - int(start.split(\".\")[n])) * m return calc(0, 256",
"(int(end.split(\".\")[n]) - int(start.split(\".\")[n])) * m return calc(0, 256 * 256",
"- int(start.split(\".\")[n])) * m return calc(0, 256 * 256 *",
"m return calc(0, 256 * 256 * 256) + calc(1,",
"256 * 256 * 256) + c... |
[
"a bug in adm feature passing scale into dwt_quant_step #",
"= cls.get_scores_key('vifdiff_den_scale3') vifdiff_scale0_scores_key = cls.get_scores_key('vifdiff_scale0') vifdiff_scale1_scores_key = cls.get_scores_key('vifdiff_scale1') vifdiff_scale2_scores_key =",
"val = None atom_feature_scores_dict[atom_featur... |
[] |
[
"by susy at 2019/11/8 \"\"\" from dao.dao import DataDao import",
"# -*- coding: utf-8 -*- \"\"\" Created by susy at",
"-*- coding: utf-8 -*- \"\"\" Created by susy at 2019/11/8",
"from dao.models import PanAccounts from cfg import PAN_SERVICE, MASTER_ACCOUNT_ID class",
"import pytz from dao.models import P... |
[
"value, mask=None): if mask is not None: mask = mask.unsqueeze(1)",
"= nn.Dropout(p=dropout) def forward(self, query, key, value, mask=None): if mask",
"nn.Linear(d_model, d_model) self.dropout = nn.Dropout(p=dropout) def forward(self, query, key, value,",
"2) for l, x in zip(self.linear_layers, (query, key, ... |
[
"strategy == \"mh\": print(\"**** \", str(system), \" Job \", str(job),",
"help='Suffix (chosen epoch, e.g. 1981)', required=True) parser.add_argument('-j', '--job', help='Job (0/1)',",
"elif strategy == \"naive\": print(\"**** \", str(system), \" Job \",",
"= argparse.ArgumentParser() parser.add_argument('-s... |
[
"cdll lib = cdll.LoadLibrary(\"../target/release/libembed.dylib\") #=> for Mac #lib = cdll.LoadLibrary(\"../target/release/libembed.so\")",
"# $ python embed.py from ctypes import cdll lib =",
"lib = cdll.LoadLibrary(\"../target/release/libembed.dylib\") #=> for Mac #lib = cdll.LoadLibrary(\"../target/release/l... |
[
"local_var_params: body_params = local_var_params['body'] if isinstance(request, SdkStreamRequest): body_params = request.get_file_stream()",
"SdkStreamRequest): body_params = request.get_file_stream() response_headers = [] header_params['Content-Type'] = http_utils.select_header_content_type(",
"the HTTP reque... |
[
"= False, decreasing: bool = False, multi_spikes_threshold: float = .7,",
"Improvement Objective'] = (self.diagnostics.loc[ self.iter - 1, 'Objective Function'] -",
"after iteration Returns ------- \"\"\" if self.iter == 0: self.diagnostics",
"= init_reweighting_prec self.init_reweighting_prec = init_reweight... |
[
"g.serialize(format='trix') g3 = ConjunctiveGraph() from StringIO import StringIO g3.parse(StringIO(r), format='trix')",
"r2 = URIRef('resource:2') label = URIRef('predicate:label') g1 = Graph(identifier =",
"isinstance(q[3].identifier, URIRef): tg=Graph(store=g.store, identifier=q[3].identifier) else: # BNode,... |
[
"print \"\\033[35mRevision Updated in the database as %d\\033[0m\" % checkout_revision",
"updatelib_inst.updateDatabaseVersion(checkout_version) print \"\\033[35mVersion Updated in the database %s\\033[0m\" % checkout_version",
"sinedon import dbupgrade, dbconfig import updatelib project_dbupgrade = dbupgrade.D... |
[
"from fightchurn.listings.chap8.listing_8_2_logistic_regression import save_regression_summary, save_dataset_predictions def regression_cparam(data_set_path, C_param): X,y =",
"from sklearn.linear_model import LogisticRegression from fightchurn.listings.chap8.listing_8_2_logistic_regression import prepare_data, s... |
