code
stringlengths
101
5.91M
class MPClassFuncOnDemand(): def __init__(self, class_handle, class_func_name, **class_kwargs): self.class_handle = class_handle self.class_func_name = class_func_name self.class_kwargs = class_kwargs self.class_func = None self.s2c = multiprocessing.Queue() self.c2s = multiprocessing.Queue() self.lock = multiprocessing.Lock() io.add_process_messages_callback(self.io_callback) def io_callback(self): while (not self.c2s.empty()): (func_args, func_kwargs) = self.c2s.get() if (self.class_func is None): self.class_func = getattr(self.class_handle(**self.class_kwargs), self.class_func_name) self.s2c.put(self.class_func(*func_args, **func_kwargs)) def __call__(self, *args, **kwargs): with self.lock: self.c2s.put((args, kwargs)) return self.s2c.get() def __getstate__(self): return {'s2c': self.s2c, 'c2s': self.c2s, 'lock': self.lock}
def nondefault_trainer_args(opt): parser = argparse.ArgumentParser() parser = Trainer.add_argparse_args(parser) args = parser.parse_args([]) return sorted((k for k in vars(args) if (getattr(opt, k) != getattr(args, k))))
class FixedBernoulli(torch.distributions.Bernoulli): def log_probs(self, actions): return super.log_prob(actions).view(actions.size(0), (- 1)).sum((- 1)).unsqueeze((- 1)) def entropy(self): return super().entropy().sum((- 1)) def mode(self): return torch.gt(self.probs, 0.5).float()
class ModelParallelTransformerDecoder(TransformerDecoder): def build_decoder_layer(self, args, no_encoder_attn=False): return ModelParallelTransformerDecoderLayer(args, no_encoder_attn) def output_layer(self, features, **kwargs): if (not self.share_input_output_embed): raise NotImplementedError('Model parallel training currently requires --share-decoder-input-output-embed') features = copy_to_model_parallel_region(features) x = self.output_projection(features) if (getattr(self.args, 'criterion') != 'vocab_parallel_cross_entropy'): x = gather_from_model_parallel_region(x).contiguous() return x
def closure_sgd(): global i, out_avg, sgd_out, sgd_expm_out, check_point net_input = net_input_saved out = net(net_input) if (out_avg is None): out_avg = out.detach() else: out_avg = ((out_avg * exp_weight) + (out.detach() * (1 - exp_weight))) total_loss = mse((out * mask_var), (img_var * mask_var)) total_loss.backward() out_np = out.detach().cpu().numpy()[0] out_avg_np = out_avg.detach().cpu().numpy()[0] psrn_gt = compare_psnr(img_np, out_np) psrn_gt_sm = compare_psnr(img_np, out_avg_np) sgd_psnr_list.append(psrn_gt) sgd_expm_psnr_list.append(psrn_gt_sm) if (i == (check_point - 1)): sgd_out = out_np sgd_expm_out = out_avg_np if ((i % show_every) == 0): np_plot(out.detach().cpu().numpy()[0], ('Iter: %d; gt %.2f' % (i, psrn_gt))) i += 1 return total_loss
def check_k0(freqs, k0=None, rtol=0.01, atol=1e-07): k0 = (k0 if (k0 is not None) else get_k0(freqs)) df = (freqs[1] - freqs[0]) f0 = (k0 * df) assert (abs((f0 - freqs[0])) < ((rtol * df) + atol))
def ortho_weight(ndim): W = numpy.random.randn(ndim, ndim) (u, s, v) = numpy.linalg.svd(W) return u.astype('float32')
class FuseMatMulRequantizeDequantizeNewAPITransformer(GraphRewriterBase): def __init__(self, model, device='cpu'): super().__init__(model) self.device = device self.graph_analyzer = GraphAnalyzer() self.graph_analyzer.graph = self.model self.graph_info = self.graph_analyzer.parse_graph() self.eps = 1e-05 def do_transformation(self): fuse_pattern = [['_QuantizedMatMul'], ['Requantize', 'RequantizePerChannel'], ['Dequantize'], ('Softmax',)] uint8_type = dtypes.quint8.as_datatype_enum int8_type = dtypes.qint8.as_datatype_enum float32_type = dtypes.float32.as_datatype_enum qint32_type = dtypes.qint32.as_datatype_enum target_nodes = self.graph_analyzer.query_fusion_pattern_nodes(fuse_pattern) for i in target_nodes: quantized_node_name = i[0] quantized_node = self.graph_info[quantized_node_name].node requantize_node_name = i[1] requantize_node = self.graph_info[requantize_node_name].node requested_output_min_name = requantize_node.input[3] requested_output_max_name = requantize_node.input[4] deq_node_name = i[2] quantized_node_op = i[(- 1)][0] attr_fused_ops = ''.join((x for x in quantized_node.attr['fused_ops'].SerializeToString().decode('UTF-8', 'ignore').strip() if x.isprintable())) if ('BiasAddAdd' not in attr_fused_ops): continue new_node = node_def_pb2.NodeDef() new_node.op = quantized_node_op new_node.name = requantize_node_name for (_, value) in enumerate(quantized_node.input): new_node.input.append(value) new_node.input.append(requested_output_min_name) new_node.input.append(requested_output_max_name) if ('T1' in quantized_node.attr): new_node.attr['T1'].CopyFrom(quantized_node.attr['T1']) if ('T2' in quantized_node.attr): new_node.attr['T2'].CopyFrom(quantized_node.attr['T2']) if ('U' in quantized_node.attr): new_node.attr['U'].CopyFrom(quantized_node.attr['U']) if ('transpose_b' in quantized_node.attr): new_node.attr['transpose_b'].CopyFrom(quantized_node.attr['transpose_b']) if ('transpose_a' in quantized_node.attr): new_node.attr['transpose_a'].CopyFrom(quantized_node.attr['transpose_a']) if ('input_quant_mode' in quantized_node.attr): new_node.attr['input_quant_mode'].CopyFrom(quantized_node.attr['input_quant_mode']) if ('output_quant_mode' in quantized_node.attr): new_node.attr['output_quant_mode'].CopyFrom(quantized_node.attr['output_quant_mode']) top_node_name = Helper.node_name_from_input(quantized_node.input[0]) max_filter_node = None min_filter_node = None if (':2' not in new_node.input[7]): max_filter_node = self.graph_info[new_node.input[7]].node if (':1' not in new_node.input[6]): min_filter_node = self.graph_info[new_node.input[6]].node last_node = self.graph_info[new_node.input[0]].node weight_node = self.graph_info[Helper.node_name_from_input(new_node.input[1])].node bias_node = self.graph_info[Helper.node_name_from_input(new_node.input[2])].node if (not (last_node.op == 'QuantizedConcatV2')): max_input_node = self.graph_info[last_node.input[(- 1)]].node min_input_node = self.graph_info[last_node.input[(- 2)]].node type_bias = float32_type if ((not (last_node.op == 'QuantizedConcatV2')) and (max_input_node.op == 'Enter')): min_input_parent_name = Helper.node_name_from_input(min_input_node.input[0]) max_input_parent_name = Helper.node_name_from_input(max_input_node.input[0]) min_input_parent_node = self.graph_info[min_input_parent_name].node max_input_parent_node = self.graph_info[max_input_parent_name].node if ((min_input_parent_node.op != 'Const') or (max_input_parent_node.op != 'Const')): continue min_input_node = min_input_parent_node max_input_node = max_input_parent_node if (max_filter_node and min_filter_node and (max_filter_node.op == 'Enter')): min_filter_parent_name = Helper.node_name_from_input(min_filter_node.input[0]) max_filter_parent_name = Helper.node_name_from_input(max_filter_node.input[0]) min_filter_parent_node = self.graph_info[min_filter_parent_name].node max_filter_parent_node = self.graph_info[max_filter_parent_name].node if ((min_filter_parent_node.op != 'Const') or (max_filter_parent_node.op != 'Const')): continue min_filter_node = min_filter_parent_node max_filter_node = max_filter_parent_node if (weight_node.op == 'Enter'): weight_parent_name = Helper.node_name_from_input(weight_node.input[0]) weight_parent_node = self.graph_info[weight_parent_name].node if (weight_parent_node.op != 'Const'): continue weight_node = weight_parent_node bias_enter_node = None if (bias_node.op == 'Enter'): bias_enter_node = bias_node bias_parent_name = Helper.node_name_from_input(bias_node.input[0]) bias_parent_node = self.graph_info[bias_parent_name].node if (bias_parent_node.op != 'Const'): continue bias_node = bias_parent_node if (max_filter_node and min_filter_node and (max_filter_node.op == 'Const') and (weight_node.op == 'Const') and (not (last_node.op == 'QuantizedConcatV2'))): min_input_value = min_input_node.attr['value'].tensor.float_val[0] max_input_value = max_input_node.attr['value'].tensor.float_val[0] if (requantize_node.op.find('PerChannel') != (- 1)): max_filter_tensor = tensor_util.MakeNdarray(max_filter_node.attr['value'].tensor) min_filter_tensor = tensor_util.MakeNdarray(min_filter_node.attr['value'].tensor) else: max_filter_value = max_filter_node.attr['value'].tensor.float_val[0] min_filter_value = min_filter_node.attr['value'].tensor.float_val[0] weights_tensor = tensor_util.MakeNdarray(weight_node.attr['value'].tensor) bias_tensor = tensor_util.MakeNdarray(bias_node.attr['value'].tensor) is_min_first = bool((quantized_node.attr['input_quant_mode'].s == b'MIN_FIRST')) input_range = ((max_input_value - min_input_value) if is_min_first else max(abs(max_input_value), abs(min_input_value))) if ((- self.eps) <= input_range <= self.eps): input_range += self.eps if ((- self.eps) <= (max_input_value - min_input_value) <= self.eps): max_input_value += self.eps if (requantize_node.op.find('PerChannel') != (- 1)): int32_bias = Helper.generate_int32_bias_for_matmul_per_channel(bias_tensor, weights_tensor, max_input_value, min_input_value, max_filter_tensor, min_filter_tensor) else: int32_bias = Helper.generate_int32_bias_for_matmul(bias_tensor, weights_tensor, input_range, max_input_value, min_input_value, max_filter_value, min_filter_value) bias_node.attr['dtype'].CopyFrom(attr_value_pb2.AttrValue(type=(float32_type if (self.device == 'gpu') else qint32_type))) bias_node.attr['value'].CopyFrom(attr_value_pb2.AttrValue(tensor=tensor_util.make_tensor_proto((bias_tensor if (self.device == 'gpu') else int32_bias), (dtypes.float32 if (self.device == 'gpu') else dtypes.int32), bias_tensor.shape))) bias_node.attr['value'].tensor.dtype = (float32_type if (self.device == 'gpu') else qint32_type) type_bias = (float32_type if (self.device == 'gpu') else qint32_type) if bias_enter_node: bias_enter_node.attr['T'].CopyFrom(attr_value_pb2.AttrValue(type=(float32_type if (self.device == 'gpu') else qint32_type))) else: type_bias = float32_type new_node.attr['Tbias'].CopyFrom(attr_value_pb2.AttrValue(type=type_bias)) Helper.set_attr_string_list(new_node, 'fused_ops', [b'BiasAdd', b'Add', b'Dequantize']) Helper.set_attr_type_list(new_node, 'Thost_inputs', [uint8_type, int8_type, type_bias, float32_type, float32_type, float32_type, float32_type, float32_type, float32_type, float32_type]) Helper.set_attr_type_list(new_node, 'Thost_outputs', [float32_type]) new_node.attr['Tout'].CopyFrom(attr_value_pb2.AttrValue(type=float32_type)) self.graph_analyzer.remove_node(requantize_node_name) if self.graph_info[deq_node_name].outputs: self.graph_analyzer.replace_single_node(new_node, [top_node_name], quantized_node_name, self.graph_info[deq_node_name].outputs, deq_node_name) self.graph_analyzer.remove_node(deq_node_name) else: self.graph_analyzer.remove_node(deq_node_name) new_node.name = deq_node_name self.graph_analyzer.replace_single_node(new_node, [top_node_name], quantized_node_name, [], deq_node_name) self.graph_analyzer.remove_node(quantized_node_name) return self.graph_analyzer.dump_graph()
class DiagonalGaussianDensity(Density): def __init__(self, mean, stddev, num_fixed_samples=0): super().__init__() assert (mean.shape == stddev.shape) self.register_buffer('mean', mean) self.register_buffer('stddev', stddev) if (num_fixed_samples > 0): self.register_buffer('_fixed_samples', self.sample(num_fixed_samples)) def shape(self): return self.mean.shape def p_parameters(self): return [] def q_parameters(self): return [] def _fix_random_u(self): return (self, self.sample(num_samples=1)[0]) def fix_u(self, u): assert (not u) return self def _elbo(self, z, detach_q_params, detach_q_samples): log_prob = diagonal_gaussian_log_prob(z, self.mean.expand_as(z), self.stddev.expand_as(z)) return {'log-p': log_prob, 'log-q': z.new_zeros((z.shape[0], 1)), 'z': z} def _sample(self, num_samples): return diagonal_gaussian_sample(self.mean.expand(num_samples, *self.shape), self.stddev.expand(num_samples, *self.shape)) def _fixed_sample(self, noise): return (noise if (noise is not None) else self._fixed_samples)
def load_data(tr: Training, verbose=0): list_train_ds = tf.data.Dataset.list_files([str((((tr.train_data_dir + '/') + ds) + '/*/images/*')) for ds in tr.train_datasets]) list_test_ds = tf.data.Dataset.list_files([str((((tr.test_data_dir + '/') + ds) + '/*/images/*')) for ds in tr.test_datasets]) train_data = dataloader.dataloader(tr.train_data_dir, tr.train_datasets, tr.hyperparameters.POLICY) test_data = dataloader.dataloader(tr.test_data_dir, tr.test_datasets, tr.hyperparameters.POLICY) if verbose: for f in list_train_ds.take(5): print(f.numpy()) print() for f in list_test_ds.take(5): print(f.numpy()) print() print(('Number of train samples: %d' % len(train_data.labels))) print(('Number of test samples: %d' % len(test_data.labels))) def process_train_path(file_path): (cmd_input, label) = train_data.get_label(tf.strings.regex_replace(file_path, '[/\\\\]', '/')) img = utils.load_img(file_path, tr.hyperparameters.IS_CROP) img = data_augmentation.augment_img(img) if tr.hyperparameters.FLIP_AUG: (img, cmd_input, label) = data_augmentation.flip_sample(img, cmd_input, label) if tr.hyperparameters.CMD_AUG: cmd_input = data_augmentation.augment_cmd(cmd_input) return ((img, cmd_input), label) def process_test_path(file_path): (cmd_input, label) = test_data.get_label(tf.strings.regex_replace(file_path, '[/\\\\]', '/')) img = utils.load_img(file_path, tr.hyperparameters.IS_CROP) return ((img, cmd_input), label) labeled_ds = list_train_ds.map(process_train_path, num_parallel_calls=4) for ((image, cmd_input), label) in labeled_ds.take(1): shape = image.numpy().shape tr.NETWORK_IMG_HEIGHT = shape[0] tr.NETWORK_IMG_WIDTH = shape[1] print('Image shape: ', shape) print('Command: ', cmd_input.numpy()) print('Label: ', label.numpy()) tr.train_ds = utils.prepare_for_training(ds=labeled_ds, batch_sz=tr.hyperparameters.TRAIN_BATCH_SIZE, shuffle_buffer_sz=(100 * tr.hyperparameters.TRAIN_BATCH_SIZE), prefetch_buffer_sz=(10 * tr.hyperparameters.TRAIN_BATCH_SIZE)) test_ds = list_test_ds.map(process_test_path, num_parallel_calls=4) test_ds = test_ds.batch(tr.hyperparameters.TEST_BATCH_SIZE) tr.test_ds = test_ds.prefetch(buffer_size=(10 * tr.hyperparameters.TRAIN_BATCH_SIZE))
