markdown
stringlengths
0
1.02M
code
stringlengths
0
832k
output
stringlengths
0
1.02M
license
stringlengths
3
36
path
stringlengths
6
265
repo_name
stringlengths
6
127
**Note**: Computing the optimal number of topics for LDA will take a pretty long amount of timeThe time complexity of this process : **O( m * x * p)**where:* m = no. of tokens in the corpus* x = summation of no. of topics given in the range* p = no. of passes for LDA model
lda_model.compute_optimal_topics(processed_data,2,4,1,path) lda = lda_model.Optimal_lda_model('./results/lda_tuning_results1.csv') lda.print_topics()
_____no_output_____
MIT
LDA_New.ipynb
parikshitsaikia1619/LDA_modeing_IQVIA
Step 6: Visualization
import pyLDAvis import pyLDAvis.gensim_models as gensimvis pyLDAvis.enable_notebook() vis = gensimvis.prepare(lda, new_corpus, id_word) vis
_____no_output_____
MIT
LDA_New.ipynb
parikshitsaikia1619/LDA_modeing_IQVIA
Monte Carlo MethodsIn this notebook, you will write your own implementations of many Monte Carlo (MC) algorithms. While we have provided some starter code, you are welcome to erase these hints and write your code from scratch. Part 0: Explore BlackjackEnvWe begin by importing the necessary packages.
import sys import gym import numpy as np from collections import defaultdict from plot_utils import plot_blackjack_values, plot_policy
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Use the code cell below to create an instance of the [Blackjack](https://github.com/openai/gym/blob/master/gym/envs/toy_text/blackjack.py) environment.
env = gym.make('Blackjack-v0')
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Each state is a 3-tuple of:- the player's current sum $\in \{0, 1, \ldots, 31\}$,- the dealer's face up card $\in \{1, \ldots, 10\}$, and- whether or not the player has a usable ace (`no` $=0$, `yes` $=1$).The agent has two potential actions:``` STICK = 0 HIT = 1```Verify this by running the code cell below.
print(env.observation_space) print(env.action_space)
Tuple(Discrete(32), Discrete(11), Discrete(2)) Discrete(2)
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Execute the code cell below to play Blackjack with a random policy. (_The code currently plays Blackjack three times - feel free to change this number, or to run the cell multiple times. The cell is designed for you to get some experience with the output that is returned as the agent interacts with the environment._)
for i_episode in range(10): state = env.reset() while True: print(state) action = env.action_space.sample() state, reward, done, info = env.step(action) if done: print('End game! Reward: ', reward) print('You won :)\n') if reward > 0 else print('You lost :...
(12, 10, False) End game! Reward: -1.0 You lost :( (15, 9, False) (20, 9, False) End game! Reward: 1.0 You won :) (20, 5, False) End game! Reward: -1 You lost :( (8, 4, False) (19, 4, True) (14, 4, False) End game! Reward: 1.0 You won :) (19, 9, False) (20, 9, False) End game! Reward: 1.0 You won :) (14, 6, F...
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Part 1: MC PredictionIn this section, you will write your own implementation of MC prediction (for estimating the action-value function). We will begin by investigating a policy where the player _almost_ always sticks if the sum of her cards exceeds 18. In particular, she selects action `STICK` with 80% probability ...
def generate_episode_from_limit_stochastic(bj_env): episode = [] state = bj_env.reset() while True: probs = [0.8, 0.2] if state[0] > 18 else [0.2, 0.8] action = np.random.choice(np.arange(2), p=probs) next_state, reward, done, info = bj_env.step(action) episode.append((state,...
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Execute the code cell below to play Blackjack with the policy. (*The code currently plays Blackjack three times - feel free to change this number, or to run the cell multiple times. The cell is designed for you to gain some familiarity with the output of the `generate_episode_from_limit_stochastic` function.*)
for i in range(10): print(generate_episode_from_limit_stochastic(env))
[((15, 3, False), 1, -1)] [((13, 9, False), 1, -1)] [((15, 9, False), 1, 0), ((18, 9, False), 1, -1)] [((18, 5, True), 1, 0), ((12, 5, False), 1, 0), ((21, 5, False), 0, 1.0)] [((21, 3, True), 0, 1.0)] [((11, 4, False), 1, 0), ((18, 4, False), 1, 0), ((21, 4, False), 0, 1.0)] [((9, 3, False), 1, 0), ((12, 3, False), 0,...
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Now, you are ready to write your own implementation of MC prediction. Feel free to implement either first-visit or every-visit MC prediction; in the case of the Blackjack environment, the techniques are equivalent.Your algorithm has three arguments:- `env`: This is an instance of an OpenAI Gym environment.- `num_episo...
def mc_prediction_q(env, num_episodes, generate_episode, gamma=1.0): # initialize empty dictionaries of arrays returns_sum = defaultdict(lambda: np.zeros(env.action_space.n)) N = defaultdict(lambda: np.zeros(env.action_space.n)) Q = defaultdict(lambda: np.zeros(env.action_space.n)) # loop over episo...
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Use the cell below to obtain the action-value function estimate $Q$. We have also plotted the corresponding state-value function.To check the accuracy of your implementation, compare the plot below to the corresponding plot in the solutions notebook **Monte_Carlo_Solution.ipynb**.
# obtain the action-value function Q = mc_prediction_q(env, 500000, generate_episode_from_limit_stochastic) # obtain the corresponding state-value function V_to_plot = dict((k,(k[0]>18)*(np.dot([0.8, 0.2],v)) + (k[0]<=18)*(np.dot([0.2, 0.8],v))) \ for k, v in Q.items()) # plot the state-value function plot_b...
Episode 500000/500000.
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Part 2: MC ControlIn this section, you will write your own implementation of constant-$\alpha$ MC control. Your algorithm has four arguments:- `env`: This is an instance of an OpenAI Gym environment.- `num_episodes`: This is the number of episodes that are generated through agent-environment interaction.- `alpha`: Th...
def generate_policy(q_table): policy = {} for state in q_table.keys(): policy[state] = np.argmax(q_table[state]) return policy def take_action(state, policy, epsilon): probs = [0.8, 0.2] if state[0] > 18 else [0.2, 0.8] action = np.random.choice(np.arange(2), p=probs) if (state in policy...
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Use the cell below to obtain the estimated optimal policy and action-value function. Note that you should fill in your own values for the `num_episodes` and `alpha` parameters.
# obtain the estimated optimal policy and action-value function policy, Q = mc_control(env, 1000000, 0.01)
Episode 1000000/1000000.
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Next, we plot the corresponding state-value function.
# obtain the corresponding state-value function V = dict((k,np.max(v)) for k, v in Q.items()) # plot the state-value function plot_blackjack_values(V)
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
Finally, we visualize the policy that is estimated to be optimal.