[
"3*6*16)) #%% [markdown] # ## ACP # Réduction de la",
"l in clustering.kmeans_labels] plt.scatter(clustering.pca_reduction[:, 0], clustering.pca_reduction[:, 1], c = col)",
"#%% from IPython.display import Image #%% for cl in range(0,19):",
"# img = Tools.display_mosaic(image_array, nrow = 10) # fig =",
"Zo... |
[
"handler_error = logging.FileHandler(os.path.join(this_folder, error_log_file)) # set levels handler_info.setLevel(logging.INFO) handler_error.setLevel(logging.ERROR) #",
"'%(message).180s', datefmt='%Y-%m-%d %H:%M:%S') format_error = \\ logging.Formatter( '%(asctime)s %(levelname)s '",
"= logging.FileHandler(o... |
[
"True, use_propensity = True, test_size = 0) uplift_model.best_params_net y_test, x_test,",
"rmse_tolerance = .05 num_obs = 10000 param_grid = dict(num_nodes=[8], dropout=[.1],",
"t) def test_prepare_data_optimized_loss_one_col_tmt(self): num_obs = 1000 y, x, t =",
"= np.mean((optim_treatments_no_cuttoff_cat=... |
[
"y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1) x1 =",
"from example_module import COLORS, increment class ExampleTest(unittest.TestCase): \"\"\" #TODO \"\"\"",
"#TODO \"\"\" def test_increment(self): x0 = 0 y0 = increment(x0)",
"= 0 y0 = increment(x0) #y0 == 1 self.assertEqual(y0, 1)",
"\"\"\" #TODO \... |
[
"from django.utils.translation import ugettext as _ from desktop.auth import forms",
"+ request.get_host() + '/login/oauth_authenticated/' })) if resp['status'] != '200': raise",
"is_first_login_ever = first_login_ever() backend_names = get_backend_names() is_active_directory = 'LdapBackend' in",
"exc_info=e)... |
[
"= name def get_follow_list(self): return self.follow_list def set_follow_list(self, follow_list): self.follow_list",
"== other.get_follow_list() \\ and self.intention == other.get_intention() and self.lane ==",
"get_name(self): return self.name def set_name(self, name): self.name = name def",
"= intention de... |
[
"re.IGNORECASE) def get_url(url, headers=0, gid=0, sheet=None): parts = parse.urlparse(url) if",
"headers = int(qs['headers'][-1]) if 'gid' in qs: gid = qs['gid'][-1]",
"path = parts.path path = '/'.join((path.rstrip('/'), 'gviz/tq')) qs = parse.parse_qs(parts.query)",
"from __future__ import division from __... |
[
"len(boxes) == 0 and len(labels) == 0: boxes.append([0, 0, 2,",
"self.coco[\"images\"][idx] ) img_path = Path(self.img_folder) / ann_info[\"file_name\"].replace(\".png\", \".jpg\") ann_path =",
"= Image.open(img_path).convert(\"RGB\") w, h = img.size if \"segments_info\" in ann_info:",
"# Instead of finding b... |
[
"middle = left + (right - left) // 2 if",
"equals to nth-item \"\"\" if len(array) <= index: return index",
"search recursively here if left is None and right is",
"0, len(array) - 1 if len(array) == 0: return None",
"O(n) because you are returning the function continuously until index",
"return middle el... |
[
"b, c, dt)) RM.commit() def update_memory(self, a, b): cursor.execute('''UPDATE memory",
"into memory (concept, location, person,DATE) values (?, ?, ?, ?)\",",
"= ? WHERE concept = ? ''', (b, a)) RM.commit()",
"from pattern.en import pprint from pattern.en import parsetree from pattern.en",
"(b, a)) RM.comm... |
[
"= bivariate_normal(X, Y, 1.5, 0.5, 1, 1) Z = 10",
"# to make the bad region transparent. This is the",
"palette.set* lines, you will see # all the defaults; under",
"palette.set_under('g', 1.0) palette.set_bad('b', 1.0) # Alternatively, we could use #",