_model def vit_base_r26_s32_224(pretrained=False, **kwargs): backbone = _resnetv2((2, 2, 2, 2), **kwargs) model_kwargs = dict(embed_dim=768, depth=12, num_heads=12, **kwargs) model = _create_vision_transformer_hybrid('vit_base_r26_s32_224', backbone=backbone, pretrained=pretrained, **model_kwargs) return model
def main(): args = init_args() (x_val, y_val, z_val) = blh2xyz(args.b, args.l, args.h) message = 'X: {:.5f}, Y: {:.5f}, Z: {:.5f}' print(message.format(x_val, y_val, z_val)) return 0
def test_get_metrics_returns_dict(): (ground_truth, retrieved) = return_ground_incorrect_retrievals() assert isinstance(get_all_metrics(ground_truth, retrieved), dict) assert (len(get_all_metrics(ground_truth, retrieved).values()) == 3)
_module() class Rotate(GeomTransform): def __init__(self, prob: float=1.0, level: Optional[int]=None, min_mag: float=0.0, max_mag: float=30.0, reversal_prob: float=0.5, img_border_value: Union[(int, float, tuple)]=128, mask_border_value: int=0, seg_ignore_label: int=255, interpolation: str='bilinear') -> None: assert (0.0 <= min_mag <= 180.0), f'min_mag for Rotate should be in range [0,180], got {min_mag}.' assert (0.0 <= max_mag <= 180.0), f'max_mag for Rotate should be in range [0,180], got {max_mag}.' super().__init__(prob=prob, level=level, min_mag=min_mag, max_mag=max_mag, reversal_prob=reversal_prob, img_border_value=img_border_value, mask_border_value=mask_border_value, seg_ignore_label=seg_ignore_label, interpolation=interpolation) def _get_homography_matrix(self, results: dict, mag: float) -> np.ndarray: img_shape = results['img_shape'] center = (((img_shape[1] - 1) * 0.5), ((img_shape[0] - 1) * 0.5)) cv2_rotation_matrix = cv2.getRotationMatrix2D(center, (- mag), 1.0) return np.concatenate([cv2_rotation_matrix, np.array([0, 0, 1]).reshape((1, 3))]).astype(np.float32) def _transform_img(self, results: dict, mag: float) -> None: results['img'] = mmcv.imrotate(results['img'], mag, border_value=self.img_border_value, interpolation=self.interpolation) def _transform_masks(self, results: dict, mag: float) -> None: results['gt_masks'] = results['gt_masks'].rotate(results['img_shape'], mag, border_value=self.mask_border_value, interpolation=self.interpolation) def _transform_seg(self, results: dict, mag: float) -> None: results['gt_seg_map'] = mmcv.imrotate(results['gt_seg_map'], mag, border_value=self.seg_ignore_label, interpolation='nearest')
class DDPMMeanLoss(nn.Module): def __init__(self, *, reduce_mean: bool=True, likelihood_weighting: bool=False, eps_weighting: bool=False): super().__init__() self.reduce_mean = reduce_mean self.likelihood_weighting = likelihood_weighting self.eps_weighting = eps_weighting def forward(self, x, model, sde: VPSDE, t, **model_kwargs): assert isinstance(sde, VPSDE), 'DDPM training only works for VPSDEs.' noise = torch.randn_like(x) perturbed_data = ((x * sde.sqrt_alphas_cumprod.to(x.device)[(t, None, None, None)]) + (sde.sqrt_1m_alphas_cumprod.to(x.device)[(t, None, None, None)] * noise)) mu = ((perturbed_data - ((sde.discrete_betas.to(x.device)[(t, None, None, None)] * noise) / sde.sqrt_1m_alphas_cumprod.to(x.device)[(t, None, None, None)])) / torch.sqrt(sde.alphas.to(x.device)[(t, None, None, None)])) mu_pred = model(perturbed_data, t, **model_kwargs) if self.likelihood_weighting: weighting = (1.0 / (2.0 * sde.discrete_betas.to(x.device)[(t, None, None, None)])) elif self.eps_weighting: weighting = ((sde.alphas.to(x.device)[(t, None, None, None)] * (1.0 - sde.alphas_cumprod.to(x.device)[(t, None, None, None)])) / (sde.discrete_betas.to(x.device)[(t, None, None, None)] ** 2.0)) else: weighting = 1.0 losses = (torch.square((mu_pred - mu)) * weighting) if self.reduce_mean: loss = torch.mean(losses) else: loss = torch.mean(torch.sum(losses.reshape(losses.shape[0], (- 1)), dim=(- 1))) return loss
def to_str(segment): assert (len(segment) == 3) return '[{0:.3f}, {1:.3f}, {2}]'.format(segment[0], segment[1], segment[2])
def train(args, net, loss_function, data_iterator): ctx = args.ctx[0] local_cfg = cfg.STATIC_GRAPH net.initialize(init=mx.init.Xavier(magnitude=3), ctx=ctx) trainer = gluon.Trainer(net.collect_params(), local_cfg.MODEL.TRAIN.OPTIMIZER, {'learning_rate': local_cfg.MODEL.TRAIN.LR, 'wd': local_cfg.MODEL.TRAIN.WD}) train_loss_logger = MetricLogger(['iter', 'loss'], ['%d', '%.4f'], os.path.join(args.save_dir, ('train_loss%d.csv' % args.save_id))) valid_loss_logger = MetricLogger(['iter', 'loss', 'f1', 'acc', 'is_best'], ['%d', '%.4f', '%.4f', '%.4f', '%d'], os.path.join(args.save_dir, ('valid_loss%d.csv' % args.save_id))) test_loss_logger = MetricLogger(['iter', 'loss', 'f1', 'acc'], ['%d', '%.4f', '%.4f', '%.4f'], os.path.join(args.save_dir, ('test_loss%d.csv' % args.save_id))) best_valid_f1 = 0 best_valid_iter_info = [] best_test_iter_info = [] no_better_valid = 0 iter_id = 1 epoch = 1 train_moving_avg_loss = 0.0 data_iterator.begin_epoch('train') for iter in range(1, local_cfg.MODEL.TRAIN.MAX_ITER): if data_iterator.epoch_finished: print(('Epoch %d finished! It has %d iterations.' % (epoch, iter_id))) data_iterator.begin_epoch('train') iter_id = 1 epoch += 1 else: iter_id += 1 (layer0_features_nd, end_points_l, indptr_l, indices_in_merged_l, labels_nd, node_ids_l) = data_iterator.sample() with mx.autograd.record(): if net._output_inner_result: (logits, gate_l, sharpness_l, attend_weights_wo_gate_l) = net(layer0_features_nd, end_points_l, indptr_l, indices_in_merged_l) else: logits = net(layer0_features_nd, end_points_l, indptr_l, indices_in_merged_l) loss = loss_function(logits, labels_nd) loss = nd.mean(loss) loss.backward() if (iter == 1): logging.info(('Total Param Number: %d' % gluon_total_param_num(net))) gluon_log_net_info(net, save_path=os.path.join(args.save_dir, ('net_info%d.txt' % args.save_id))) if (local_cfg.MODEL.TRAIN.GRAD_CLIP <= 0): gnorm = get_global_norm([v.grad() for v in net.collect_params().values()]) else: gnorm = gluon.utils.clip_global_norm([v.grad() for v in net.collect_params().values()], max_norm=local_cfg.MODEL.TRAIN.GRAD_CLIP) trainer.step(batch_size=1) iter_train_loss = loss.asscalar() train_moving_avg_loss += iter_train_loss logging.info(('[iter=%d]: loss=%.4f, gnorm=%g' % (iter, iter_train_loss, gnorm))) train_loss_logger.log(iter=iter, loss=iter_train_loss) if ((iter % local_cfg.MODEL.TRAIN.VALID_ITER) == 0): (valid_loss, valid_f1, valid_accuracy) = eval_classification(net=net, loss_function=loss_function, data_iterator=data_iterator, num_class=data_iterator.num_class, mode='valid') logging.info(('Iter %d, Epoch %d,: train_moving_loss=%.4f, valid loss=%.4f, f1=%.4f, accuracy=%.4f' % (iter, epoch, (train_moving_avg_loss / local_cfg.MODEL.TRAIN.VALID_ITER), valid_loss, valid_f1, valid_accuracy))) train_moving_avg_loss = 0.0 if (valid_f1 > best_valid_f1): logging.info('> Best Iter') is_best = True best_valid_f1 = valid_f1 best_iter = iter best_valid_iter_info = [best_iter, valid_loss, valid_f1, valid_accuracy] no_better_valid = 0 net.save_params(filename=os.path.join(args.save_dir, ('best_valid%d.params' % args.save_id))) (test_loss, test_f1, test_accuracy) = eval_classification(net=net, loss_function=loss_function, data_iterator=data_iterator, num_class=data_iterator.num_class, mode='test') test_loss_logger.log(iter=iter, loss=test_loss, f1=test_f1, acc=test_accuracy) best_test_iter_info = [best_iter, test_loss, test_f1, test_accuracy] logging.info(('Iter %d, Epoch %d: test loss=%.4f, f1=%.4f, accuracy=%.4f' % (iter, epoch, test_loss, test_f1, test_accuracy))) else: is_best = False no_better_valid += 1 if (no_better_valid > local_cfg.MODEL.TRAIN.EARLY_STOPPING_PATIENCE): logging.info('Early stopping threshold reached. Stop training.') valid_loss_logger.log(iter=iter, loss=valid_loss, f1=valid_f1, acc=valid_accuracy, is_best=is_best) break elif (no_better_valid > local_cfg.MODEL.TRAIN.DECAY_PATIENCE): new_lr = max((trainer.learning_rate * local_cfg.MODEL.TRAIN.LR_DECAY_FACTOR), local_cfg.MODEL.TRAIN.MIN_LR) if (new_lr < trainer.learning_rate): logging.info(('Change the LR to %g' % new_lr)) trainer.set_learning_rate(new_lr) no_better_valid = 0 valid_loss_logger.log(iter=iter, loss=valid_loss, f1=valid_f1, acc=valid_accuracy, is_best=is_best) logging.info(('Best Valid: [Iter, Loss, F1, ACC] = %s' % str(best_valid_iter_info))) logging.info(('Best Test : [Iter, Loss, F1, ACC] = %s' % str(best_test_iter_info))) valid_loss_logger.log(iter=best_valid_iter_info[0], loss=best_valid_iter_info[1], f1=best_valid_iter_info[2], acc=best_valid_iter_info[3], is_best=True) test_loss_logger.log(iter=best_test_iter_info[0], loss=best_test_iter_info[1], f1=best_test_iter_info[2], acc=best_test_iter_info[3]) if ((args.emails is not None) and (len(args.emails) > 0)): for email_address in args.emails.split(','): send_msg(title=os.path.basename(args.save_dir), text=((((('Test: [Iter, Loss, F1, ACC] = %s\n' % str(best_test_iter_info)) + ('Valid: [Iter, Loss, F1, ACC] = %s\n' % str(best_valid_iter_info))) + ('Save Dir: %s\n' % args.save_dir)) + '\nConfig:\n') + ordered_dump()), dst_address=email_address) return
class TextualResEncoder(nn.Module): def __init__(self, input_nc=3, ngf=32, z_nc=256, img_f=256, L=6, layers=5, norm='none', activation='ReLU', use_spect=True, use_coord=False, image_dim=256, text_dim=256, multi_peak=True, pool_attention='max'): super(TextualResEncoder, self).__init__() self.layers = layers self.z_nc = z_nc self.L = L norm_layer = get_norm_layer(norm_type=norm) nonlinearity = get_nonlinearity_layer(activation_type=activation) self.block0 = ResBlockEncoderOptimized(input_nc, ngf, norm_layer, nonlinearity, use_spect, use_coord) self.word_attention = ImageTextAttention(idf=image_dim, cdf=text_dim, multi_peak=multi_peak, pooling=pool_attention) mult = 1 for i in range((layers - 1)): mult_prev = mult mult = min((2 ** (i + 2)), (img_f // ngf)) block = ResBlock((ngf * mult_prev), (ngf * mult), (ngf * mult_prev), norm_layer, nonlinearity, 'down', use_spect, use_coord) setattr(self, ('encoder' + str(i)), block) for i in range(self.L): block = ResBlock((ngf * mult), (ngf * mult), (ngf * mult), norm_layer, nonlinearity, 'none', use_spect, use_coord) setattr(self, ('infer_prior' + str(i)), block) self.posterior = ResBlock((2 * text_dim), (2 * z_nc), ((ngf * mult) * 2), norm_layer, nonlinearity, 'none', use_spect, use_coord) self.prior = ResBlock((2 * text_dim), (2 * z_nc), ((ngf * mult) * 2), norm_layer, nonlinearity, 'none', use_spect, use_coord) def forward(self, img_m, sentence_embedding, word_embeddings, text_mask, image_mask, img_c=None): if (type(img_c) != type(None)): img = torch.cat([img_m, img_c], dim=0) else: img = img_m out = self.block0(img) feature = [out] for i in range((self.layers - 1)): model = getattr(self, ('encoder' + str(i))) out = model(out) feature.append(out) image_mask = task.scale_img(image_mask, size=[feature[(- 1)].size(2), feature[(- 1)].size(3)]) if (image_mask.size(1) == 3): image_mask = image_mask.chunk(3, dim=1)[0] if (type(img_c) != type(None)): (f_m_g, f_m_rec) = feature[(- 1)].chunk(2) img_mask_g = image_mask img_mask_rec = (1 - img_mask_g) weighted_word_embedding_rec = self.word_attention(f_m_rec, word_embeddings, mask=text_mask, image_mask=img_mask_rec, inverse_attention=False) weighted_word_embedding_g = self.word_attention(f_m_g, word_embeddings, mask=text_mask, image_mask=img_mask_g, inverse_attention=True) weighted_word_embedding = torch.cat([weighted_word_embedding_g, weighted_word_embedding_rec]) (distribution, f_text) = self.two_paths(out, sentence_embedding, weighted_word_embedding) return (distribution, feature, f_text) else: f_m = feature[(- 1)] weighted_word_embedding = self.word_attention(f_m, word_embeddings, mask=text_mask, image_mask=image_mask, inverse_attention=True) (distribution, f_text) = self.one_path(out, sentence_embedding, weighted_word_embedding) return (distribution, feature, f_text) def one_path(self, f_in, sentence_embedding, weighted_word_embedding): f_m = f_in distribution = [] for i in range(self.L): infer_prior = getattr(self, ('infer_prior' + str(i))) f_m = infer_prior(f_m) (ix, iw) = (f_m.size(2), f_m.size(3)) sentence_dim = sentence_embedding.size(1) sentence_embedding_replication = sentence_embedding.view((- 1), sentence_dim, 1, 1).repeat(1, 1, ix, iw) f_text = torch.cat([sentence_embedding_replication, weighted_word_embedding], dim=1) o = self.prior(f_text) (q_mu, q_std) = torch.split(o, self.z_nc, dim=1) distribution.append([q_mu, F.softplus(q_std)]) return (distribution, f_text) def two_paths(self, f_in, sentence_embedding, weighted_word_embedding): (f_m, f_c) = f_in.chunk(2) (weighted_word_embedding_m, weighted_word_embedding_c) = weighted_word_embedding.chunk(2) distributions = [] (ix, iw) = (f_c.size(2), f_c.size(3)) sentence_dim = sentence_embedding.size(1) sentence_embedding_replication = sentence_embedding.view((- 1), sentence_dim, 1, 1).repeat(1, 1, ix, iw) f_text_c = torch.cat([sentence_embedding_replication, weighted_word_embedding_c], dim=1) o = self.posterior(f_text_c) (p_mu, p_std) = torch.split(o, self.z_nc, dim=1) (distribution, f_text_m) = self.one_path(f_m, sentence_embedding, weighted_word_embedding_m) distributions.append([p_mu, F.softplus(p_std), distribution[0][0], distribution[0][1]]) return (distributions, torch.cat([f_text_m, f_text_c], dim=0))