# plot the policy plot_policy(policy)
_____no_output_____
MIT
monte-carlo/Monte_Carlo.ipynb
romOlivo/NanodegreeExercises
IntroductionIn this notebook, we implement [YOLOv4](https://arxiv.org/pdf/2004.10934.pdf) for training on your own dataset.We also recommend reading our blog post on [Training YOLOv4 on custom data](https://blog.roboflow.ai/training-yolov4-on-a-custom-dataset/) side by side.We will take the following steps to impleme...
# CUDA: Let's check that Nvidia CUDA drivers are already pre-installed and which version is it. !/usr/local/cuda/bin/nvcc --version # We need to install the correct cuDNN according to this output !nvidia-smi # Change the number depending on what GPU is listed above, under NVIDIA-SMI > Name. # Tesla K80: 30 # Tesla P100...
env: compute_capability=75
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
STEP 1. Install cuDNN according to the current CUDA versionColab added cuDNN as an inherent install - so you don't have to do a thing - major win Step 2: Installing Darknet for YOLOv4 on Colab
%cd /content/ %rm -rf darknet #we clone the fork of darknet maintained by roboflow #small changes have been made to configure darknet for training !git clone https://github.com/roboflow-ai/darknet.git %cd /content/darknet/ %rm Makefile #colab occasionally shifts dependencies around, at the time of authorship, this Make...
CUDNN_VERSION=7.6.5.32 __EGL_VENDOR_LIBRARY_DIRS=/usr/lib64-nvidia:/usr/share/glvnd/egl_vendor.d/ LD_LIBRARY_PATH=/usr/lib64-nvidia CLOUDSDK_PYTHON=python3 LANG=en_US.UTF-8 HOSTNAME=d41f1c72eaaf OLDPWD=/ CLOUDSDK_CONFIG=/content/.config NVIDIA_VISIBLE_DEVICES=all DATALAB_SETTINGS_OVERRIDES={"kernelManagerProxyPort":600...
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
Set up Custom Dataset for YOLOv4 We'll use Roboflow to convert our dataset from any format to the YOLO Darknet format. 1. To do so, create a free [Roboflow account](https://app.roboflow.ai).2. Upload your images and their annotations (in any format: VOC XML, COCO JSON, TensorFlow CSV, etc).3. Apply preprocessing and a...
from google.colab import drive drive.mount('/content/drive') #if you already have YOLO darknet format, you can skip this step %cd /content/darknet #!curl -L [YOUR LINK HERE] > roboflow.zip; unzip roboflow.zip; rm roboflow.zip !cp /content/drive/MyDrive/Aquarium-darknet.zip . !pwd !unzip ./Aquarium-darknet.zip #Set up...
/content/darknet
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
Write Custom Training Config for YOLOv4
#we build config dynamically based on number of classes #we build iteratively from base config files. This is the same file shape as cfg/yolo-obj.cfg def file_len(fname): with open(fname) as f: for i, l in enumerate(f): pass return i + 1 num_classes = file_len('train/_darknet.labels') print("writing conf...
[net] batch=64 subdivisions=24 width=416 height=416 channels=3 momentum=0.949 decay=0.0005 angle=0 saturation = 1.5 exposure = 1.5 hue = .1 learning_rate=0.001 burn_in=1000 max_batches=14000 policy=steps steps=11200.0,12600.0 scales=.1,.1 #cutmix=1 mosaic=1 #:104x104 54:52x52 85:26x26 104:13x13 for 416 [convolution...
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
Train Custom YOLOv4 Detector
!./darknet detector train data/obj.data cfg/custom-yolov4-detector.cfg yolov4.conv.137 -dont_show -map #If you get CUDA out of memory adjust subdivisions above! #adjust max batches down for shorter training above
(next mAP calculation at 2568 iterations) Last accuracy mAP@0.5 = 59.40 %, best = 59.40 % 2461: 10.640112, 8.513406 avg loss, 0.001000 rate, 8.034781 seconds, 118128 images, 16.766305 hours left Loaded: 0.000033 seconds ^C
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
Infer Custom Objects with Saved YOLOv4 Weights
#define utility function def imShow(path): import cv2 import matplotlib.pyplot as plt %matplotlib inline image = cv2.imread(path) height, width = image.shape[:2] resized_image = cv2.resize(image,(3*width, 3*height), interpolation = cv2.INTER_CUBIC) fig = plt.gcf() fig.set_size_inches(18, 10) plt.axi...
_____no_output_____
MIT
YOLOv4_Darknet_Roboflow1.ipynb
qwerlarlgus/Darknet
Programming Exercise 2: Logistic Regression IntroductionIn this exercise, you will implement logistic regression and apply it to two different datasets. Before starting on the programming exercise, we strongly recommend watching the video lectures and completing the review questions for the associated topics.All the i...
# used for manipulating directory paths import os # Scientific and vector computation for python import numpy as np # Plotting library from matplotlib import pyplot # Optimization module in scipy from scipy import optimize # library written for this exercise providing additional functions for assignment submission,...
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Submission and GradingAfter completing each part of the assignment, be sure to submit your solutions to the grader. The following is a breakdown of how each part of this exercise is scored.| Section | Part | Submission function | Points | :- |:- ...
# Load data # The first two columns contains the exam scores and the third column # contains the label. data = np.loadtxt(os.path.join('Data', 'ex2data1.txt'), delimiter=',') X, y = data[:, 0:2], data[:, 2]
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
1.1 Visualizing the dataBefore starting to implement any learning algorithm, it is always good to visualize the data if possible. We display the data on a 2-dimensional plot by calling the function `plotData`. You will now complete the code in `plotData` so that it displays a figure where the axes are the two exam sc...
def plotData(X, y): """ Plots the data points X and y into a new figure. Plots the data points with * for the positive examples and o for the negative examples. Parameters ---------- X : array_like An Mx2 matrix representing the dataset. y : array_like Label value...
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Now, we call the implemented function to display the loaded data:
plotData(X, y) # add axes labels pyplot.xlabel('Exam 1 score') pyplot.ylabel('Exam 2 score') pyplot.legend(['Admitted', 'Not admitted']) pass
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
1.2 Implementation 1.2.1 Warmup exercise: sigmoid functionBefore you start with the actual cost function, recall that the logistic regression hypothesis is defined as:$$ h_\theta(x) = g(\theta^T x)$$where function $g$ is the sigmoid function. The sigmoid function is defined as: $$g(z) = \frac{1}{1+e^{-z}}$$.Your first...
def sigmoid(z): """ Compute sigmoid function given the input z. Parameters ---------- z : array_like The input to the sigmoid function. This can be a 1-D vector or a 2-D matrix. Returns ------- g : array_like The computed sigmoid function. g has the sa...