"range to which the regular palette color scale is ap... |
[
"List, Dict from datetime import datetime class Authenticate(BaseModel): access_token: str",
"typing import List, Dict from datetime import datetime class Authenticate(BaseModel):",
"import List, Dict from datetime import datetime class Authenticate(BaseModel): access_token:",
"validator, Field from typing im... |
[
"\"dankmemes\", \"wholesomememes\", \"memes\", \"terriblefacebookmemes\", \"historymemes\", \"me_irl\", \"2meirl4meirl\", \"fellowkids\", \"tumblr\" ]",
"requests.get(f\"https://api.reddit.com/r/{chosen_sub}/top.json?sort=top&t=day&limit=500\", headers={'User-agent': 'Super Bot 9000'}) r = r.json() boxed =",
"r... |
[
"Registry MultiSZ multi_sz = cli.get_reg_multi_sz(r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', 'Machine') # Output Lines self.output",
"[r'HKLM:\\SYSTEM\\CurrentControlSet\\control\\SecurePipeServers\\winreg\\allowedExactPaths', ('Machine=')] + multi_sz # Recommended... |
[
"MikanException(Exception): \"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot convert a",
"<reponame>dzzhvks94vd2/mikan class MikanException(Exception): \"\"\"Generic Mikan exception\"\"\" class ConversionError(MikanException, ValueError): \"\"\"Cannot",
"class ... |
[
"path('password_reset/done/', auth_views.PasswordResetDoneView.as_view(), name='password_reset_done'), path('reset/<uidb64>/<token>/', auth_views.PasswordResetConfirmView.as_view(), name='password_reset_confirm'), path('reset/done/', auth_views.PasswordResetCompleteView.as_view(), name='password_reset_complete'), p... |
[
"def generate_docker_compose(host, conf_path, extra_vars, extra_vars_file): template = 'template/docker/compose/docker-compose.yml' ansible_arg =",
"_to_backup(local_path) show_diff(local_path, backup_path) upload_conf_file(backup_path, host, container, conf_path, relative_to) restart_container(host, container)",... |
[
"edit distance computes the cost it takes to get from",
"len(source) m = len(target) D = {} # Initialization for",
"j in range(0, m+1): D[0,j] = j for i in",
"in range(1, m+1): if source[i-1] == target[j-1]: D[i,j] = D[i-1,",
"i in range(0, n+1): D[i,0] = i for j in",
"computes the cost it takes to get fr... |
[
"outfile, threads=1, maxmatch=True): delta = f\"{outfile}.tmp.delta\" delta_1 = f\"{outfile}.tmp.1delta\" subprocess.check_output(f\"rm",
": new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in",
"new_record.POS + len(new_record.REF) ] ) if new_record.CHROM not in vcf_records:",
"os.unlink(delt... |
[
"-> Dict[str, Any]: return asdict(self) @dataclass(frozen=True) class Contributor: name: str",
"model for various reports\"\"\" @classmethod def key(cls: Type) -> str:",
"def to_table(self) -> List[str]: return [ 'Pull requests:', '--------------', 'Open",
"+ [f'{c.name.ljust(24)}{c.commit_count}' for c in se... |
[
"datetime_string = self.get(datum, self.field_name) if isinstance(datetime_string, datetime.datetime): to_compare = datetime_string",
"any((self.start, self.end)): return None query_params = dict() if self.start: query_params[\"{}__gte\".format(self.field_name)]",
"if query_params: return queryset.filter(**quer... |
[
"= reduction self.loss_weight = loss_weight if self.use_sigmoid: self.loss_function = binary_cross_entropy",
"diff * diff / beta, diff - 0.5 * beta)",
"0 diff = torch.abs(pred - target) loss = torch.where(diff <",
"torch.nonzero(labels >= 1).squeeze() if inds.numel() > 0: bin_labels[inds, labels[inds] -",
"... |