_model def fbnetv3_d(pretrained=False, **kwargs): model = _gen_fbnetv3('fbnetv3_d', pretrained=pretrained, **kwargs) return model
def data_provider(dataset_name, train_data_paths, valid_data_paths, batch_size, img_width, is_training=True): if (dataset_name not in datasets_map): raise ValueError(('Name of dataset unknown %s' % dataset_name)) train_data_list = train_data_paths.split(',') valid_data_list = valid_data_paths.split(',') if (dataset_name == 'mnist'): test_input_param = {'paths': valid_data_list, 'minibatch_size': batch_size, 'input_data_type': 'float32', 'is_output_sequence': True, 'name': (dataset_name + 'test iterator')} test_input_handle = datasets_map[dataset_name].InputHandle(test_input_param) test_input_handle.begin(do_shuffle=False) if is_training: train_input_param = {'paths': train_data_list, 'minibatch_size': batch_size, 'input_data_type': 'float32', 'is_output_sequence': True, 'name': (dataset_name + ' train iterator')} train_input_handle = datasets_map[dataset_name].InputHandle(train_input_param) train_input_handle.begin(do_shuffle=True) return (train_input_handle, test_input_handle) else: return test_input_handle
class TestMSAColumnGlobalAttention(unittest.TestCase): def test_shape(self): batch_size = consts.batch_size n_seq = consts.n_seq n_res = consts.n_res c_m = consts.c_m c = 44 no_heads = 4 msagca = MSAColumnGlobalAttention(c_m, c, no_heads) x = torch.rand((batch_size, n_seq, n_res, c_m)) shape_before = x.shape x = msagca(x, chunk_size=None) shape_after = x.shape self.assertTrue((shape_before == shape_after)) _utils.skip_unless_alphafold_installed() def test_compare(self): def run_msa_col_global_att(msa_act, msa_mask): config = compare_utils.get_alphafold_config() c_e = config.model.embeddings_and_evoformer.evoformer msa_col = alphafold.model.modules.MSAColumnGlobalAttention(c_e.msa_column_attention, config.model.global_config, name='msa_column_global_attention') act = msa_col(msa_act=msa_act, msa_mask=msa_mask) return act f = hk.transform(run_msa_col_global_att) n_res = consts.n_res n_seq = consts.n_seq c_e = consts.c_e msa_act = np.random.rand(n_seq, n_res, c_e) msa_mask = np.random.randint(low=0, high=2, size=(n_seq, n_res)) params = compare_utils.fetch_alphafold_module_weights(('alphafold/alphafold_iteration/evoformer/extra_msa_stack/' + 'msa_column_global_attention')) params = tree_map((lambda n: n[0]), params, jax.numpy.DeviceArray) out_gt = f.apply(params, None, msa_act, msa_mask).block_until_ready() out_gt = torch.as_tensor(np.array(out_gt.block_until_ready())) model = compare_utils.get_global_pretrained_openfold() out_repro = model.extra_msa_stack.blocks[0].msa_att_col(torch.as_tensor(msa_act, dtype=torch.float32).cuda(), chunk_size=4, mask=torch.as_tensor(msa_mask, dtype=torch.float32).cuda()).cpu() self.assertTrue(torch.max((torch.abs((out_gt - out_repro)) < consts.eps)))
class DDIMScheduler(SchedulerMixin, ConfigMixin): _compatibles = [e.name for e in KarrasDiffusionSchedulers] order = 1 _to_config def __init__(self, num_train_timesteps: int=1000, beta_start: float=0.0001, beta_end: float=0.02, beta_schedule: str='linear', trained_betas: Optional[Union[(np.ndarray, List[float])]]=None, clip_sample: bool=True, set_alpha_to_one: bool=True, steps_offset: int=0, prediction_type: str='epsilon', thresholding: bool=False, dynamic_thresholding_ratio: float=0.995, clip_sample_range: float=1.0, sample_max_value: float=1.0, timestep_spacing: str='leading', rescale_betas_zero_snr: bool=False): if (trained_betas is not None): self.betas = torch.tensor(trained_betas, dtype=torch.float32) elif (beta_schedule == 'linear'): self.betas = torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32) elif (beta_schedule == 'scaled_linear'): self.betas = (torch.linspace((beta_start ** 0.5), (beta_end ** 0.5), num_train_timesteps, dtype=torch.float32) ** 2) elif (beta_schedule == 'squaredcos_cap_v2'): self.betas = betas_for_alpha_bar(num_train_timesteps) else: raise NotImplementedError(f'{beta_schedule} does is not implemented for {self.__class__}') if rescale_betas_zero_snr: self.betas = rescale_zero_terminal_snr(self.betas) self.alphas = (1.0 - self.betas) self.alphas_cumprod = torch.cumprod(self.alphas, dim=0) self.final_alpha_cumprod = (torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]) self.init_noise_sigma = 1.0 self.num_inference_steps = None self.timesteps = torch.from_numpy(np.arange(0, num_train_timesteps)[::(- 1)].copy().astype(np.int64)) def scale_model_input(self, sample: torch.FloatTensor, timestep: Optional[int]=None) -> torch.FloatTensor: return sample def _get_variance(self, timestep, prev_timestep): alpha_prod_t = self.alphas_cumprod[timestep] alpha_prod_t_prev = (self.alphas_cumprod[prev_timestep] if (prev_timestep >= 0) else self.final_alpha_cumprod) beta_prod_t = (1 - alpha_prod_t) beta_prod_t_prev = (1 - alpha_prod_t_prev) variance = ((beta_prod_t_prev / beta_prod_t) * (1 - (alpha_prod_t / alpha_prod_t_prev))) return variance def _threshold_sample(self, sample: torch.FloatTensor) -> torch.FloatTensor: dtype = sample.dtype (batch_size, channels, *remaining_dims) = sample.shape if (dtype not in (torch.float32, torch.float64)): sample = sample.float() sample = sample.reshape(batch_size, (channels * np.prod(remaining_dims))) abs_sample = sample.abs() s = torch.quantile(abs_sample, self.config.dynamic_thresholding_ratio, dim=1) s = torch.clamp(s, min=1, max=self.config.sample_max_value) s = s.unsqueeze(1) sample = (torch.clamp(sample, (- s), s) / s) sample = sample.reshape(batch_size, channels, *remaining_dims) sample = sample.to(dtype) return sample def set_timesteps(self, num_inference_steps: int, device: Union[(str, torch.device)]=None): if (num_inference_steps > self.config.num_train_timesteps): raise ValueError(f'`num_inference_steps`: {num_inference_steps} cannot be larger than `self.config.train_timesteps`: {self.config.num_train_timesteps} as the unet model trained with this scheduler can only handle maximal {self.config.num_train_timesteps} timesteps.') self.num_inference_steps = num_inference_steps if (self.config.timestep_spacing == 'linspace'): timesteps = np.linspace(0, (self.config.num_train_timesteps - 1), num_inference_steps).round()[::(- 1)].copy().astype(np.int64) elif (self.config.timestep_spacing == 'leading'): step_ratio = (self.config.num_train_timesteps // self.num_inference_steps) timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::(- 1)].copy().astype(np.int64) timesteps += self.config.steps_offset elif (self.config.timestep_spacing == 'trailing'): step_ratio = (self.config.num_train_timesteps / self.num_inference_steps) timesteps = np.round(np.arange(self.config.num_train_timesteps, 0, (- step_ratio))).astype(np.int64) timesteps -= 1 else: raise ValueError(f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'leading' or 'trailing'.") self.timesteps = torch.from_numpy(timesteps).to(device) def step(self, model_output: torch.FloatTensor, timestep: int, sample: torch.FloatTensor, eta: float=0.0, use_clipped_model_output: bool=False, generator=None, variance_noise: Optional[torch.FloatTensor]=None, return_dict: bool=True) -> Union[(DDIMSchedulerOutput, Tuple)]: if (self.num_inference_steps is None): raise ValueError("Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler") prev_timestep = (timestep - (self.config.num_train_timesteps // self.num_inference_steps)) alpha_prod_t = self.alphas_cumprod[timestep] alpha_prod_t_prev = (self.alphas_cumprod[prev_timestep] if (prev_timestep >= 0) else self.final_alpha_cumprod) beta_prod_t = (1 - alpha_prod_t) if (self.config.prediction_type == 'epsilon'): pred_original_sample = ((sample - ((beta_prod_t ** 0.5) * model_output)) / (alpha_prod_t ** 0.5)) pred_epsilon = model_output elif (self.config.prediction_type == 'sample'): pred_original_sample = model_output pred_epsilon = ((sample - ((alpha_prod_t ** 0.5) * pred_original_sample)) / (beta_prod_t ** 0.5)) elif (self.config.prediction_type == 'v_prediction'): pred_original_sample = (((alpha_prod_t ** 0.5) * sample) - ((beta_prod_t ** 0.5) * model_output)) pred_epsilon = (((alpha_prod_t ** 0.5) * model_output) + ((beta_prod_t ** 0.5) * sample)) else: raise ValueError(f'prediction_type given as {self.config.prediction_type} must be one of `epsilon`, `sample`, or `v_prediction`') if self.config.thresholding: pred_original_sample = self._threshold_sample(pred_original_sample) elif self.config.clip_sample: pred_original_sample = pred_original_sample.clamp((- self.config.clip_sample_range), self.config.clip_sample_range) variance = self._get_variance(timestep, prev_timestep) std_dev_t = (eta * (variance ** 0.5)) if use_clipped_model_output: pred_epsilon = ((sample - ((alpha_prod_t ** 0.5) * pred_original_sample)) / (beta_prod_t ** 0.5)) pred_sample_direction = ((((1 - alpha_prod_t_prev) - (std_dev_t ** 2)) ** 0.5) * pred_epsilon) prev_sample = (((alpha_prod_t_prev ** 0.5) * pred_original_sample) + pred_sample_direction) if (eta > 0): if ((variance_noise is not None) and (generator is not None)): raise ValueError('Cannot pass both generator and variance_noise. Please make sure that either `generator` or `variance_noise` stays `None`.') if (variance_noise is None): variance_noise = randn_tensor(model_output.shape, generator=generator, device=model_output.device, dtype=model_output.dtype) variance = (std_dev_t * variance_noise) prev_sample = (prev_sample + variance) if (not return_dict): return (prev_sample,) return DDIMSchedulerOutput(prev_sample=prev_sample, pred_original_sample=pred_original_sample) def add_noise(self, original_samples: torch.FloatTensor, noise: torch.FloatTensor, timesteps: torch.IntTensor) -> torch.FloatTensor: alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device, dtype=original_samples.dtype) timesteps = timesteps.to(original_samples.device) sqrt_alpha_prod = (alphas_cumprod[timesteps] ** 0.5) sqrt_alpha_prod = sqrt_alpha_prod.flatten() while (len(sqrt_alpha_prod.shape) < len(original_samples.shape)): sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze((- 1)) sqrt_one_minus_alpha_prod = ((1 - alphas_cumprod[timesteps]) ** 0.5) sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten() while (len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape)): sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze((- 1)) noisy_samples = ((sqrt_alpha_prod * original_samples) + (sqrt_one_minus_alpha_prod * noise)) return noisy_samples def get_velocity(self, sample: torch.FloatTensor, noise: torch.FloatTensor, timesteps: torch.IntTensor) -> torch.FloatTensor: alphas_cumprod = self.alphas_cumprod.to(device=sample.device, dtype=sample.dtype) timesteps = timesteps.to(sample.device) sqrt_alpha_prod = (alphas_cumprod[timesteps] ** 0.5) sqrt_alpha_prod = sqrt_alpha_prod.flatten() while (len(sqrt_alpha_prod.shape) < len(sample.shape)): sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze((- 1)) sqrt_one_minus_alpha_prod = ((1 - alphas_cumprod[timesteps]) ** 0.5) sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten() while (len(sqrt_one_minus_alpha_prod.shape) < len(sample.shape)): sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze((- 1)) velocity = ((sqrt_alpha_prod * noise) - (sqrt_one_minus_alpha_prod * sample)) return velocity def __len__(self): return self.config.num_train_timesteps
def test_setr_up_head(capsys): with pytest.raises(AssertionError): SETRUPHead(num_classes=19, kernel_size=2) with pytest.raises(AssertionError): SETRUPHead(in_channels=(4, 4), channels=2, num_classes=19) head = SETRUPHead(in_channels=4, channels=2, norm_cfg=dict(type='SyncBN'), num_classes=19, init_cfg=dict(type='Kaiming')) super(SETRUPHead, head).init_weights() img_size = (4, 4) patch_size = 2 head = SETRUPHead(in_channels=4, channels=2, num_classes=19, num_convs=1, up_scale=4, kernel_size=1, norm_cfg=dict(type='BN')) (h, w) = ((img_size[0] // patch_size), (img_size[1] // patch_size)) x = [torch.randn(1, 4, h, w)] if torch.cuda.is_available(): (head, x) = to_cuda(head, x) out = head(x) assert (out.shape == (1, head.num_classes, (h * 4), (w * 4))) x = [torch.randn(1, 4, h, (w * 2))] if torch.cuda.is_available(): (head, x) = to_cuda(head, x) out = head(x) assert (out.shape == (1, head.num_classes, (h * 4), (w * 8)))
_module() class FCNHead(BaseDecodeHead): def __init__(self, num_convs=2, kernel_size=3, concat_input=True, dilation=1, **kwargs): assert ((num_convs >= 0) and (dilation > 0) and isinstance(dilation, int)) self.num_convs = num_convs self.concat_input = concat_input self.kernel_size = kernel_size super(FCNHead, self).__init__(**kwargs) if (num_convs == 0): assert (self.in_channels == self.channels) conv_padding = ((kernel_size // 2) * dilation) convs = [] for i in range(num_convs): _in_channels = (self.in_channels if (i == 0) else self.channels) convs.append(ConvModule(_in_channels, self.channels, kernel_size=kernel_size, padding=conv_padding, dilation=dilation, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg)) if (len(convs) == 0): self.convs = nn.Identity() else: self.convs = nn.Sequential(*convs) if self.concat_input: self.conv_cat = ConvModule((self.in_channels + self.channels), self.channels, kernel_size=kernel_size, padding=(kernel_size // 2), conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg, act_cfg=self.act_cfg) def _forward_feature(self, inputs): x = self._transform_inputs(inputs) feats = self.convs(x) if self.concat_input: feats = self.conv_cat(torch.cat([x, feats], dim=1)) return feats def forward(self, inputs): output = self._forward_feature(inputs) output = self.cls_seg(output) return output