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
The following cell evaluates the sigmoid function at `z=0`. You should get a value of 0.5. You can also try different values for `z` to experiment with the sigmoid function.
# Test the implementation of sigmoid function here z = 0 g = sigmoid(z) print('g(', z, ') = ', g)
g( 0 ) = 0.5
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
After completing a part of the exercise, you can submit your solutions for grading by first adding the function you modified to the submission object, and then sending your function to Coursera for grading. The submission script will prompt you for your login e-mail and submission token. You can obtain a submission tok...
# appends the implemented function in part 1 to the grader object grader[1] = sigmoid # send the added functions to coursera grader for getting a grade on this part grader.grade()
Submitting Solutions | Programming Exercise logistic-regression Login (email address): vvijaytanmai@gmail.com Token: c71I4EgVY0bY6fIL Part Name | Score | Feedback --------- | ----- | -------- Sigmoid Function | 5 ...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
1.2.2 Cost function and gradientNow you will implement the cost function and gradient for logistic regression. Before proceeding we add the intercept term to X.
# Setup the data matrix appropriately, and add ones for the intercept term m, n = X.shape # Add intercept term to X X = np.concatenate([np.ones((m, 1)), X], axis=1)
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Now, complete the code for the function `costFunction` to return the cost and gradient. Recall that the cost function in logistic regression is$$ J(\theta) = \frac{1}{m} \sum_{i=1}^{m} \left[ -y^{(i)} \log\left(h_\theta\left( x^{(i)} \right) \right) - \left( 1 - y^{(i)}\right) \log \left( 1 - h_\theta\left( x^{(i)} \ri...
def costFunction(theta, X, y): """ Compute cost and gradient for logistic regression. Parameters ---------- theta : array_like The parameters for logistic regression. This a vector of shape (n+1, ). X : array_like The input dataset of shape (m x n+1) where m is...
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Once you are done call your `costFunction` using two test cases for $\theta$ by executing the next cell.
# Initialize fitting parameters initial_theta = np.zeros(n+1) cost, grad = costFunction(initial_theta, X, y) print('Cost at initial theta (zeros): {:.3f}'.format(cost)) print('Expected cost (approx): 0.693\n') print('Gradient at initial theta (zeros):') print('\t[{:.4f}, {:.4f}, {:.4f}]'.format(*grad)) print('Expect...
Cost at initial theta (zeros): 0.693 Expected cost (approx): 0.693 Gradient at initial theta (zeros): [-0.1000, -12.0092, -11.2628] Expected gradients (approx): [-0.1000, -12.0092, -11.2628] Cost at test theta: 0.218 Expected cost (approx): 0.218 Gradient at test theta: [0.043, 2.566, 2.647] Expected gradients (a...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
*You should now submit your solutions.*
grader[2] = costFunction grader[3] = costFunction grader.grade()
Submitting Solutions | Programming Exercise logistic-regression Use token from last successful submission (vvijaytanmai@gmail.com)? (Y/n): y Part Name | Score | Feedback --------- | ----- | -------- Sigmoid Function...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
1.2.3 Learning parameters using `scipy.optimize`In the previous assignment, you found the optimal parameters of a linear regression model by implementing gradient descent. You wrote a cost function and calculated its gradient, then took a gradient descent step accordingly. This time, instead of taking gradient descent...
# set options for optimize.minimize options= {'maxiter': 400} # see documention for scipy's optimize.minimize for description about # the different parameters # The function returns an object `OptimizeResult` # We use truncated Newton algorithm for optimization which is # equivalent to MATLAB's fminunc # See https:/...
Cost at theta found by optimize.minimize: 0.203 Expected cost (approx): 0.203 theta: [-25.161, 0.206, 0.201] Expected theta (approx): [-25.161, 0.206, 0.201]
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Once `optimize.minimize` completes, we want to use the final value for $\theta$ to visualize the decision boundary on the training data as shown in the figure below. ![](Figures/decision_boundary1.png)To do so, we have written a function `plotDecisionBoundary` for plotting the decision boundary on top of training data....
# Plot Boundary utils.plotDecisionBoundary(plotData, theta, X, y)
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
1.2.4 Evaluating logistic regressionAfter learning the parameters, you can use the model to predict whether a particular student will be admitted. For a student with an Exam 1 score of 45 and an Exam 2 score of 85, you should expect to see an admissionprobability of 0.776. Another way to evaluate the quality of the pa...
def predict(theta, X): """ Predict whether the label is 0 or 1 using learned logistic regression. Computes the predictions for X using a threshold at 0.5 (i.e., if sigmoid(theta.T*x) >= 0.5, predict 1) Parameters ---------- theta : array_like Parameters for logistic regression....
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
After you have completed the code in `predict`, we proceed to report the training accuracy of your classifier by computing the percentage of examples it got correct.
# Predict probability for a student with score 45 on exam 1 # and score 85 on exam 2 prob = sigmoid(np.dot([1, 45, 85], theta)) print('For a student with scores 45 and 85,' 'we predict an admission probability of {:.3f}'.format(prob)) print('Expected value: 0.775 +/- 0.002\n') # Compute accuracy on our train...
For a student with scores 45 and 85,we predict an admission probability of 0.776 Expected value: 0.775 +/- 0.002 Train Accuracy: 89.00 % Expected accuracy (approx): 89.00 %
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
*You should now submit your solutions.*
grader[4] = predict grader.grade()
Submitting Solutions | Programming Exercise logistic-regression Use token from last successful submission (vvijaytanmai@gmail.com)? (Y/n): y Part Name | Score | Feedback --------- | ----- | -------- Sigmoid Function...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
2 Regularized logistic regressionIn this part of the exercise, you will implement regularized logistic regression to predict whether microchips from a fabrication plant passes quality assurance (QA). During QA, each microchip goes through various tests to ensure it is functioning correctly.Suppose you are the product ...
# Load Data # The first two columns contains the X values and the third column # contains the label (y). data = np.loadtxt(os.path.join('Data', 'ex2data2.txt'), delimiter=',') X = data[:, :2] y = data[:, 2]
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
2.1 Visualize the dataSimilar to the previous parts of this exercise, `plotData` is used to generate a figure, where the axes are the two test scores, and the positive (y = 1, accepted) and negative (y = 0, rejected) examples are shown withdifferent markers.
plotData(X, y) # Labels and Legend pyplot.xlabel('Microchip Test 1') pyplot.ylabel('Microchip Test 2') # Specified in plot order pyplot.legend(['y = 1', 'y = 0'], loc='upper right') pass
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
The above figure shows that our dataset cannot be separated into positive and negative examples by a straight-line through the plot. Therefore, a straight-forward application of logistic regression will not perform well on this dataset since logistic regression will only be able to find a linear decision boundary. 2.2 ...
# Note that mapFeature also adds a column of ones for us, so the intercept # term is handled X = utils.mapFeature(X[:, 0], X[:, 1])
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
2.3 Cost function and gradientNow you will implement code to compute the cost function and gradient for regularized logistic regression. Complete the code for the function `costFunctionReg` below to return the cost and gradient.Recall that the regularized cost function in logistic regression is$$ J(\theta) = \frac{1}{...
def costFunctionReg(theta, X, y, lambda_): """ Compute cost and gradient for logistic regression with regularization. Parameters ---------- theta : array_like Logistic regression parameters. A vector with shape (n, ). n is the number of features including any intercept. If we h...
_____no_output_____
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Once you are done with the `costFunctionReg`, we call it below using the initial value of $\theta$ (initialized to all zeros), and also another test case where $\theta$ is all ones.
# Initialize fitting parameters initial_theta = np.zeros(X.shape[1]) # Set regularization parameter lambda to 1 # DO NOT use `lambda` as a variable name in python # because it is a python keyword lambda_ = 1 # Compute and display initial cost and gradient for regularized logistic # regression cost, grad = costFunctio...
Cost at initial theta (zeros): 0.693 Expected cost (approx) : 0.693 Gradient at initial theta (zeros) - first five values only: [0.0085, 0.0188, 0.0001, 0.0503, 0.0115] Expected gradients (approx) - first five values only: [0.0085, 0.0188, 0.0001, 0.0503, 0.0115] ------------------------------ Cost at test t...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
*You should now submit your solutions.*
grader[5] = costFunctionReg grader[6] = costFunctionReg grader.grade()
Submitting Solutions | Programming Exercise logistic-regression Use token from last successful submission (vvijaytanmai@gmail.com)? (Y/n): y Part Name | Score | Feedback --------- | ----- | -------- Sigmoid Function...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
2.3.1 Learning parameters using `scipy.optimize.minimize`Similar to the previous parts, you will use `optimize.minimize` to learn the optimal parameters $\theta$. If you have completed the cost and gradient for regularized logistic regression (`costFunctionReg`) correctly, you should be able to step through the next p...
# Initialize fitting parameters initial_theta = np.zeros(X.shape[1]) # Set regularization parameter lambda to 1 (you should vary this) lambdas = [0,0.5,1,10,100] # set options for optimize.minimize options= {'maxiter': 100} for lambda_ in lambdas: res = optimize.minimize(costFunctionReg, ...
Train Accuracy: 86.4 % Expected accuracy (with lambda = 1): 83.1 % (approx) Train Accuracy: 82.2 % Expected accuracy (with lambda = 1): 83.1 % (approx) Train Accuracy: 83.1 % Expected accuracy (with lambda = 1): 83.1 % (approx) Train Accuracy: 74.6 % Expected accuracy (with lambda = 1): 83.1 % (approx) Train Accura...
MIT
learning-phase/machine-learning/week-3/tanmai/.ipynb_checkpoints/exercise2-checkpoint.ipynb
Saharsh007/open-qas
Working directoryThe notebooks that you save here persist in the host filesystem.
import os print(os.getcwd()) for d in os.listdir(): print(f'\t {d}')
/home/jovyan/work .ipynb_checkpoints demo.ipynb
Apache-2.0
work/demo.ipynb
vladiuz1/py3-jupyter-docker-compose
requirements.txtThe `requirements.txt` file contains list of modules you want to use in your notebook. The demo `requirements.txt` has flask module, check if below runs without errors:
import flask print('No problem, flask module installed.')
No problem, flask module installed.
Apache-2.0
work/demo.ipynb
vladiuz1/py3-jupyter-docker-compose
Imports
%load_ext autoreload %autoreload 2 import ib_insync print(ib_insync.__all__) import helpers.hdbg as dbg import helpers.hprint as pri import core.explore as exp import im.ib.data.extract.gateway.utils as ibutils
_____no_output_____
BSD-3-Clause
im/ib/data/extract/gateway/notebooks/Task114_Download_futures_price_data_from_IB.ipynb
alphamatic/amp
Connect
ib = ibutils.ib_connect(client_id=100, is_notebook=True)
_____no_output_____
BSD-3-Clause
im/ib/data/extract/gateway/notebooks/Task114_Download_futures_price_data_from_IB.ipynb
alphamatic/amp
Historical data
import logging #dbg.init_logger(verbosity=logging.DEBUG) dbg.init_logger(verbosity=logging.INFO) import datetime # start_ts = pd.to_datetime(pd.Timestamp("2018-02-01 06:00:00").tz_localize(tz="America/New_York")) # import datetime # #datetime.datetime.combine(start_ts, datetime.time()) # dt = start_ts.to_pydatetime() ...
_____no_output_____
BSD-3-Clause
im/ib/data/extract/gateway/notebooks/Task114_Download_futures_price_data_from_IB.ipynb
alphamatic/amp
%load_ext autoreload %autoreload 2 import ib_insync print(ib_insync.__all__) import helpers.hdbg as dbg import helpers.hprint as pri import core.explore as exp import im.ib.data.extract.gateway.utils as ibutils import logging dbg.init_logger(verbosity=logging.DEBUG) #dbg.init_logger(verbosity=logging.INFO) import pa...
(datetime.datetime(2021, 2, 18, 23, 59, 59, tzinfo=<DstTzInfo 'America/New_York' EST-1 day, 19:00:00 STD>), Timestamp('2021-02-18 18:00:00-0500', tz='America/New_York')) (Timestamp('2021-02-18 18:00:00-0500', tz='America/New_York'), Timestamp('2021-02-17 18:00:00-0500', tz='America/New_York')) (Timestamp('2021-02-17 18...