def get_dataset_params(is_gcloud=False, tfrecord_dir=constants.NVIDIA_CELEBA_HQ_DATASET_PATH_GCLOUD, **kwargs): if is_gcloud: return CelebAHQDatasetParams(gcs_bucket=constants.GCLOUD_BUCKET, tfrecord_dir=tfrecord_dir, **kwargs) else: return CelebAHQDatasetParams(**kwargs)
def bg_white(seg, raw, blur_level=3, gaussian=81): seg = cv2.blur(seg, (blur_level, blur_level)) empty = np.ones_like(seg) seg_bg = ((empty - seg) * 255) seg_bg = cv2.GaussianBlur(seg_bg, (gaussian, gaussian), 0) background_mask = cv2.cvtColor((255 - cv2.cvtColor(seg, cv2.COLOR_BGR2GRAY)), cv2.COLOR_GRAY2BGR) masked_fg = ((raw * (1 / 255)) * (seg * (1 / 255))) masked_bg = ((seg_bg * (1 / 255)) * (background_mask * (1 / 255))) frame = np.uint8((cv2.add(masked_bg, masked_fg) * 255)) return frame
def test_disaggregated_scores_are_determinstic(): no_aggregation = calculate_rouge(PRED, TGT, bootstrap_aggregation=False, rouge_keys=['rouge2', 'rougeL']) assert isinstance(no_aggregation, defaultdict) no_aggregation_just_r2 = calculate_rouge(PRED, TGT, bootstrap_aggregation=False, rouge_keys=['rouge2']) assert (pd.DataFrame(no_aggregation['rouge2']).fmeasure.mean() == pd.DataFrame(no_aggregation_just_r2['rouge2']).fmeasure.mean())
def _find_my_group(grouped_ranks): index = _find_my_group_index(grouped_ranks) return grouped_ranks[index]
_module() class OBBRetinaHead(OBBAnchorHead): def __init__(self, num_classes, in_channels, stacked_convs=4, conv_cfg=None, norm_cfg=None, anchor_generator=dict(type='AnchorGenerator', octave_base_scale=4, scales_per_octave=3, ratios=[0.5, 1.0, 2.0], strides=[8, 16, 32, 64, 128]), **kwargs): self.stacked_convs = stacked_convs self.conv_cfg = conv_cfg self.norm_cfg = norm_cfg super(OBBRetinaHead, self).__init__(num_classes, in_channels, bbox_type='obb', reg_dim=5, anchor_generator=anchor_generator, **kwargs) def _init_layers(self): self.relu = nn.ReLU(inplace=True) self.cls_convs = nn.ModuleList() self.reg_convs = nn.ModuleList() for i in range(self.stacked_convs): chn = (self.in_channels if (i == 0) else self.feat_channels) self.cls_convs.append(ConvModule(chn, self.feat_channels, 3, stride=1, padding=1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg)) self.reg_convs.append(ConvModule(chn, self.feat_channels, 3, stride=1, padding=1, conv_cfg=self.conv_cfg, norm_cfg=self.norm_cfg)) self.retina_cls = nn.Conv2d(self.feat_channels, (self.num_anchors * self.cls_out_channels), 3, padding=1) self.retina_reg = nn.Conv2d(self.feat_channels, (self.num_anchors * self.reg_dim), 3, padding=1) def init_weights(self): for m in self.cls_convs: normal_init(m.conv, std=0.01) for m in self.reg_convs: normal_init(m.conv, std=0.01) bias_cls = bias_init_with_prob(0.01) normal_init(self.retina_cls, std=0.01, bias=bias_cls) normal_init(self.retina_reg, std=0.01) def forward_single(self, x): cls_feat = x reg_feat = x for cls_conv in self.cls_convs: cls_feat = cls_conv(cls_feat) for reg_conv in self.reg_convs: reg_feat = reg_conv(reg_feat) cls_score = self.retina_cls(cls_feat) bbox_pred = self.retina_reg(reg_feat) return (cls_score, bbox_pred)
class NfCfg(): depths: Tuple[(int, int, int, int)] channels: Tuple[(int, int, int, int)] alpha: float = 0.2 gamma_in_act: bool = False stem_type: str = '3x3' stem_chs: Optional[int] = None group_size: Optional[int] = 8 attn_layer: Optional[str] = 'se' attn_kwargs: dict = field(default_factory=(lambda : dict(reduction_ratio=0.5, divisor=8))) attn_gain: float = 2.0 width_factor: float = 0.75 bottle_ratio: float = 2.25 efficient: bool = True num_features: int = 1280 ch_div: int = 8 skipinit: bool = False act_layer: str = 'silu'
class PositionSortLauncher(): def __init__(self): pass def _no_opposite(self, handle): return (not _has_opposite(self.env, handle)) def _by_cities(self): self._city_n = ([(- 1)] * len(self.env.agents)) timer = 0 for (handle, agent) in enumerate(self.env.agents): (cur_x, cur_y) = agent.initial_position for prev in range(handle): (x, y) = self.env.agents[prev].initial_position if ((abs((x - cur_x)) < 10) and (abs((y - cur_y)) < 10)): self._city_n[handle] = self._city_n[prev] if (self._city_n[handle] == (- 1)): self._city_n[handle] = timer timer += 1 def reset(self, env): self.env = env self.order = [i for i in range(len(self.env.agents))] self._by_cities() self.order.sort(key=(lambda handle: self._city_n[handle])) self.ready_to_depart = ([0] * len(self.env.agents)) self.cur_pos = 0 self.send_more = 0 self.max_city = (- 1) def update(self): for handle in range(len(self.env.agents)): if (((self.env.agents[handle].status == RailAgentStatus.DONE_REMOVED) or self.env.obs_builder.deadlock_checker.is_deadlocked(handle)) and (self.ready_to_depart[handle] == 1)): self.ready_to_depart[handle] = 2 self.send_more += 1 if (self.send_more >= (((- self.max_city) * 2) - 2)): self.max_city += 1 for pos in range(len(self.env.agents)): handle = self.order[pos] if ((self.ready_to_depart[handle] == 0) and (self._city_n[handle] <= self.max_city) and self._no_opposite(handle)): self.send_more -= 1 self.ready_to_depart[handle] = 1 def is_ready(self, handle): return (self.ready_to_depart[handle] != 0)
def m_dreg_looser(model, x, K=1): S = compute_microbatch_split(x, K) x_split = zip(*[_x.split(S) for _x in x]) (lw, zss) = zip(*[_m_dreg_looser(model, _x, K) for _x in x_split]) lw = torch.cat(lw, 2) zss = torch.cat(zss, 2) with torch.no_grad(): grad_wt = (lw - torch.logsumexp(lw, 1, keepdim=True)).exp() if zss.requires_grad: zss.register_hook((lambda grad: (grad_wt.unsqueeze((- 1)) * grad))) return (grad_wt * lw).mean(0).sum()
def torch2numpy(input): assert isinstance(input, torch.Tensor), type(input) return input.detach().cpu().numpy()
def process_feature(feature: example_pb2.Feature, typename: str, typename_mapping: Dict, key: str) -> np.ndarray: field = feature.ListFields()[0] (inferred_typename, value) = (field[0].name, field[1].value) if (typename is not None): tf_typename = typename_mapping[typename] if (tf_typename != inferred_typename): reversed_mapping = {v: k for (k, v) in typename_mapping.items()} raise TypeError(f"Incompatible type '{typename}' for `{key}` (should be '{reversed_mapping[inferred_typename]}').") if (inferred_typename == 'bytes_list'): value = np.frombuffer(value[0], dtype=np.uint8) elif (inferred_typename == 'float_list'): value = np.array(value, dtype=np.float32) elif (inferred_typename == 'int64_list'): value = np.array(value, dtype=np.int64) return value
def save_checkpoint(obj, directory, step_num, use_thread=False): if use_thread: warnings.warn('use_threads set to True, but done synchronously still') os.makedirs(directory, exist_ok=True) torch.save(obj, checkpoint_name(directory), pickle_module=pickle) torch.save(obj, checkpoint_name(directory, step_num), pickle_module=pickle)
def prepare_t5(tokenizer, data_dir, max_input_length, max_output_length, lower=True): train_file = f'{data_dir}/train' dev_file = f'{data_dir}/dev' test_file = f'{data_dir}/test' train_out = f'{data_dir}/train_{max_input_length}_{max_output_length}.t5' dev_out = f'{data_dir}/dev_{max_input_length}_{max_output_length}.t5' test_out = f'{data_dir}/test_{max_input_length}_{max_output_length}.t5' if path.exists((train_file + '.source')): print(f'prepare {train_out}') (train_examples, train_eval) = process_file_t5(train_file, tokenizer, lower=lower) build_features_t5(train_examples, 'train', train_out, tokenizer, max_input_length=max_input_length, max_output_length=max_output_length) if path.exists((dev_file + '.source')): print(f'prepare {dev_out}') (dev_examples, dev_eval) = process_file_t5(dev_file, tokenizer, lower=lower) build_features_t5(dev_examples, 'dev', dev_out, tokenizer, max_input_length=max_input_length, max_output_length=max_output_length) if path.exists((test_file + '.source')): print(f'prepare {test_out}') (test_examples, test_eval) = process_file_t5(test_file, tokenizer, lower=lower) build_features_t5(test_examples, 'test', test_out, tokenizer, max_input_length=max_input_length, max_output_length=max_output_length)
class simpleMLP(nn.Module): def __init__(self, i_c=1, n_c=10): super(simpleMLP, self).__init__() self.flatten = Expression((lambda tensor: tensor.view(tensor.shape[0], (- 1)))) self.fc1 = nn.Linear((28 * 28), 256, bias=True) self.fc2 = nn.Linear(256, 128, bias=True) self.fc3 = nn.Linear(128, n_c, bias=True) def forward(self, x_i, _eval=False): x_o = self.flatten(x_i) x_o = torch.relu(self.fc1(x_o)) x_o = torch.relu(self.fc2(x_o)) return self.fc3(x_o)
_features_generator('morgan_count') def morgan_counts_features_generator(mol: Molecule, radius: int=MORGAN_RADIUS, num_bits: int=MORGAN_NUM_BITS) -> np.ndarray: mol = (Chem.MolFromSmiles(mol) if (type(mol) == str) else mol) features_vec = AllChem.GetHashedMorganFingerprint(mol, radius, nBits=num_bits) features = np.zeros((1,)) DataStructs.ConvertToNumpyArray(features_vec, features) return features
def initgen(mesh_size, freq=3, boundary='Periodic', dtype=None, device=None, batch_size=1): xs = [] for k in range(batch_size): xs.append(_initgen(mesh_size, freq=freq, boundary=boundary, dtype=dtype, device=device)) x = torch.stack(xs, dim=0) if (batch_size == 1): return x[0] else: return x
class RankingAndFitnessSelection(Selection[(List[S], List[S])]): def __init__(self, max_population_size: int, reference_point: S, dominance_comparator: Comparator=DominanceComparator()): super(RankingAndFitnessSelection, self).__init__() self.max_population_size = max_population_size self.dominance_comparator = dominance_comparator self.reference_point = reference_point def hypesub(self, l, A, actDim, bounds, pvec, alpha, k): h = [0 for _ in range(l)] Adim = [a[(actDim - 1)] for a in A] indices_sort = sorted(range(len(Adim)), key=Adim.__getitem__) S = [A[j] for j in indices_sort] pvec = [pvec[j] for j in indices_sort] for i in range(1, (len(S) + 1)): if (i < len(S)): extrusion = (S[i][(actDim - 1)] - S[(i - 1)][(actDim - 1)]) else: extrusion = (bounds[(actDim - 1)] - S[(i - 1)][(actDim - 1)]) if (actDim == 1): if (i > k): break if (all(alpha) >= 0): for p in pvec[0:i]: h[p] = (h[p] + (extrusion * alpha[(i - 1)])) elif (extrusion > 0): h = [(h[j] + (extrusion * self.hypesub(l, S[0:i], (actDim - 1), bounds, pvec[0:i], alpha, k)[j])) for j in range(l)] return h def compute_hypervol_fitness_values(self, population: List[S], reference_point: S, k: int): points = [ind.objectives for ind in population] bounds = reference_point.objectives population_size = len(points) if (k < 0): k = population_size actDim = len(bounds) pvec = range(population_size) alpha = [] for i in range(1, (k + 1)): alpha.append((np.prod([(float((k - j)) / (population_size - j)) for j in range(1, i)]) / i)) f = self.hypesub(population_size, points, actDim, bounds, pvec, alpha, k) for i in range(len(population)): population[i].attributes['fitness'] = f[i] return population def execute(self, front: List[S]) -> List[S]: if (front is None): raise Exception('The front is null') elif (len(front) == 0): raise Exception('The front is empty') ranking = FastNonDominatedRanking(self.dominance_comparator) ranking.compute_ranking(front) ranking_index = 0 new_solution_list = [] while (len(new_solution_list) < self.max_population_size): if (len(ranking.get_subfront(ranking_index)) < (self.max_population_size - len(new_solution_list))): subfront = ranking.get_subfront(ranking_index) new_solution_list = (new_solution_list + subfront) ranking_index += 1 else: subfront = ranking.get_subfront(ranking_index) parameter_K = (len(subfront) - (self.max_population_size - len(new_solution_list))) while (parameter_K > 0): subfront = self.compute_hypervol_fitness_values(subfront, self.reference_point, parameter_K) subfront = sorted(subfront, key=(lambda x: x.attributes['fitness']), reverse=True) subfront = subfront[:(- 1)] parameter_K = (parameter_K - 1) new_solution_list = (new_solution_list + subfront) return new_solution_list def get_name(self) -> str: return 'Ranking and fitness selection'
def train_AugTune(args, io): train_loader = DataLoader(ModelNet40(args, partition='train'), num_workers=8, batch_size=args.batch_size, shuffle=True, drop_last=True) test_loader = DataLoader(ModelNet40(args, partition='test'), num_workers=8, batch_size=args.test_batch_size, shuffle=True, drop_last=False) device = torch.device(('cuda' if args.cuda else 'cpu')) if (args.model == 'pointnet'): model = PointNet(args).to(device) elif (args.model == 'dgcnn'): model = DGCNN(args).to(device) else: raise Exception('Not implemented') print(str(model)) model = nn.DataParallel(model) print("Let's use", torch.cuda.device_count(), 'GPUs!') if args.use_sgd: print('Use SGD') opt = optim.SGD(model.parameters(), lr=(args.lr * 100), momentum=args.momentum, weight_decay=0.0001) else: print('Use Adam') opt = optim.Adam(model.parameters(), lr=args.lr, weight_decay=0.0001) scheduler = CosineAnnealingLR(opt, args.epochs, eta_min=args.lr) criterion = cal_loss best_test_acc = 0 for epoch in range(args.epochs): train_loss = 0.0 count = 0.0 model.train() train_pred = [] train_true = [] for (origin, data, label) in train_loader: (origin, data, label) = (origin.to(device), data.to(device), label.to(device).squeeze()) origin = origin.permute(0, 2, 1) data = data.permute(0, 2, 1) batch_size = data.size()[0] with torch.no_grad(): pred_origin = model(origin) pred_data = model(data) c_origin = (pred_origin.exp() * F.one_hot(label, pred_origin.shape[(- 1)])).sum(1) c_data = (pred_data.exp() * F.one_hot(label, pred_data.shape[(- 1)])).sum(1) c_target = torch.max(((1 - args.l) * c_origin), c_data) alpha = ((c_target - c_data) / ((c_origin - c_data) + 0.0001)).unsqueeze(1) alpha = torch.clamp(alpha, min=0, max=1).reshape((- 1), 1, 1) data = ((alpha * origin) + ((1 - alpha) * data)) data = normalize_point_cloud_batch(data) data = translate_pointcloud_batch(data) opt.zero_grad() logits = model(data) loss = criterion(logits, label) loss.backward() opt.step() preds = logits.max(dim=1)[1] count += batch_size train_loss += (loss.item() * batch_size) train_true.append(label.cpu().numpy()) train_pred.append(preds.detach().cpu().numpy()) scheduler.step() train_true = np.concatenate(train_true) train_pred = np.concatenate(train_pred) outstr = ('Train %d, loss: %.6f, train acc: %.6f, train avg acc: %.6f' % (epoch, ((train_loss * 1.0) / count), metrics.accuracy_score(train_true, train_pred), metrics.balanced_accuracy_score(train_true, train_pred))) io.cprint(outstr) test_loss = 0.0 count = 0.0 model.eval() test_pred = [] test_true = [] for (data, label) in test_loader: (data, label) = (data.to(device), label.to(device).squeeze()) data = data.permute(0, 2, 1) batch_size = data.size()[0] logits = model(data) loss = criterion(logits, label) preds = logits.max(dim=1)[1] count += batch_size test_loss += (loss.item() * batch_size) test_true.append(label.cpu().numpy()) test_pred.append(preds.detach().cpu().numpy()) test_true = np.concatenate(test_true) test_pred = np.concatenate(test_pred) test_acc = metrics.accuracy_score(test_true, test_pred) avg_per_class_acc = metrics.balanced_accuracy_score(test_true, test_pred) outstr = ('Test %d, loss: %.6f, test acc: %.6f, test avg acc: %.6f' % (epoch, ((test_loss * 1.0) / count), test_acc, avg_per_class_acc)) io.cprint(outstr) if (test_acc >= best_test_acc): best_test_acc = test_acc torch.save(model.state_dict(), ('checkpoints/%s/models/model.t7' % args.exp_name))