BSD-3-Clause
im/ib/data/extract/gateway/notebooks/Task114_Download_futures_price_data_from_IB.ipynb
alphamatic/amp
As explained in the [Composing Data](Composing_Data.ipynb) and [Containers](Containers.ipynb) tutorials, HoloViews allows you to build up hierarchical containers that express the natural relationships between your data items, in whatever multidimensional space best characterizes your application domain. Once your data...
import numpy as np import holoviews as hv hv.notebook_extension() %opts Layout [fig_size=125] Points [size_index=None] (s=50) Scatter3D [size_index=None] %opts Bounds (linewidth=2 color='k') {+axiswise} Text (fontsize=16 color='k') Image (cmap='Reds')
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
1D Elements: Slicing and indexing Certain Chart elements support both single-dimensional indexing and slicing: ``Scatter``, ``Curve``, ``Histogram``, and ``ErrorBars``. Here we'll look at how we can easily slice a ``Histogram`` to select a subregion of it:
np.random.seed(42) edges, data = np.histogram(np.random.randn(100)) hist = hv.Histogram(edges, data) subregion = hist[0:1] hist * subregion
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
The two bins in a different color show the selected region, overlaid on top of the full histogram. We can also access the value for a specific bin in the ``Histogram``. A continuous-valued index that falls inside a particular bin will return the corresponding value or frequency.
hist[0.25], hist[0.5], hist[0.55]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
We can slice a ``Curve`` the same way:
xs = np.linspace(0, np.pi*2, 21) curve = hv.Curve((xs, np.sin(xs))) subregion = curve[np.pi/2:np.pi*1.5] curve * subregion * hv.Scatter(curve)
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Here again the region in a different color is the specified subregion, and we've also marked each discrete point with a dot using the ``Scatter`` ``Element``. As before we can also get the value for a specific sample point; whatever x-index is provided will snap to the closest sample point and return the dependent val...
curve[4.05], curve[4.1], curve[4.17], curve[4.3]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
It is important to note that an index (or a list of indices, as for the 2D and 3D cases below) will always return the raw indexed (dependent) value, i.e. a number. A slice (indicated with `:`), on the other hand, will retain the Element type even in cases where the plot might not be useful, such as having only a singl...
curve[4:4.5]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
2D and 3D Elements: slicing For data defined in a 2D space, there are 2D equivalents of the 1D Curve and Scatter types. A ``Points``, for example, can be thought of as a number of points in a 2D space.
r = np.arange(0, 1, 0.005) xs, ys = (r * fn(85*np.pi*r) for fn in (np.cos, np.sin)) paths = hv.Points((xs, ys)) paths + paths[0:1, 0:1]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
However, indexing is not supported in this space, because there could be many possible points near a given set of coordinates, and finding the nearest one would require a search across potentially incommensurable dimensions, which is poorly defined and difficult to support.Slicing in 3D works much like slicing in 2D, b...
xs = np.linspace(0, np.pi*8, 201) scatter = hv.Scatter3D((xs, np.sin(xs), np.cos(xs))) scatter + scatter[5:10, :, 0:]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
2D Raster and Image: slicing and indexingRaster and the various other image-like objects (Images, RGB, HSV, etc.) can all sliced and indexed, as can Surface, because they all have an underlying regular grid of key dimension values:
%opts Image (cmap='Blues') Bounds (color='red') np.random.seed(0) extents = (0, 0, 10, 10) img = hv.Image(np.random.rand(10, 10), bounds=extents) img_slice = img[1:9,4:5] box = hv.Bounds((1,4,9,5)) img*box + img_slice img[4.2,4.2], img[4.3,4.2], img[5.0,4.2]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
SamplingSampling is essentially a process of indexing an Element at multiple index locations, and collecting the results. Thus any Element that can be indexed can also be sampled. Compared to regular indexing, sampling is different in that multiple indices may be supplied at the same time. Also, indexing will only...
img_coords = hv.Points(img.table(), extents=extents) labeled_img = img * img_coords * hv.Points([img.closest([(5.1,4.9)])]).opts(style=dict(color='r')) img + labeled_img + img.sample([(5.1,4.9)]) img[5.1,4.9]
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Here, the output of the indexing operation is the value (0.1965823616800535) from the location closest to the specified , whereas ``.sample()`` returns a Table that lists both the coordinates *and* the value, and slicing (in previous section) returns an Element of the same type, not a Table.Next we can try sampling alo...
sampled = img.sample(y=5) labeled_img = img * img_coords * hv.Points(zip(sampled['x'], [img.closest(y=5)]*10)) img + labeled_img + sampled
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Sampling works on any regularly sampled Element type. For example, we can select multiple samples along the x-axis of a Curve.
xs = np.arange(10) samples = [2, 4, 6, 8] curve = hv.Curve(zip(xs, np.sin(xs))) curve_samples = hv.Scatter(zip(xs, [0] * 10)) * hv.Scatter(zip(samples, [0]*len(samples))) curve + curve_samples + curve.sample(samples)
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Sampling HoloMapsSampling is often useful when you have more data than you wish to visualize or analyze at one time. First, let's create a HoloMap containing a number of observations of some noisy data.
obs_hmap = hv.HoloMap({i: hv.Image(np.random.randn(10, 10), bounds=extents) for i in range(3)}, kdims=['Observation'])
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
HoloMaps also provide additional functionality to perform regular sampling on your data. In this case we'll take 3x3 subsamples of each of the Images.
sample_style = dict(edgecolors='k', alpha=1) all_samples = obs_hmap.table().to.scatter3d().opts(style=dict(alpha=0.15)) sampled = obs_hmap.sample((3,3)) subsamples = sampled.to.scatter3d().opts(style=sample_style) all_samples * subsamples + sampled
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
By supplying bounds in as a (left, bottom, right, top) tuple we can also sample a subregion of our images:
sampled = obs_hmap.sample((3,3), bounds=(2,5,5,10)) subsamples = sampled.to.scatter3d().opts(style=sample_style) all_samples * subsamples + sampled
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Since this kind of sampling is only well supported for continuous coordinate systems, we can only apply this kind of sampling to Image types for now. Sampling ChartsSampling Chart-type Elements like Curve, Scatter, Histogram is only supported by providing an explicit list of samples, since those Elements have no underl...
xs = np.arange(10) extents = (0, 0, 2, 10) curve = hv.HoloMap({(i) : hv.Curve(zip(xs, np.sin(xs)*i)) for i in np.linspace(0.5, 1.5, 3)}, kdims=['Observation']) all_samples = curve.table().to.points() sampled = curve.sample([0, 2, 4, 6, 8]) sampling = all_samples * sampled.to.point...
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
Alternatively, you can always deconstruct your data into a Table (see the [Columnar Data](Columnar_Data.ipynb) tutorial) and perform ``select`` operations instead. This is also the easiest way to sample ``NdElement`` types like Bars. Individual samples should be supplied as a set, while ranges can be specified as a two...
sampled = curve.table().select(Observation=(0, 1.1), x={0, 2, 4, 6, 8}) sampling = all_samples * sampled.to.points(extents=extents).opts(style=dict(color='r')) sampling + sampled
_____no_output_____
BSD-3-Clause
doc/Tutorials/Sampling_Data.ipynb
stuarteberg/holoviews
C6.01: Introduction
from sklearn.datasets.samples_generator import make_blobs n_clusters = 2 centers = [[-1,-1],[1,1]] stdevs = [0.4, 0.6] X, y = make_blobs(n_samples = 20, centers = centers, cluster_std = stdevs, random_state = 1200000) from sklearn.cluster import KMeans model = KMeans(n_clusters = n_clusters) preds = model.fit_predict...
_____no_output_____
MIT
course/3_unsupervised_learning/01_clustering/UL6_PCA_Storyboard_Assets.ipynb
claudiocmp/udacity-dsnd
C6.09: Dimensionality Reduction
n_points = [4, 7, 9, 9, 7, 4] n_rooms = [3, 4, 5, 6, 7, 8] size_lower = [ 600, 800, 1000, 1300, 1600, 2000] size_upper = [1000, 1300, 1600, 2000, 2400, 3000] np.random.seed(147258369) rooms = [] sizes = [] for i in range(6): # use beta dist to generate sizes, find bounds with # lower @ 0.2, upper @ 0.8. ...
_____no_output_____
MIT
course/3_unsupervised_learning/01_clustering/UL6_PCA_Storyboard_Assets.ipynb
claudiocmp/udacity-dsnd
C6.10: PCA Properties
alt_slope = np.array([1.3,1.1]) alt_slope = alt_slope / np.sqrt((alt_slope ** 2).sum()) alt_center = np.array([-.35,-.35]) X_alt = np.matmul( np.dot(X - alt_center, alt_slope).reshape(-1,1), alt_slope.reshape(1,-1) ) + alt_center plt.figure(figsize = [15,15]) plt.scatter(X[:,0], X[:,1], s = 64, c = plot_colors[-1]) p...
_____no_output_____
MIT
course/3_unsupervised_learning/01_clustering/UL6_PCA_Storyboard_Assets.ipynb
claudiocmp/udacity-dsnd
Real caseFirst, the real case is relatively simple.The integral that we want to do is:$$k_\Delta(\tau) = \frac{1}{\Delta^2}\int_{t_i-\Delta/2}^{t_i+\Delta/2} \mathrm{d}t \,\int_{t_j-\Delta/2}^{t_j+\Delta/2}\mathrm{d}t^\prime\,k(t - t^\prime)$$For celerite kernels it helps to make the assumtion that $t_j + \Delta/2 < t...
import sympy as sm cr = sm.symbols("cr", positive=True) ti, tj, dt, t, tp = sm.symbols("ti, tj, dt, t, tp", real=True) k = sm.exp(-cr*(t - tp)) k0 = k.subs([(t, ti), (tp, tj)]) kint = sm.simplify(sm.integrate( sm.integrate(k, (t, ti-dt/2, ti+dt/2)), (tp, tj-dt/2, tj+dt/2)) / dt**2) res = sm.simplify(kint / k0)...
(exp(2*cr*dt) - 2*exp(cr*dt) + 1)*exp(-cr*dt)/(cr**2*dt**2)
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
This is the factor that we want.Let's make sure that it is identical to what we have in the note.
kD = 2 * (sm.cosh(cr*dt) - 1) / (cr*dt)**2 sm.simplify(res.expand() - kD.expand())
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Excellent.Let's double check that this reduces to the original kernel in the limit $\Delta \to 0$:
sm.limit(kD, dt, 0)
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Complex caseThe complex cases proceeds similarly, but it's a bit more involved.In this case,$$k(\tau) = (a + i\,b)\,\exp(-(c+i\,d)\,(t_i-t_j))$$
a, b, c, d = sm.symbols("a, b, c, d", real=True, positive=True) k = sm.exp(-(c + sm.I*d) * (t - tp)) k0 = k.subs([(t, ti), (tp, tj)]) kint = sm.simplify(sm.integrate(k, (t, ti-dt/2, ti+dt/2)) / dt) kint = sm.simplify(sm.integrate(kint.expand(), (tp, tj-dt/2, tj+dt/2)) / dt) print(sm.simplify(kint / k0))
(2*cos(I*dt*(c + I*d)) - 2)/(dt**2*(c**2 + 2*I*c*d - d**2))
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
That doesn't look so bad!But, I'm going to re-write it by hand and make sure that it's correct:
coeff = (c-sm.I*d)**2 / (dt*(c**2+d**2))**2 coeff *= (sm.exp((c+sm.I*d)*dt) + sm.exp(-(c+sm.I*d)*dt)-2) sm.simplify(coeff * k0 - kint)
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Good.Now we need to work out nice expressions for the real and imaginary parts of this.First, the real part.I found that it was easiest to look at the prefactors for the trig functions directly and simplify those.Here we go:
res = (a+sm.I*b) * coeff A = sm.simplify((res.expand(complex=True) + sm.conjugate(res).expand(complex=True)) / 2) sm.simplify(sm.poly(A, sm.cos(dt*d)).coeff_monomial(sm.cos(dt*d))) sm.simplify(sm.poly(sm.poly(A, sm.cos(dt*d)).coeff_monomial(1), sm.sin(dt*d)).coeff_monomial(sm.sin(dt*d))) sm.simplify(sm...
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Then, same thing for the imaginary part:
B = sm.simplify(-sm.I * (res.expand(complex=True) - sm.conjugate(res).expand(complex=True)) / 2) sm.simplify(sm.poly(B, sm.cos(dt*d)).coeff_monomial(sm.cos(dt*d))) sm.simplify(sm.poly(sm.poly(B, sm.cos(dt*d)).coeff_monomial(1), sm.sin(dt*d)).coeff_monomial(sm.sin(dt*d))) sm.simplify(sm.poly(sm....
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Ok.Now let's make sure that the simplified expressions are right.
C1 = (a*c**2 - a*d**2 + 2*b*c*d) C2 = (b*c**2 - b*d**2 - 2*a*c*d) cos_term = (sm.exp(c*dt) + sm.exp(-c*dt)) * sm.cos(d*dt) - 2 sin_term = (sm.exp(c*dt) - sm.exp(-c*dt)) * sm.sin(d*dt) denom = dt**2 * (c**2 + d**2)**2 A0 = (C1 * cos_term - C2 * sin_term) / denom B0 = (C2 * cos_term + C1 * sin_term) / denom sm.simplify(...