def _prepare_output_docstrings(output_type, config_class): docstrings = output_type.__doc__ lines = docstrings.split('\n') i = 0 while ((i < len(lines)) and (re.search('^\\s*(Args|Parameters):\\s*$', lines[i]) is None)): i += 1 if (i < len(lines)): docstrings = '\n'.join(lines[(i + 1):]) docstrings = _convert_output_args_doc(docstrings) full_output_type = f'{output_type.__module__}.{output_type.__name__}' intro = (TF_RETURN_INTRODUCTION if output_type.__name__.startswith('TF') else PT_RETURN_INTRODUCTION) intro = intro.format(full_output_type=full_output_type, config_class=config_class) return (intro + docstrings)
class TestFoldBatchnorm(unittest.TestCase): tf.compat.v1.disable_eager_execution() x = tf.compat.v1.placeholder(tf.float32, [1, 224, 224, 3], name='input') conv_weights = tf.compat.v1.get_variable('weight', [3, 3, 3, 32], initializer=tf.compat.v1.random_normal_initializer()) conv_bias = tf.compat.v1.get_variable('bias', [32], initializer=tf.compat.v1.random_normal_initializer()) beta = tf.compat.v1.get_variable(name='beta', shape=[32], initializer=tf.compat.v1.random_normal_initializer()) gamma = tf.compat.v1.get_variable(name='gamma', shape=[32], initializer=tf.compat.v1.random_normal_initializer()) conv1 = tf.nn.conv2d(x, conv_weights, strides=[1, 1, 1, 1], padding='SAME') conv_bias = tf.nn.bias_add(conv1, conv_bias) normed = tf.compat.v1.layers.batch_normalization(conv_bias) with tf.compat.v1.Session() as sess: sess.run(tf.compat.v1.global_variables_initializer()) output_graph_def = graph_util.convert_variables_to_constants(sess=sess, input_graph_def=sess.graph_def, output_node_names=[normed.name.split(':')[0]]) output_graph_def = QuantizeGraphHelper.remove_training_nodes(output_graph_def, protected_nodes=[normed.name.split(':')[0]]) graph_def = copy.deepcopy(output_graph_def) fold_graph_def = FoldBatchNormNodesOptimizer(output_graph_def).do_transformation() def test_fold_output_values(self): input_data = np.random.randn(1, 224, 224, 3) graph = tf.compat.v1.Graph() fold_graph = tf.compat.v1.Graph() with graph.as_default(): tf.compat.v1.import_graph_def(self.graph_def, name='') with tf.compat.v1.Session(graph=graph) as sess: sess.run(tf.compat.v1.global_variables_initializer()) x = graph.get_tensor_by_name('input:0') normed = graph.get_tensor_by_name('batch_normalization/FusedBatchNormV3:0') y = sess.run(normed, feed_dict={x: input_data}) with fold_graph.as_default(): tf.compat.v1.import_graph_def(self.fold_graph_def, name='') with tf.compat.v1.Session(graph=fold_graph) as sess: sess.run(tf.compat.v1.global_variables_initializer()) x = fold_graph.get_tensor_by_name('input:0') normed = fold_graph.get_tensor_by_name('batch_normalization/FusedBatchNormV3:0') y_fold = sess.run(normed, feed_dict={x: input_data}) assert np.allclose(y, y_fold, rtol=1e-05, atol=1e-05) def test_do_transform(self): for node in self.fold_graph_def.node: assert (node.op not in ['FusedBatchNormV3'])
def get_wsi_loader(data_dir, batch_size=1, shuffle=False, num_threads=2, train_eval_test='val', splitter_path='./', device_id=0, num_gpus=1, seed=1, bag_size=1024, label_csv_path='./', split_num=0): eii = ExternalInputCallable(data_dir=data_dir, batch_size=batch_size, split_num=split_num, splitter_path=splitter_path, shuffle=shuffle, device_id=device_id, num_gpus=num_gpus, train_eval_test=train_eval_test, bag_size=bag_size, label_csv_path=label_csv_path) img_size = 224 if (train_eval_test == 'train'): pipe = TrainPipeline(batch_size=(batch_size * bag_size), eii=eii, num_threads=num_threads, device_id=device_id, seed=(seed + device_id), img_size=img_size) elif (train_eval_test == 'val'): pipe = ValPipeline(batch_size=(batch_size * bag_size), eii=eii, num_threads=num_threads, device_id=device_id, seed=(seed + device_id), img_size=img_size) else: pipe = ValPipeline(batch_size=(batch_size * bag_size), eii=eii, num_threads=num_threads, device_id=device_id, seed=(seed + device_id), use_gpu=True, img_size=img_size) pipe.build() loader = DALIClassificationIterator(pipe, size=(eii.size * bag_size), auto_reset=True, last_batch_padded=True, prepare_first_batch=False) return loader
class AlignTextModelTester(): def __init__(self, parent, batch_size=12, seq_length=7, is_training=True, use_input_mask=True, use_token_type_ids=True, vocab_size=99, hidden_size=32, num_hidden_layers=5, num_attention_heads=4, intermediate_size=37, hidden_act='gelu', hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, max_position_embeddings=512, type_vocab_size=16, type_sequence_label_size=2, initializer_range=0.02, scope=None): self.parent = parent self.batch_size = batch_size self.seq_length = seq_length self.is_training = is_training self.use_input_mask = use_input_mask self.use_token_type_ids = use_token_type_ids self.vocab_size = vocab_size self.hidden_size = hidden_size self.num_hidden_layers = num_hidden_layers self.num_attention_heads = num_attention_heads self.intermediate_size = intermediate_size self.hidden_act = hidden_act self.hidden_dropout_prob = hidden_dropout_prob self.attention_probs_dropout_prob = attention_probs_dropout_prob self.max_position_embeddings = max_position_embeddings self.type_vocab_size = type_vocab_size self.type_sequence_label_size = type_sequence_label_size self.initializer_range = initializer_range self.scope = scope def prepare_config_and_inputs(self): input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size) input_mask = None if self.use_input_mask: input_mask = random_attention_mask([self.batch_size, self.seq_length]) token_type_ids = None if self.use_token_type_ids: token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size) config = self.get_config() return (config, input_ids, token_type_ids, input_mask) def get_config(self): return AlignTextConfig(vocab_size=self.vocab_size, hidden_size=self.hidden_size, num_hidden_layers=self.num_hidden_layers, num_attention_heads=self.num_attention_heads, intermediate_size=self.intermediate_size, hidden_act=self.hidden_act, hidden_dropout_prob=self.hidden_dropout_prob, attention_probs_dropout_prob=self.attention_probs_dropout_prob, max_position_embeddings=self.max_position_embeddings, type_vocab_size=self.type_vocab_size, is_decoder=False, initializer_range=self.initializer_range) def create_and_check_model(self, config, input_ids, token_type_ids, input_mask): model = AlignTextModel(config=config) model.to(torch_device) model.eval() with torch.no_grad(): result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids) result = model(input_ids, token_type_ids=token_type_ids) result = model(input_ids) self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.seq_length, self.hidden_size)) self.parent.assertEqual(result.pooler_output.shape, (self.batch_size, self.hidden_size)) def prepare_config_and_inputs_for_common(self): config_and_inputs = self.prepare_config_and_inputs() (config, input_ids, token_type_ids, input_mask) = config_and_inputs inputs_dict = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask': input_mask} return (config, inputs_dict)
class TestCompletionDataset(unittest.TestCase): def setUpClass(self): self.tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) class TestArgs(): train_on_inputs = False task = 'completion' max_seq_length = 512 max_source_length = 256 dataset_name = 'tatsu-lab/alpaca' self.test_args = TestArgs() self.sample_datasets = DatasetDict() raw_datasets = load_dataset('tatsu-lab/alpaca') self.sample_datasets['train'] = raw_datasets['train'].select(range(100)) def test_process(self): (raw_datasets, preprocess_fn) = data_utils.preprocess_dataset(self.sample_datasets, self.tokenizer, self.test_args, self.test_args) column_names = list(raw_datasets['train'].features) tokenized_datasets = raw_datasets.map(preprocess_fn, batched=True, remove_columns=column_names) self.assertTrue(isinstance(tokenized_datasets, DatasetDict))
(frozen=True) class Task(): id: Optional[int] name: str description: str version: str problem: str origin: str config: dict assets: List[Asset] measures: Dict[(str, Measure)] def measure(self, name: str) -> pd.DataFrame: return self.measures[name].values def scalar_measure(self, name: str) -> Union[(Number, str, dict)]: return self.measures[name].values['val'].iloc[0] def data(self, aliases: Iterable[str]) -> List[object]: from powerlift.bench.store import BytesParser outputs = [] alias_map = self.config['aliases'] name_to_asset = {asset.name: asset for asset in self.assets} for alias in aliases: name = alias_map[alias] asset = name_to_asset[name] parsed = BytesParser.deserialize(asset.mimetype, asset.embedded) outputs.append(parsed) return outputs
def get_stat_in_paths(paths, dict_name, scalar_name): if (len(paths) == 0): return np.array([[]]) if (type(paths[0][dict_name]) == dict): return [path[dict_name][scalar_name] for path in paths] return [[info[scalar_name] for info in path[dict_name]] for path in paths]
def read_csv(filename, loss_name='val/loss'): import codecs import csv fit_out = {} with codecs.open(filename, encoding='utf-8-sig') as f: for row in csv.DictReader(f, skipinitialspace=True): if row[loss_name]: fit_out[row['epoch']] = {'val_loss': row[loss_name]} return fit_out
class Lexicon(lazydict): def __init__(self, path=''): self._path = path def path(self): return self._path def load(self): dict.update(self, (x.split(' ')[:2] for x in _read(self._path) if (len(x.split(' ')) > 1)))
class CamembertConfig(RobertaConfig): pretrained_config_archive_map = CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP model_type = 'camembert'
class MobileBertForQuestionAnswering(metaclass=DummyObject): _backends = ['torch'] def __init__(self, *args, **kwargs): requires_backends(self, ['torch'])
def evaluate(model): model.compile(run_eagerly=False) postprocess = LabelShift(label_shift=1) from neural_compressor import METRICS metrics = METRICS('tensorflow') metric = metrics['topk']() latency_list = [] def eval_func(dataloader, metric): warmup = 5 iteration = None if (FLAGS.benchmark and (FLAGS.mode == 'performance')): iteration = FLAGS.iters for (idx, (inputs, labels)) in enumerate(dataloader): start = time.time() predictions = model.predict_on_batch(inputs) end = time.time() latency_list.append((end - start)) (predictions, labels) = postprocess((predictions, labels)) metric.update(predictions, labels) if (iteration and (idx >= iteration)): break latency = (np.array(latency_list[warmup:]).mean() / eval_dataloader.batch_size) return latency latency = eval_func(eval_dataloader, metric) if FLAGS.benchmark: logger.info('\n{} mode benchmark result:'.format(FLAGS.mode)) for (i, res) in enumerate(latency_list): logger.debug('Iteration {} result {}:'.format(i, res)) if (FLAGS.benchmark and (FLAGS.mode == 'performance')): logger.info('Batch size = {}'.format(eval_dataloader.batch_size)) logger.info('Latency: {:.3f} ms'.format((latency * 1000))) logger.info('Throughput: {:.3f} images/sec'.format((1.0 / latency))) acc = metric.result() return acc
def factors(n): lst = [] i = 1 while (i <= n): if ((n % i) == 0): lst.append(i) i += 1 return lst
(t='double', spline='Spline', returns='double') def scale_factor(t=(- 1)): if (not enable_Hubble): return 1 if (t == (- 1)): t = universals.t spline = temporal_splines.t_a if (spline is None): abort('The function a(t) has not been tabulated. Have you called init_time?') return spline.eval(t)
def processed(f): (f) def wrapper(*args, **kwargs): func = Process(target=f, args=args, kwargs=kwargs) func.daemon = False func.start() return func return wrapper
def dataset_renderer_worker(log_ids: List[str], start_idx: int, end_idx: int, worker_id: int, kwargs: Mapping[(str, Any)]) -> None: logging.info(f'Worker {worker_id} started...') local_dataset_dir = kwargs['local_dataset_dir'] config = kwargs['config'] dataloader = kwargs['dataloader'] use_gpu = (isinstance(config, BevRenderingConfig) and (config.recompute_segmentation or (config.projection_method == 'ray_tracing'))) if use_gpu: if (not torch.cuda.is_available()): raise RuntimeError('CUDA is not supported on your platform.') num_gpus = torch.cuda.device_count() gpu_id = (worker_id // (config.num_processes // num_gpus)) os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu_id) logging.info('Creating Argoverse dataloader...') chunk_sz = (end_idx - start_idx) for idx in range(start_idx, end_idx): if ((idx % 10) == 0): pct_completed = (((idx - start_idx) / chunk_sz) * 100) logging.info(f'Completed {pct_completed:.2f}%') log_id = log_ids[idx] try: log_dir = ((Path(local_dataset_dir) / 'logs') / log_id) render_log_dataset(config=config, local_dataset_dir=local_dataset_dir, log_id=log_id, log_dir=log_dir, dataloader=dataloader) except Exception as e: logging.exception(f'Extraction failed for {log_id}')