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Finally let's rewrite things in terms of hyperbolic trig functions.
sm.simplify(2*(sm.cosh(c*dt) * sm.cos(d*dt) - 1).expand() - cos_term.expand()) sm.simplify(2*(sm.sinh(c*dt) * sm.sin(d*dt)).expand() - sin_term.expand())
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Looks good!Let's make sure that this actually reproduces the target integral:
sm.simplify(((a+sm.I*b)*kint/k0 - (A+sm.I*B)).expand(complex=True))
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Finally, let's make sure that this reduces to the original kernel when $\Delta \to 0$:
sm.limit(A, dt, 0), sm.limit(B, dt, 0)
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Overlapping exposures & the power spectrumIf we directly evaluate the power spectrum of this kernel, we'll have some issues because there will be power from lags where our assumption of non-overlapping exposures will break down.Instead, we can evaluate the correct power spectrum by realizing that the integrals that we...
omega = sm.symbols("omega", real=True) sm.simplify(sm.integrate(sm.exp(sm.I * t * omega) / dt, (t, -dt / 2, dt / 2)))
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Therefore, the integrated power spectrum is$$S_\Delta(\omega) = \frac{\sin^2(\Delta\,\omega/2)}{(\Delta\,\omega/2)^2}\,S(\omega) = \mathrm{sinc}^2(\Delta\,\omega/2)\,S(\omega)$$ For overlapping exposures, some care must be taken when computing the autocorrelation because of the absolute value.This also means that celer...
tau = sm.symbols("tau", real=True, positive=True) kp = sm.exp(-cr*(t - tp)) km = sm.exp(-cr*(tp - t)) k1 = sm.simplify(sm.integrate( sm.integrate(kp, (tp, tj-dt/2, tj+dt/2)), (t, tj+dt/2, ti+dt/2)) / dt**2) k2 = sm.simplify(sm.integrate( sm.integrate(kp, (tp, tj-dt/2, t)), (t, ti-dt/2, tj+dt/2)) / dt**2...
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Ok. That's the result for the real case. Now let's work through the result for the complex case.
arg1 = ((a+sm.I*b) * kint.subs([(cr, c+sm.I*d)])).expand(complex=True) arg2 = ((a-sm.I*b) * kint.subs([(cr, c-sm.I*d)])).expand(complex=True) res = sm.simplify((arg1 + arg2) / 2) res C1 = (a*c**2 - a*d**2 + 2*b*c*d) C2 = (b*c**2 - b*d**2 - 2*a*c*d) denom = dt**2 * (c**2 + d**2)**2 dpt = dt + tau dmt = dt - tau cos_ter...
_____no_output_____
MIT
paper/proofs/celerite-integral.ipynb
exowanderer/exoplanet
Module1> module 1
#export def nothing(): pass
_____no_output_____
Apache-2.0
module1.ipynb
hamelsmu/nbdev_export_demo
目标规划**基本上分为两种思路: 加权系数法(转化为单一目标), 优先等级法(按重要程度不同, 转化为单目标模型)**正负偏差变量: $d_i^+=max\{f_i-d_i^0,0\},\ d_i^-=-min\{f_i-d_i^0,0\}$, 显然恒有$d_i^+\times d_i^-=0$
# 例16.3 # 该题的规划模型如下: # min z =P[0]*dminus[0] + P[1]*(dplus[1]+dminus[1]) + P[2]*(3 * (dplus[2]+dminus[2]) + dplus[3]) # 2*x[0] + 2*x[1] <= 12 # 200*x[0] + 300*x[1] + dminus[0] - dplus[0] = 1500 # 2*x[0] - x[1] + dminus[1] - dplus[1] = 0 # 4*x[0] + dminus[2] - dplus[2] = 16 # 5*x[1] + dminus[3] - dpl...
_____no_output_____
MIT
.ipynb_checkpoints/16 目标规划-checkpoint.ipynb
ZDDWLIG/math-model
Fully-Connected Neural NetsIn this exercise we will implement fully-connected networks using a modular approach. For each layer we will implement a `forward` and a `backward` function. The `forward` function will receive inputs, weights, and other parameters and will return both an output and a `cache` object storing ...
# As usual, a bit of setup from __future__ import print_function import time import numpy as np import matplotlib.pyplot as plt from cs231n.classifiers.fc_net import * from cs231n.data_utils import get_CIFAR10_data from cs231n.gradient_check import eval_numerical_gradient, eval_numerical_gradient_array from cs231n.solv...
('X_train: ', (49000, 3, 32, 32)) ('y_train: ', (49000,)) ('X_val: ', (1000, 3, 32, 32)) ('y_val: ', (1000,)) ('X_test: ', (1000, 3, 32, 32)) ('y_test: ', (1000,))
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Affine layer: forwardOpen the file `cs231n/layers.py` and implement the `affine_forward` function.Once you are done you can test your implementaion by running the following:
# Test the affine_forward function num_inputs = 2 input_shape = (4, 5, 6) output_dim = 3 input_size = num_inputs * np.prod(input_shape) weight_size = output_dim * np.prod(input_shape) x = np.linspace(-0.1, 0.5, num=input_size).reshape(num_inputs, *input_shape) w = np.linspace(-0.2, 0.3, num=weight_size).reshape(np.p...
Testing affine_forward function: difference: 9.769849468192957e-10
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Affine layer: backwardNow implement the `affine_backward` function and test your implementation using numeric gradient checking.
# Test the affine_backward function np.random.seed(231) x = np.random.randn(10, 2, 3) w = np.random.randn(6, 5) b = np.random.randn(5) dout = np.random.randn(10, 5) dx_num = eval_numerical_gradient_array(lambda x: affine_forward(x, w, b)[0], x, dout) dw_num = eval_numerical_gradient_array(lambda w: affine_forward(x, w...
Testing affine_backward function: dx error: 5.399100368651805e-11 dw error: 9.904211865398145e-11 db error: 2.4122867568119087e-11
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
ReLU activation: forwardImplement the forward pass for the ReLU activation function in the `relu_forward` function and test your implementation using the following:
# Test the relu_forward function x = np.linspace(-0.5, 0.5, num=12).reshape(3, 4) out, _ = relu_forward(x) correct_out = np.array([[ 0., 0., 0., 0., ], [ 0., 0., 0.04545455, 0.13636364,], [ 0.22727273, 0.31818182, 0...
Testing relu_forward function: difference: 4.999999798022158e-08
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
ReLU activation: backwardNow implement the backward pass for the ReLU activation function in the `relu_backward` function and test your implementation using numeric gradient checking:
np.random.seed(231) x = np.random.randn(10, 10) dout = np.random.randn(*x.shape) dx_num = eval_numerical_gradient_array(lambda x: relu_forward(x)[0], x, dout) _, cache = relu_forward(x) dx = relu_backward(dout, cache) # The error should be on the order of e-12 print('Testing relu_backward function:') print('dx error...
Testing relu_backward function: dx error: 3.2756349136310288e-12
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Inline Question 1: We've only asked you to implement ReLU, but there are a number of different activation functions that one could use in neural networks, each with its pros and cons. In particular, an issue commonly seen with activation functions is getting zero (or close to zero) gradient flow during backpropagation...
from cs231n.layer_utils import affine_relu_forward, affine_relu_backward np.random.seed(231) x = np.random.randn(2, 3, 4) w = np.random.randn(12, 10) b = np.random.randn(10) dout = np.random.randn(2, 10) out, cache = affine_relu_forward(x, w, b) dx, dw, db = affine_relu_backward(dout, cache) dx_num = eval_numerical_g...