class CiderScorer(object): def copy(self): new = CiderScorer(n=self.n) new.ctest = copy.copy(self.ctest) new.crefs = copy.copy(self.crefs) return new def copy_empty(self): new = CiderScorer(df_mode='corpus', n=self.n, sigma=self.sigma) new.df_mode = self.df_mode new.ref_len = self.ref_len new.document_frequency = self.document_frequency return new def __init__(self, df_mode='corpus', test=None, refs=None, n=4, sigma=6.0): self.n = n self.sigma = sigma self.crefs = [] self.ctest = [] self.df_mode = df_mode self.ref_len = None if (self.df_mode != 'corpus'): pkl_file = cPickle.load(open(os.path.join('data', (df_mode + '.p')), 'rb'), **(dict(encoding='latin1') if six.PY3 else {})) self.ref_len = np.log(float(pkl_file['ref_len'])) self.document_frequency = pkl_file['document_frequency'] self.cook_append(test, refs) def clear(self): self.crefs = [] self.ctest = [] def cook_append(self, test, refs): if (refs is not None): self.crefs.append(cook_refs(refs)) if (test is not None): self.ctest.append(cook_test(test)) else: self.ctest.append(None) def size(self): assert (len(self.crefs) == len(self.ctest)), ('refs/test mismatch! %d<>%d' % (len(self.crefs), len(self.ctest))) return len(self.crefs) def __iadd__(self, other): if (type(other) is tuple): self.cook_append(other[0], other[1]) else: self.ctest.extend(other.ctest) self.crefs.extend(other.crefs) return self def compute_doc_freq(self): for refs in self.crefs: for ngram in set([ngram for ref in refs for (ngram, count) in ref.items()]): self.document_frequency[ngram] += 1 def compute_cider(self): def counts2vec(cnts): vec = [defaultdict(float) for _ in range(self.n)] length = 0 norm = [0.0 for _ in range(self.n)] for (ngram, term_freq) in cnts.items(): df = np.log(max(1.0, self.document_frequency[ngram])) n = (len(ngram) - 1) vec[n][ngram] = (float(term_freq) * (self.ref_len - df)) norm[n] += pow(vec[n][ngram], 2) if (n == 1): length += term_freq norm = [np.sqrt(n) for n in norm] return (vec, norm, length) def sim(vec_hyp, vec_ref, norm_hyp, norm_ref, length_hyp, length_ref): delta = float((length_hyp - length_ref)) val = np.array([0.0 for _ in range(self.n)]) for n in range(self.n): for (ngram, count) in vec_hyp[n].items(): val[n] += (min(vec_hyp[n][ngram], vec_ref[n][ngram]) * vec_ref[n][ngram]) if ((norm_hyp[n] != 0) and (norm_ref[n] != 0)): val[n] /= (norm_hyp[n] * norm_ref[n]) assert (not math.isnan(val[n])) val[n] *= (np.e ** ((- (delta ** 2)) / (2 * (self.sigma ** 2)))) return val if (self.df_mode == 'corpus'): self.ref_len = np.log(float(len(self.crefs))) scores = [] for (test, refs) in zip(self.ctest, self.crefs): (vec, norm, length) = counts2vec(test) score = np.array([0.0 for _ in range(self.n)]) for ref in refs: (vec_ref, norm_ref, length_ref) = counts2vec(ref) score += sim(vec, vec_ref, norm, norm_ref, length, length_ref) score_avg = np.mean(score) score_avg /= len(refs) score_avg *= 10.0 scores.append(score_avg) return scores def compute_score(self, option=None, verbose=0): if (self.df_mode == 'corpus'): self.document_frequency = defaultdict(float) self.compute_doc_freq() assert (len(self.ctest) >= max(self.document_frequency.values())) score = self.compute_cider() return (np.mean(np.array(score)), np.array(score))
class Config(): library_path = None library_file = None compatibility_check = False loaded = False def set_library_path(path): if Config.loaded: raise Exception('library path must be set before before using any other functionalities in libclang.') Config.library_path = path def set_library_file(filename): if Config.loaded: raise Exception('library file must be set before before using any other functionalities in libclang.') Config.library_file = filename def set_compatibility_check(check_status): if Config.loaded: raise Exception('compatibility_check must be set before before using any other functionalities in libclang.') Config.compatibility_check = check_status def lib(self): lib = self.get_cindex_library() register_functions(lib, (not Config.compatibility_check)) Config.loaded = True return lib def get_filename(self): if Config.library_file: return Config.library_file import platform name = platform.system() if (name == 'Darwin'): file = 'libclang.dylib' elif (name == 'Windows'): file = 'libclang.dll' else: file = 'libclang.so' if Config.library_path: file = ((Config.library_path + '/') + file) return file def get_cindex_library(self): try: library = cdll.LoadLibrary(self.get_filename()) except OSError as e: msg = (str(e) + '. To provide a path to libclang use Config.set_library_path() or Config.set_library_file().') raise LibclangError(msg) return library def function_exists(self, name): try: getattr(self.lib, name) except AttributeError: return False return True
class AdaINorm2d(_AdaINorm): def _check_input_dim(self, input): if (input.dim() != 4): raise ValueError('expected 4D input (got {}D input)'.format(input.dim()))
def remove_tmp_file(func): (func) def wrapper(*args, **kwargs): onnx_file = 'tmp.onnx' kwargs['onnx_file'] = onnx_file try: result = func(*args, **kwargs) finally: if os.path.exists(onnx_file): os.remove(onnx_file) return result return wrapper
def adjust_axes(r, t, fig, axes): bb = t.get_window_extent(renderer=r) text_width_inches = (bb.width / fig.dpi) current_fig_width = fig.get_figwidth() new_fig_width = (current_fig_width + text_width_inches) propotion = (new_fig_width / current_fig_width) x_lim = axes.get_xlim() axes.set_xlim([x_lim[0], (x_lim[1] * propotion)])
(scope='module') def lapicque_hidden_reset_none_instance(): return snn.Lapicque(beta=0.5, init_hidden=True, reset_mechanism='none')
def gen_voxel(cropped, com_2d, cube, voxel_len): (H, W) = cropped.shape x = np.arange(H) y = np.arange(W) (x, y) = np.meshgrid(x, y, indexing='ij') z = cropped.copy() mask = np.bitwise_and((cropped >= (com_2d[2] - (cube[2] / 2.0))), (cropped < (com_2d[2] + (cube[2] / 2.0)))) mask = mask.reshape((- 1)) x = x.reshape((- 1))[mask] y = y.reshape((- 1))[mask] z = z.reshape((- 1))[mask] x = (x / H) y = (y / W) z = (((z - com_2d[2]) + (cube[2] / 2)) / cube[2]) voxel = np.zeros([voxel_len, voxel_len, voxel_len], dtype=np.int) x = (x * voxel_len).astype(np.int) y = (y * voxel_len).astype(np.int) z = (z * voxel_len).astype(np.int) voxel[(x, y, z)] = 1 return voxel
def eval_test(model, test_loader): model.eval() test_loss = 0 correct = 0 with torch.no_grad(): for (data, target) in test_loader: (data, target) = (data.cuda(), target.cuda()) output = model(data) test_loss += F.cross_entropy(output, target, size_average=False).item() pred = output.max(1, keepdim=True)[1] correct += pred.eq(target.view_as(pred)).sum().item() test_loss /= len(test_loader.dataset) print('Test: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)'.format(test_loss, correct, len(test_loader.dataset), ((100.0 * correct) / len(test_loader.dataset)))) test_accuracy = (correct / len(test_loader.dataset)) return (test_loss, test_accuracy)
def get_flat_arcs(index, sen): sid = sen.id arcs = [] for (j, token) in enumerate(sen): form = token.form lemma = token.lemma upos = token.upos xpos = token.xpos deprel = token.deprel if token.scope: for (head, lbl) in token.scope: lbl = (lbl[3:] if lbl.startswith('IN:') else lbl) arcs.append(FlatArc(index, sid, form, lemma, upos, xpos, deprel, j, lbl)) return arcs
class OneInstanceLauncher(Launcher): def launch(self, args, memory_prefix_list): processes = [] cmd = [] cmd_for_print = [] processes = [] tmp_log_path = '' cores = 1 current_path = os.path.abspath(os.getcwd()) batch_size_list = [] if (args.batch_size == 'auto'): batch_size_list = '1,2,4,8,16,32,64,128'.split(',') else: batch_size_list = args.batch_size.split(',') first_write = True instance = 1 if (args.instance_num == '1'): cores = self.cores_per_socket core_list = np.arange(0, cores) min_latency = min_config = Configs(1, instance, cores, 'disabled', 'default', 'cycle_buffer', '', args.mode) for batch_str in batch_size_list: batch_size = int(batch_str) for (mp_list_idx, mp_list_item) in enumerate(memory_prefix_list): cmd_prefix = get_cmd_prefix(core_list) cmd_prefix = (replace_instance_num(memory_prefix_list[mp_list_idx], instance) + cmd_prefix) cmd.clear() cmd_for_print.clear() set_cmd_prefix(cmd, core_list) cmd_for_print.append(cmd_prefix) cmd.append(sys.executable) cmd.append('-u') cmd_for_print.append(sys.executable) cmd_for_print.append('-u') weight_sharing = get_weight_sharing(cmd_prefix) memory_allocator = get_memory_allocator(cmd_prefix) memory_planning = get_memory_planning(cmd_prefix) import shlex cmd.append(shlex.quote(args.program)) cmd_for_print.append(args.program) batch_size = replace_batch(cmd, args, batch_size) replace_batch(cmd_for_print, args, batch_size) tmp_config = Configs(batch_size, instance, cores, weight_sharing, memory_allocator, memory_planning, memory_prefix_list[mp_list_idx], args.mode) tmp_log_path = get_tmp_log_path(tmp_config, current_path, 0) if (os.path.exists((current_path + '/all_latency')) == 0): os.mkdir((current_path + '/all_latency')) cmd.append(('--log_file=' + tmp_log_path)) cmd_for_print.append(('--log_file=' + tmp_log_path)) cmd_s = ' '.join(cmd_for_print) tmp_config.set_cmds(cmd_s) env_cmd = self.launcher_env set_numactl_env(env_cmd, core_list) set_jemalloc_env(env_cmd, memory_allocator, self.project_path) set_unified_buffer_env(env_cmd, memory_planning) set_weight_sharing(env_cmd, weight_sharing) set_instance_num(env_cmd, instance) process = subprocess.Popen(cmd, env=env_cmd, shell=False, stdout=subprocess.PIPE) processes.append(process) for process in processes: process.wait() if (process.returncode != 0): raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd_s) output_log_name = get_formal_log_path(args.output_file, current_path) tmp_latency = latency_mode_grab_log(current_path, output_log_name, tmp_config, False, first_write, 0) first_write = False if (tmp_latency < min_latency): min_latency = tmp_latency min_config.set_batch(batch_size) min_config.set_instance(instance) min_config.set_cores_per_instance(cores) min_config.set_weight_sharing(weight_sharing) min_config.set_memory_allocator(memory_allocator) min_config.set_memory_planning(memory_planning) min_config.set_cmds(cmd_s) latency_mode_grab_log(current_path, output_log_name, min_config, True, first_write, 0) elif (args.instance_num == 'auto'): cores = self.cores_per_socket min_latency = min_config = Configs(1, instance, cores, 'disabled', 'default', 'cycle_buffer', '', args.mode) for batch_str in batch_size_list: batch_size = int(batch_str) cores_iterator = int((cores / 2)) while (cores_iterator <= cores): core_list = np.arange(0, cores_iterator) cmd_prefix = get_cmd_prefix(core_list) for (mp_list_idx, mp_list_item) in enumerate(memory_prefix_list): cmd_prefix = (replace_instance_num(memory_prefix_list[mp_list_idx], instance) + cmd_prefix) cmd.clear() cmd_for_print.clear() set_cmd_prefix(cmd, core_list) cmd_for_print.append(cmd_prefix) cmd.append(sys.executable) cmd.append('-u') cmd_for_print.append(sys.executable) cmd_for_print.append('-u') weight_sharing = get_weight_sharing(cmd_prefix) memory_allocator = get_memory_allocator(cmd_prefix) memory_planning = get_memory_planning(cmd_prefix) import shlex cmd.append(shlex.quote(args.program)) cmd_for_print.append(args.program) batch_size = replace_batch(cmd, args, batch_size) replace_batch(cmd_for_print, args, batch_size) tmp_config = Configs(batch_size, instance, cores_iterator, weight_sharing, memory_allocator, memory_planning, memory_prefix_list[mp_list_idx], args.mode) tmp_log_path = get_tmp_log_path(tmp_config, current_path, 0) if (os.path.exists((current_path + '/all_latency')) == 0): os.mkdir((current_path + '/all_latency')) log_file_path = '--log_file={}'.format(tmp_log_path) cmd.append(log_file_path) cmd_for_print.append(('--log_file=' + tmp_log_path)) cmd.append(('--log_file=' + tmp_log_path)) cmd_s = ' '.join(cmd) tmp_config.set_cmds(cmd_s) env_cmd = self.launcher_env set_numactl_env(env_cmd, core_list) set_jemalloc_env(env_cmd, memory_allocator, self.project_path) set_unified_buffer_env(env_cmd, memory_planning) set_weight_sharing(env_cmd, weight_sharing) set_instance_num(env_cmd, instance) process = subprocess.Popen(cmd, env=env_cmd, shell=False, stdout=subprocess.PIPE) processes.append(process) for process in processes: process.wait() if (process.returncode != 0): raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) output_log_name = get_formal_log_path(args.output_file, current_path) latency_tmp = latency_mode_grab_log(current_path, output_log_name, tmp_config, False, first_write, 0) first_write = False if (latency_tmp < min_latency): min_latency = latency_tmp min_config.set_batch(batch_size) min_config.set_instance(instance) min_config.set_cores_per_instance(cores_iterator) min_config.set_weight_sharing(weight_sharing) min_config.set_memory_allocator(memory_allocator) min_config.set_memory_planning(memory_planning) min_config.set_cmds(tmp_config.get_cmds()) cores_iterator += 1 output_log_name = get_min_latency_output_log_path(min_config, current_path) if (args.output_file != ''): output_log_name = args.output_file latency_mode_grab_log(current_path, output_log_name, min_config, True, first_write, 0) else: print('Latency mode only support instance=auto or instance=1 !!!')