Testing affine_relu_forward and affine_relu_backward: dx error: 2.299579177309368e-11 dw error: 8.162011105764925e-11 db error: 7.826724021458994e-12
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Loss layers: Softmax and SVMNow implement the loss and gradient for softmax and SVM in the `softmax_loss` and `svm_loss` function in `cs231n/layers.py`. These should be similar to what you implemented in `cs231n/classifiers/softmax.py` and `cs231n/classifiers/linear_svm.py`.You can make sure that the implementations a...
np.random.seed(231) num_classes, num_inputs = 10, 50 x = 0.001 * np.random.randn(num_inputs, num_classes) y = np.random.randint(num_classes, size=num_inputs) dx_num = eval_numerical_gradient(lambda x: svm_loss(x, y)[0], x, verbose=False) loss, dx = svm_loss(x, y) # Test svm_loss function. Loss should be around 9 and ...
Testing svm_loss: loss: 8.999602749096233 dx error: 1.4021566006651672e-09 Testing softmax_loss: loss: 2.3025458445007376 dx error: 8.234144091578429e-09
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Two-layer networkOpen the file `cs231n/classifiers/fc_net.py` and complete the implementation of the `TwoLayerNet` class. Read through it to make sure you understand the API. You can run the cell below to test your implementation.
np.random.seed(231) N, D, H, C = 3, 5, 50, 7 X = np.random.randn(N, D) y = np.random.randint(C, size=N) std = 1e-3 model = TwoLayerNet(input_dim=D, hidden_dim=H, num_classes=C, weight_scale=std) print('Testing initialization ... ') W1_std = abs(model.params['W1'].std() - std) b1 = model.params['b1'] W2_std = abs(mode...
Testing initialization ... Testing test-time forward pass ... Testing training loss (no regularization) Running numeric gradient check with reg = 0.0 W1 relative error: 1.83e-08 W2 relative error: 3.20e-10 b1 relative error: 9.83e-09 b2 relative error: 4.33e-10 Running numeric gradient check with reg = 0.7 W1 relat...
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
SolverOpen the file `cs231n/solver.py` and read through it to familiarize yourself with the API. You also need to imeplement the `sgd` function in `cs231n/optim.py`. After doing so, use a `Solver` instance to train a `TwoLayerNet` that achieves about `36%` accuracy on the validation set.
input_size = 32 * 32 * 3 hidden_size = 50 num_classes = 10 model = TwoLayerNet(input_size, hidden_size, num_classes) solver = None ############################################################################## # TODO: Use a Solver instance to train a TwoLayerNet that achieves about 36% # # accuracy on the validation s...
(Iteration 1 / 4900) loss: 2.300089 (Epoch 0 / 10) train acc: 0.171000; val_acc: 0.170000 (Iteration 101 / 4900) loss: 1.782419 (Iteration 201 / 4900) loss: 1.803466 (Iteration 301 / 4900) loss: 1.712676 (Iteration 401 / 4900) loss: 1.693946 (Epoch 1 / 10) train acc: 0.399000; val_acc: 0.428000 (Iteration 501 / 4900) l...
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Debug the trainingWith the default parameters we provided above, you should get a validation accuracy of about 0.36 on the validation set. This isn't very good.One strategy for getting insight into what's wrong is to plot the loss function and the accuracies on the training and validation sets during optimization.Anot...
# Run this cell to visualize training loss and train / val accuracy plt.subplot(2, 1, 1) plt.title('Training loss') plt.plot(solver.loss_history, 'o') plt.xlabel('Iteration') plt.subplot(2, 1, 2) plt.title('Accuracy') plt.plot(solver.train_acc_history, '-o', label='train') plt.plot(solver.val_acc_history, '-o', label...
_____no_output_____
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Tune your hyperparameters**What's wrong?**. Looking at the visualizations above, we see that the loss is decreasing more or less linearly, which seems to suggest that the learning rate may be too low. Moreover, there is no gap between the training and validation accuracy, suggesting that the model we used has low capa...
best_model = None best_acc = 0 ################################################################################# # TODO: Tune hyperparameters using the validation set. Store your best trained # # model in best_model. # # ...
(Iteration 1 / 4900) loss: 2.306069 (Epoch 0 / 10) train acc: 0.104000; val_acc: 0.114000 (Iteration 101 / 4900) loss: 1.724592 (Iteration 201 / 4900) loss: 1.782996 (Iteration 301 / 4900) loss: 1.663951 (Iteration 401 / 4900) loss: 1.639033 (Epoch 1 / 10) train acc: 0.443000; val_acc: 0.435000 (Iteration 501 / 4900) l...
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Test your model!Run your best model on the validation and test sets. You should achieve above 48% accuracy on the validation set and the test set.
y_val_pred = np.argmax(best_model.loss(data['X_val']), axis=1) print('Validation set accuracy: ', (y_val_pred == data['y_val']).mean()) data = get_CIFAR10_data() y_test_pred = np.argmax(best_model.loss(data["X_test"]), axis=1) print('Test set accuracy: ', (y_test_pred == data['y_test']).mean())
Test set accuracy: 0.495
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Inline Question 2: Now that you have trained a Neural Network classifier, you may find that your testing accuracy is much lower than the training accuracy. In what ways can we decrease this gap? Select all that apply.1. Train on a larger dataset.2. Add more hidden units.3. Increase the regularization strength.4. None ...
_____no_output_____
Apache-2.0
assignment1/two_layer_net.ipynb
qiaw99/CS231n-Convolutional-Neural-Networks-for-Visual-Recognition
Welcome to 101 Exercises for Python FundamentalsSolving these exercises will help make you a better programmer. Solve them in order, because each solution builds scaffolding, working code, and knowledge you can use on future problems. Read the directions carefully, and have fun!> "Learning to program takes a little b...
# Example problem: # Uncomment the line below and run this cell. # The hashtag "#" character in a line of Python code is the comment character. doing_python_right_now = True # The lines below will test your answer. If you see an error, then it means that your answer is incorrect or incomplete. assert doing_python_rig...
Exercise 4 is correct.
MIT
101_exercises.ipynb
barbmarques/python-exercises
List Operations**Hint** Recommend finding and using built-in Python functionality whenever possible.
# Exercise 5 # Given the following assigment of the list of fruits, add "tomato" to the end of the list. fruits = ["mango", "banana", "guava", "kiwi", "strawberry"] fruits.append('tomato') assert fruits == ["mango", "banana", "guava", "kiwi", "strawberry", "tomato"], "Ensure the variable contains all the strings in ...
Exercise 10 is correct
MIT
101_exercises.ipynb
barbmarques/python-exercises