_module() class ADE20KDataset(CustomDataset): CLASSES = ('wall', 'building', 'sky', 'floor', 'tree', 'ceiling', 'road', 'bed ', 'windowpane', 'grass', 'cabinet', 'sidewalk', 'person', 'earth', 'door', 'table', 'mountain', 'plant', 'curtain', 'chair', 'car', 'water', 'painting', 'sofa', 'shelf', 'house', 'sea', 'mirror', 'rug', 'field', 'armchair', 'seat', 'fence', 'desk', 'rock', 'wardrobe', 'lamp', 'bathtub', 'railing', 'cushion', 'base', 'box', 'column', 'signboard', 'chest of drawers', 'counter', 'sand', 'sink', 'skyscraper', 'fireplace', 'refrigerator', 'grandstand', 'path', 'stairs', 'runway', 'case', 'pool table', 'pillow', 'screen door', 'stairway', 'river', 'bridge', 'bookcase', 'blind', 'coffee table', 'toilet', 'flower', 'book', 'hill', 'bench', 'countertop', 'stove', 'palm', 'kitchen island', 'computer', 'swivel chair', 'boat', 'bar', 'arcade machine', 'hovel', 'bus', 'towel', 'light', 'truck', 'tower', 'chandelier', 'awning', 'streetlight', 'booth', 'television receiver', 'airplane', 'dirt track', 'apparel', 'pole', 'land', 'bannister', 'escalator', 'ottoman', 'bottle', 'buffet', 'poster', 'stage', 'van', 'ship', 'fountain', 'conveyer belt', 'canopy', 'washer', 'plaything', 'swimming pool', 'stool', 'barrel', 'basket', 'waterfall', 'tent', 'bag', 'minibike', 'cradle', 'oven', 'ball', 'food', 'step', 'tank', 'trade name', 'microwave', 'pot', 'animal', 'bicycle', 'lake', 'dishwasher', 'screen', 'blanket', 'sculpture', 'hood', 'sconce', 'vase', 'traffic light', 'tray', 'ashcan', 'fan', 'pier', 'crt screen', 'plate', 'monitor', 'bulletin board', 'shower', 'radiator', 'glass', 'clock', 'flag') PALETTE = [[120, 120, 120], [180, 120, 120], [6, 230, 230], [80, 50, 50], [4, 200, 3], [120, 120, 80], [140, 140, 140], [204, 5, 255], [230, 230, 230], [4, 250, 7], [224, 5, 255], [235, 255, 7], [150, 5, 61], [120, 120, 70], [8, 255, 51], [255, 6, 82], [143, 255, 140], [204, 255, 4], [255, 51, 7], [204, 70, 3], [0, 102, 200], [61, 230, 250], [255, 6, 51], [11, 102, 255], [255, 7, 71], [255, 9, 224], [9, 7, 230], [220, 220, 220], [255, 9, 92], [112, 9, 255], [8, 255, 214], [7, 255, 224], [255, 184, 6], [10, 255, 71], [255, 41, 10], [7, 255, 255], [224, 255, 8], [102, 8, 255], [255, 61, 6], [255, 194, 7], [255, 122, 8], [0, 255, 20], [255, 8, 41], [255, 5, 153], [6, 51, 255], [235, 12, 255], [160, 150, 20], [0, 163, 255], [140, 140, 140], [250, 10, 15], [20, 255, 0], [31, 255, 0], [255, 31, 0], [255, 224, 0], [153, 255, 0], [0, 0, 255], [255, 71, 0], [0, 235, 255], [0, 173, 255], [31, 0, 255], [11, 200, 200], [255, 82, 0], [0, 255, 245], [0, 61, 255], [0, 255, 112], [0, 255, 133], [255, 0, 0], [255, 163, 0], [255, 102, 0], [194, 255, 0], [0, 143, 255], [51, 255, 0], [0, 82, 255], [0, 255, 41], [0, 255, 173], [10, 0, 255], [173, 255, 0], [0, 255, 153], [255, 92, 0], [255, 0, 255], [255, 0, 245], [255, 0, 102], [255, 173, 0], [255, 0, 20], [255, 184, 184], [0, 31, 255], [0, 255, 61], [0, 71, 255], [255, 0, 204], [0, 255, 194], [0, 255, 82], [0, 10, 255], [0, 112, 255], [51, 0, 255], [0, 194, 255], [0, 122, 255], [0, 255, 163], [255, 153, 0], [0, 255, 10], [255, 112, 0], [143, 255, 0], [82, 0, 255], [163, 255, 0], [255, 235, 0], [8, 184, 170], [133, 0, 255], [0, 255, 92], [184, 0, 255], [255, 0, 31], [0, 184, 255], [0, 214, 255], [255, 0, 112], [92, 255, 0], [0, 224, 255], [112, 224, 255], [70, 184, 160], [163, 0, 255], [153, 0, 255], [71, 255, 0], [255, 0, 163], [255, 204, 0], [255, 0, 143], [0, 255, 235], [133, 255, 0], [255, 0, 235], [245, 0, 255], [255, 0, 122], [255, 245, 0], [10, 190, 212], [214, 255, 0], [0, 204, 255], [20, 0, 255], [255, 255, 0], [0, 153, 255], [0, 41, 255], [0, 255, 204], [41, 0, 255], [41, 255, 0], [173, 0, 255], [0, 245, 255], [71, 0, 255], [122, 0, 255], [0, 255, 184], [0, 92, 255], [184, 255, 0], [0, 133, 255], [255, 214, 0], [25, 194, 194], [102, 255, 0], [92, 0, 255]] def __init__(self, **kwargs): super(ADE20KDataset, self).__init__(img_suffix='.jpg', seg_map_suffix='.png', reduce_zero_label=True, **kwargs) def results2img(self, results, imgfile_prefix, to_label_id): mmcv.mkdir_or_exist(imgfile_prefix) result_files = [] prog_bar = mmcv.ProgressBar(len(self)) for idx in range(len(self)): result = results[idx] filename = self.img_infos[idx]['filename'] basename = osp.splitext(osp.basename(filename))[0] png_filename = osp.join(imgfile_prefix, f'{basename}.png') result = (result + 1) output = Image.fromarray(result.astype(np.uint8)) output.save(png_filename) result_files.append(png_filename) prog_bar.update() return result_files def format_results(self, results, imgfile_prefix=None, to_label_id=True): assert isinstance(results, list), 'results must be a list' assert (len(results) == len(self)), f'The length of results is not equal to the dataset len: {len(results)} != {len(self)}' if (imgfile_prefix is None): tmp_dir = tempfile.TemporaryDirectory() imgfile_prefix = tmp_dir.name else: tmp_dir = None result_files = self.results2img(results, imgfile_prefix, to_label_id) return (result_files, tmp_dir)
def get_mean_std(exp_name): root_path = '/data/sls/scratch/yuangong/avbyol/egs/vggsound/exp/' three_res = [] for repeat in ['-r1', '-r2', '-r3']: cur_res = (np.loadtxt((((root_path + exp_name) + repeat) + '/result.csv'), delimiter=',') * 100) three_res.append(cur_res) three_res = np.stack(three_res) res_mean = np.mean(three_res, axis=0) res_std = np.std(three_res, axis=0) max_idx = 24 res_mean = res_mean[(max_idx, [0, 3, 6])] res_std = res_std[(max_idx, [0, 3, 6])] return (res_mean[0], res_mean[1], res_mean[2], res_std[0], res_std[1], res_std[2])
_config def pnn_rigidity(): cfg = {'learner': {'model': 'LifelongSidetuneNetwork', 'model_kwargs': {'use_baked_encoding': False, 'base_class': 'TaskonomyEncoderWithCache', 'side_class': 'FCN5ProgressiveH', 'pnn': True, 'dense': False}}}
class VisionTextDualEncoderProcessor(ProcessorMixin): attributes = ['image_processor', 'tokenizer'] image_processor_class = 'AutoImageProcessor' tokenizer_class = 'AutoTokenizer' def __init__(self, image_processor=None, tokenizer=None, **kwargs): if ('feature_extractor' in kwargs): warnings.warn('The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor` instead.', FutureWarning) feature_extractor = kwargs.pop('feature_extractor') image_processor = (image_processor if (image_processor is not None) else feature_extractor) if (image_processor is None): raise ValueError('You have to specify an image_processor.') if (tokenizer is None): raise ValueError('You have to specify a tokenizer.') super().__init__(image_processor, tokenizer) self.current_processor = self.image_processor def __call__(self, text=None, images=None, return_tensors=None, **kwargs): if ((text is None) and (images is None)): raise ValueError('You have to specify either text or images. Both cannot be none.') if (text is not None): encoding = self.tokenizer(text, return_tensors=return_tensors, **kwargs) if (images is not None): image_features = self.image_processor(images, return_tensors=return_tensors, **kwargs) if ((text is not None) and (images is not None)): encoding['pixel_values'] = image_features.pixel_values return encoding elif (text is not None): return encoding else: return BatchEncoding(data=dict(**image_features), tensor_type=return_tensors) def batch_decode(self, *args, **kwargs): return self.tokenizer.batch_decode(*args, **kwargs) def decode(self, *args, **kwargs): return self.tokenizer.decode(*args, **kwargs) def model_input_names(self): tokenizer_input_names = self.tokenizer.model_input_names image_processor_input_names = self.image_processor.model_input_names return list(dict.fromkeys((tokenizer_input_names + image_processor_input_names))) def feature_extractor_class(self): warnings.warn('`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.', FutureWarning) return self.image_processor_class def feature_extractor(self): warnings.warn('`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.', FutureWarning) return self.image_processor
def generate_regnet_parameters(w_a, w_0, w_m, d, q=8): assert ((w_a >= 0) and (w_0 > 0) and (w_m > 1) and ((w_0 % q) == 0)) ws_cont = ((np.arange(d) * w_a) + w_0) ks = np.round((np.log((ws_cont / w_0)) / np.log(w_m))) ws_all = (w_0 * np.power(w_m, ks)) ws_all = (np.round(np.divide(ws_all, q)).astype(int) * q) (ws, ds) = np.unique(ws_all, return_counts=True) (num_stages, total_stages) = (len(ws), (ks.max() + 1)) (ws, ds, ws_all, ws_cont) = (x.tolist() for x in (ws, ds, ws_all, ws_cont)) return (ws, ds, num_stages, total_stages, ws_all, ws_cont)
class MotionEncoderBiGRUCo(nn.Module): def __init__(self, input_size, hidden_size, output_size): super(MotionEncoderBiGRUCo, self).__init__() self.input_emb = nn.Linear(input_size, hidden_size) self.gru = nn.GRU(hidden_size, hidden_size, batch_first=True, bidirectional=True) self.output_net = nn.Sequential(nn.Linear((hidden_size * 2), hidden_size), nn.LayerNorm(hidden_size), nn.LeakyReLU(0.2, inplace=True), nn.Linear(hidden_size, output_size)) self.hidden_size = hidden_size self.hidden = nn.Parameter(torch.randn((2, 1, self.hidden_size), requires_grad=True)) def forward(self, inputs, m_lens): num_samples = inputs.shape[0] input_embs = self.input_emb(inputs) hidden = self.hidden.repeat(1, num_samples, 1) cap_lens = m_lens.data.tolist() emb = pack_padded_sequence(input_embs, cap_lens, batch_first=True) (gru_seq, gru_last) = self.gru(emb, hidden) gru_last = torch.cat([gru_last[0], gru_last[1]], dim=(- 1)) return self.output_net(gru_last)
def prepare_minibatch(egs_file, minibatch_size): egs = load_egs(egs_file) random.shuffle(egs) merged_egs = kaldi.chain.MergeChainEgs(egs, str(minibatch_size)) return merged_egs
class TensorboardManager(): def __init__(self, path): self.writer = tensorboardX.SummaryWriter(path) def update(self, split, step, vals): for (k, v) in vals.items(): self.writer.add_scalar(('%s_%s' % (split, k)), v, step) def close(self): self.writer.flush() self.writer.close()
class TrendBlock(Block): def __init__(self, units, thetas_dim, past_seq_len=10, future_seq_len=5, nb_harmonics=None): super(TrendBlock, self).__init__(units, thetas_dim, past_seq_len, future_seq_len, share_thetas=True) def forward(self, x): x = super(TrendBlock, self).forward(x) backcast = trend_model(self.theta_b_fc(x), self.backcast_linspace) forecast = trend_model(self.theta_f_fc(x), self.forecast_linspace) return (backcast, forecast)
class TestStochasticSwap(QiskitTestCase): def test_multiple_registers_with_layout_adjust(self): coupling = CouplingMap([[0, 1], [1, 2]]) qr_q = QuantumRegister(2, 'q') qr_a = QuantumRegister(1, 'a') cr_c = ClassicalRegister(3, 'c') circ = QuantumCircuit(qr_q, qr_a, cr_c) circ.cx(qr_q[0], qr_a[0]) circ.cx(qr_q[1], qr_a[0]) circ.measure(qr_q[0], cr_c[0]) circ.measure(qr_q[1], cr_c[1]) circ.measure(qr_a[0], cr_c[2]) dag = circuit_to_dag(circ) layout = Layout({qr_q[0]: 0, qr_q[1]: 1, qr_a[0]: 2}) pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_multiple_registers_with_good_layout(self): coupling = CouplingMap([[0, 1], [1, 2]]) qr_q = QuantumRegister(2, 'q') qr_a = QuantumRegister(1, 'a') cr_c = ClassicalRegister(3, 'c') circ = QuantumCircuit(qr_q, qr_a, cr_c) circ.cx(qr_q[0], qr_a[0]) circ.cx(qr_q[1], qr_a[0]) circ.measure(qr_q[0], cr_c[0]) circ.measure(qr_q[1], cr_c[1]) circ.measure(qr_a[0], cr_c[2]) dag = circuit_to_dag(circ) layout = Layout({qr_q[0]: 0, qr_a[0]: 1, qr_q[1]: 2}) pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_multiple_registers_with_default_layout(self): coupling = CouplingMap([[0, 1], [1, 2]]) qr_q = QuantumRegister(2, 'q') qr_a = QuantumRegister(1, 'a') cr_c = ClassicalRegister(3, 'c') circ = QuantumCircuit(qr_q, qr_a, cr_c) circ.cx(qr_q[0], qr_a[0]) circ.cx(qr_q[1], qr_a[0]) circ.measure(qr_q[0], cr_c[0]) circ.measure(qr_q[1], cr_c[1]) circ.measure(qr_a[0], cr_c[2]) dag = circuit_to_dag(circ) layout = None pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_trivial_case(self): coupling = CouplingMap([[0, 1], [0, 2]]) qr = QuantumRegister(3, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[0], qr[1]) circuit.h(qr[0]) circuit.cx(qr[0], qr[2]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_trivial_in_same_layer(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[2], qr[3]) circuit.cx(qr[0], qr[1]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_permute_wires_1(self): coupling = CouplingMap([[0, 1], [0, 2]]) qr = QuantumRegister(3, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[1], qr[2]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_permute_wires_2(self): coupling = CouplingMap([[1, 0], [1, 2]]) qr = QuantumRegister(3, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[0], qr[2]) circuit.h(qr[0]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_permute_wires_3(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[0], qr[3]) circuit.cx(qr[3], qr[0]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_permute_wires_4(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') circuit = QuantumCircuit(qr) circuit.h(qr[3]) circuit.cx(qr[3], qr[0]) dag = circuit_to_dag(circuit) expected = QuantumCircuit(qr) expected.h(qr[3]) expected.swap(qr[2], qr[3]) expected.swap(qr[0], qr[1]) expected.cx(qr[2], qr[1]) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(circuit_to_dag(expected), after) def test_permute_wires_5(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[3], qr[0]) circuit.h(qr[3]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_permute_wires_6(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') circuit = QuantumCircuit(qr) circuit.cx(qr[3], qr[0]) circuit.h(qr[3]) circuit.cx(qr[0], qr[3]) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_overoptimization_case(self): coupling = CouplingMap([[0, 2], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') cr = ClassicalRegister(4, 'c') circuit = QuantumCircuit(qr, cr) circuit.x(qr[0]) circuit.y(qr[1]) circuit.z(qr[2]) circuit.cx(qr[0], qr[1]) circuit.cx(qr[2], qr[3]) circuit.s(qr[1]) circuit.t(qr[2]) circuit.h(qr[3]) circuit.cx(qr[1], qr[2]) circuit.measure(qr[0], cr[0]) circuit.measure(qr[1], cr[1]) circuit.measure(qr[2], cr[2]) circuit.measure(qr[3], cr[3]) dag = circuit_to_dag(circuit) expected = QuantumCircuit(qr, cr) expected.z(qr[2]) expected.y(qr[1]) expected.x(qr[0]) expected.swap(qr[1], qr[2]) expected.cx(qr[0], qr[2]) expected.swap(qr[2], qr[3]) expected.cx(qr[1], qr[2]) expected.s(qr[3]) expected.t(qr[1]) expected.h(qr[2]) expected.measure(qr[0], cr[0]) expected.swap(qr[1], qr[2]) expected.cx(qr[3], qr[2]) expected.measure(qr[1], cr[3]) expected.measure(qr[3], cr[1]) expected.measure(qr[2], cr[2]) expected_dag = circuit_to_dag(expected) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(expected_dag, after) def test_already_mapped(self): coupling = CouplingMap([[1, 0], [1, 2], [2, 3], [3, 4], [3, 14], [5, 4], [6, 5], [6, 7], [6, 11], [7, 10], [8, 7], [9, 8], [9, 10], [11, 10], [12, 5], [12, 11], [12, 13], [13, 4], [13, 14], [15, 0], [15, 0], [15, 2], [15, 14]]) qr = QuantumRegister(16, 'q') cr = ClassicalRegister(16, 'c') circ = QuantumCircuit(qr, cr) circ.cx(qr[3], qr[14]) circ.cx(qr[5], qr[4]) circ.h(qr[9]) circ.cx(qr[9], qr[8]) circ.x(qr[11]) circ.cx(qr[3], qr[4]) circ.cx(qr[12], qr[11]) circ.cx(qr[13], qr[4]) for j in range(16): circ.measure(qr[j], cr[j]) dag = circuit_to_dag(circ) pass_ = StochasticSwap(coupling, None, 20, 13) after = pass_.run(dag) self.assertEqual(circuit_to_dag(circ), after) def test_map_with_layout(self): coupling = CouplingMap([[0, 1], [1, 2]]) qra = QuantumRegister(2, 'qa') qrb = QuantumRegister(1, 'qb') cr = ClassicalRegister(3, 'r') circ = QuantumCircuit(qra, qrb, cr) circ.cx(qra[0], qrb[0]) circ.measure(qra[0], cr[0]) circ.measure(qra[1], cr[1]) circ.measure(qrb[0], cr[2]) dag = circuit_to_dag(circ) layout = Layout({qra[0]: 0, qra[1]: 1, qrb[0]: 2}) pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_congestion(self): coupling = CouplingMap([[0, 1], [1, 2], [1, 3]]) qr = QuantumRegister(2, 'q') ar = QuantumRegister(2, 'a') cr = ClassicalRegister(4, 'c') circ = QuantumCircuit(qr, ar, cr) circ.cx(qr[1], ar[0]) circ.cx(qr[0], ar[1]) circ.measure(qr[0], cr[0]) circ.h(qr) circ.h(ar) circ.cx(qr[0], qr[1]) circ.cx(ar[0], ar[1]) circ.measure(qr[0], cr[0]) circ.measure(qr[1], cr[1]) circ.measure(ar[0], cr[2]) circ.measure(ar[1], cr[3]) dag = circuit_to_dag(circ) expected = QuantumCircuit(qr, ar, cr) expected.cx(qr[1], ar[0]) expected.swap(qr[0], qr[1]) expected.cx(qr[1], ar[1]) expected.h(ar[1]) expected.h(ar[0]) expected.measure(qr[1], cr[0]) expected.h(qr[0]) expected.swap(qr[1], ar[1]) expected.h(ar[1]) expected.cx(ar[0], qr[1]) expected.measure(ar[0], cr[2]) expected.swap(qr[1], ar[1]) expected.measure(ar[1], cr[3]) expected.cx(qr[1], qr[0]) expected.measure(qr[1], cr[0]) expected.measure(qr[0], cr[1]) expected_dag = circuit_to_dag(expected) layout = Layout({qr[0]: 0, qr[1]: 1, ar[0]: 2, ar[1]: 3}) pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(expected_dag, after) def test_all_single_qubit(self): coupling = CouplingMap([[0, 1], [1, 2], [1, 3]]) qr = QuantumRegister(2, 'q') ar = QuantumRegister(2, 'a') cr = ClassicalRegister(4, 'c') circ = QuantumCircuit(qr, ar, cr) circ.h(qr) circ.h(ar) circ.s(qr) circ.s(ar) circ.t(qr) circ.t(ar) circ.measure(qr[0], cr[0]) circ.measure(qr[0], cr[0]) circ.measure(qr[1], cr[1]) circ.measure(ar[0], cr[2]) circ.measure(ar[1], cr[3]) dag = circuit_to_dag(circ) layout = Layout({qr[0]: 0, qr[1]: 1, ar[0]: 2, ar[1]: 3}) pass_ = StochasticSwap(coupling, layout, 20, 13) after = pass_.run(dag) self.assertEqual(dag, after) def test_only_output_cx_and_swaps_in_coupling_map(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') cr = ClassicalRegister(4, 'c') circuit = QuantumCircuit(qr, cr) circuit.h(qr[0]) circuit.cx(qr[0], qr[1]) circuit.cx(qr[0], qr[2]) circuit.cx(qr[0], qr[3]) circuit.measure(qr, cr) dag = circuit_to_dag(circuit) layout = Layout({qr[0]: 0, qr[1]: 1, qr[2]: 2, qr[3]: 3}) pass_ = StochasticSwap(coupling, layout, 20, 5) after = pass_.run(dag) valid_couplings = [set([layout[a], layout[b]]) for (a, b) in coupling.get_edges()] for _2q_gate in after.twoQ_gates(): self.assertIn(set(_2q_gate.qargs), valid_couplings) def test_len_coupling_vs_dag(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3], [3, 4]]) qr = QuantumRegister(4, 'q') cr = ClassicalRegister(4, 'c') circuit = QuantumCircuit(qr, cr) circuit.h(qr[0]) circuit.cx(qr[0], qr[1]) circuit.cx(qr[0], qr[2]) circuit.cx(qr[0], qr[3]) circuit.measure(qr, cr) dag = circuit_to_dag(circuit) pass_ = StochasticSwap(coupling) with self.assertRaises(TranspilerError): _ = pass_.run(dag) def test_len_layout_vs_dag(self): coupling = CouplingMap([[0, 1], [1, 2], [2, 3]]) qr = QuantumRegister(4, 'q') cr = ClassicalRegister(4, 'c') circuit = QuantumCircuit(qr, cr) circuit.h(qr[0]) circuit.cx(qr[0], qr[1]) circuit.cx(qr[0], qr[2]) circuit.cx(qr[0], qr[3]) circuit.measure(qr, cr) dag = circuit_to_dag(circuit) layout = Layout({qr[0]: 0, qr[1]: 1, qr[2]: 2}) pass_ = StochasticSwap(coupling, layout) with self.assertRaises(TranspilerError): _ = pass_.run(dag)
class ImageMirror(ImagePreprocessing): def __init__(self, bigdl_type='float'): super(ImageMirror, self).__init__(bigdl_type)
def get_available_gpus(session_config=None): if (session_config is None): session_config = get_session()._config from tensorflow.python.client import device_lib local_device_protos = device_lib.list_local_devices(session_config) return [x.name for x in local_device_protos if (x.device_type == 'GPU')]
class LayoutLMv3Processor(ProcessorMixin): attributes = ['image_processor', 'tokenizer'] image_processor_class = 'LayoutLMv3ImageProcessor' tokenizer_class = ('LayoutLMv3Tokenizer', 'LayoutLMv3TokenizerFast') def __init__(self, image_processor=None, tokenizer=None, **kwargs): if ('feature_extractor' in kwargs): warnings.warn('The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor` instead.', FutureWarning) feature_extractor = kwargs.pop('feature_extractor') image_processor = (image_processor if (image_processor is not None) else feature_extractor) if (image_processor is None): raise ValueError('You need to specify an `image_processor`.') if (tokenizer is None): raise ValueError('You need to specify a `tokenizer`.') super().__init__(image_processor, tokenizer) def __call__(self, images, text: Union[(TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput])]=None, text_pair: Optional[Union[(PreTokenizedInput, List[PreTokenizedInput])]]=None, boxes: Union[(List[List[int]], List[List[List[int]]])]=None, word_labels: Optional[Union[(List[int], List[List[int]])]]=None, add_special_tokens: bool=True, padding: Union[(bool, str, PaddingStrategy)]=False, truncation: Union[(bool, str, TruncationStrategy)]=None, max_length: Optional[int]=None, stride: int=0, pad_to_multiple_of: Optional[int]=None, return_token_type_ids: Optional[bool]=None, return_attention_mask: Optional[bool]=None, return_overflowing_tokens: bool=False, return_special_tokens_mask: bool=False, return_offsets_mapping: bool=False, return_length: bool=False, verbose: bool=True, return_tensors: Optional[Union[(str, TensorType)]]=None, **kwargs) -> BatchEncoding: if (self.image_processor.apply_ocr and (boxes is not None)): raise ValueError('You cannot provide bounding boxes if you initialized the image processor with apply_ocr set to True.') if (self.image_processor.apply_ocr and (word_labels is not None)): raise ValueError('You cannot provide word labels if you initialized the image processor with apply_ocr set to True.') features = self.image_processor(images=images, return_tensors=return_tensors) if ((text is not None) and self.image_processor.apply_ocr and (text_pair is None)): if isinstance(text, str): text = [text] text_pair = features['words'] encoded_inputs = self.tokenizer(text=(text if (text is not None) else features['words']), text_pair=(text_pair if (text_pair is not None) else None), boxes=(boxes if (boxes is not None) else features['boxes']), word_labels=word_labels, add_special_tokens=add_special_tokens, padding=padding, truncation=truncation, max_length=max_length, stride=stride, pad_to_multiple_of=pad_to_multiple_of, return_token_type_ids=return_token_type_ids, return_attention_mask=return_attention_mask, return_overflowing_tokens=return_overflowing_tokens, return_special_tokens_mask=return_special_tokens_mask, return_offsets_mapping=return_offsets_mapping, return_length=return_length, verbose=verbose, return_tensors=return_tensors, **kwargs) images = features.pop('pixel_values') if (return_overflowing_tokens is True): images = self.get_overflowing_images(images, encoded_inputs['overflow_to_sample_mapping']) encoded_inputs['pixel_values'] = images return encoded_inputs def get_overflowing_images(self, images, overflow_to_sample_mapping): images_with_overflow = [] for sample_idx in overflow_to_sample_mapping: images_with_overflow.append(images[sample_idx]) if (len(images_with_overflow) != len(overflow_to_sample_mapping)): raise ValueError(f'Expected length of images to be the same as the length of `overflow_to_sample_mapping`, but got {len(images_with_overflow)} and {len(overflow_to_sample_mapping)}') return images_with_overflow def batch_decode(self, *args, **kwargs): return self.tokenizer.batch_decode(*args, **kwargs) def decode(self, *args, **kwargs): return self.tokenizer.decode(*args, **kwargs) def model_input_names(self): return ['input_ids', 'bbox', 'attention_mask', 'pixel_values'] def feature_extractor_class(self): warnings.warn('`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.', FutureWarning) return self.image_processor_class def feature_extractor(self): warnings.warn('`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.', FutureWarning) return self.image_processor
class ROIAlign3d(nn.Module): def __init__(self, output_size, spatial_scale, sampling_ratio): super(ROIAlign3d, self).__init__() self.output_size = output_size self.spatial_scale = spatial_scale self.sampling_ratio = sampling_ratio def forward(self, input, rois): return roi_align_3d(input, rois, self.output_size, self.spatial_scale, self.sampling_ratio) def __repr__(self): tmpstr = (self.__class__.__name__ + '(') tmpstr += ('output_size=' + str(self.output_size)) tmpstr += (', spatial_scale=' + str(self.spatial_scale)) tmpstr += (', sampling_ratio=' + str(self.sampling_ratio)) tmpstr += ')' return tmpstr
class StableDiffusionOnnxPipeline(metaclass=DummyObject): _backends = ['torch', 'transformers', 'onnx'] def __init__(self, *args, **kwargs): requires_backends(self, ['torch', 'transformers', 'onnx']) def from_config(cls, *args, **kwargs): requires_backends(cls, ['torch', 'transformers', 'onnx']) def from_pretrained(cls, *args, **kwargs): requires_backends(cls, ['torch', 'transformers', 'onnx'])
def preprocess_org_min_date(path=DATA_PATH, file=DATA_FILE, path_proc=DATA_PATH_PROCESSED, min_item_support=MIN_ITEM_SUPPORT, min_session_length=MIN_SESSION_LENGTH, min_date=MIN_DATE): data = load_data((path + file)) data = filter_data(data, min_item_support, min_session_length) data = filter_min_date(data, min_date) split_data_org(data, (path_proc + file))
class DropdownSelectionWidget(Widget): def __init__(self, options, value, description, parameter_dict, **kwargs): super().__init__(**kwargs) self.drop_down = widgets.Dropdown(options=options, value=value, description=description, disabled=False) self.parameter_dict = parameter_dict self.drop_down.observe(self.on_change, names='value') self.parameter_accordion = ParameterSelectionWidget(self.parameter_dict[self.drop_down.value]) self.widget = widgets.VBox([self.drop_down, self.parameter_accordion.parameter_accordion]) def get_value(self): return self.drop_down.value def on_change(self, change): if ((change['type'] == 'change') and (change['name'] == 'value')): self.parameter_accordion.update(self.parameter_dict[change['new']]) def get_parameter_values(self): return self.parameter_accordion.get_currently_selected_parameters()
def apk(actual, predicted, k=10): if (len(predicted) > k): predicted = predicted[:k] score = 0.0 num_hits = 0.0 for (i, p) in enumerate(predicted): if ((p in actual) and (p not in predicted[:i])): num_hits += 1.0 score += (num_hits / (i + 1.0)) if (not actual): return 0.0 return (score / min(len(actual), k))
.slow def test_factorized_antisymmetry_can_be_evaluated(): (key, init_pos, slog_psis) = _make_factorized_antisymmetries() [_jit_eval_model_and_verify_output_shape(key, init_pos, slog_psi) for slog_psi in slog_psis]
class HeteroDotProductPredictor(nn.Module): def forward(self, graph, h, etype): with graph.local_scope(): graph.ndata['h'] = h graph.apply_edges(fn.u_dot_v('h', 'h', 'score'), etype=etype) return graph.edges[etype].data['score']
def get_cosine_schedule_with_warmup(optimizer: Optimizer, num_warmup_steps: int, num_training_steps: int, num_cycles: float=0.5, last_epoch: int=(- 1), min_lr_ratio: float=0.0): lr_lambda = partial(_get_cosine_schedule_with_warmup_lr_lambda, num_warmup_steps=num_warmup_steps, num_training_steps=num_training_steps, num_cycles=num_cycles, min_lr_ratio=min_lr_ratio) return LambdaLR(optimizer, lr_lambda, last_epoch)
class XNet_sb(nn.Module): def __init__(self, in_channels, num_classes): super(XNet_sb, self).__init__() (l1c, l2c, l3c, l4c, l5c) = (64, 128, 256, 512, 1024) self.b1_1_1 = nn.Sequential(conv3x3(in_channels, l1c), conv3x3(l1c, l1c), BasicBlock(l1c, l1c)) self.b1_1_2_down = down_conv(l1c, l2c) self.b1_1_3 = DoubleBasicBlock((l1c + l1c), l1c, nn.Sequential(conv1x1(in_planes=(l1c + l1c), out_planes=l1c), BatchNorm2d(l1c, momentum=BN_MOMENTUM))) self.b1_1_4 = nn.Conv2d(l1c, num_classes, kernel_size=1, stride=1, padding=0) self.b1_2_1 = DoubleBasicBlock(l2c, l2c) self.b1_2_2_down = down_conv(l2c, l3c) self.b1_2_3 = DoubleBasicBlock((l2c + l2c), l2c, nn.Sequential(conv1x1(in_planes=(l2c + l2c), out_planes=l2c), BatchNorm2d(l2c, momentum=BN_MOMENTUM))) self.b1_2_4_up = up_conv(l2c, l1c) self.b1_3_1 = DoubleBasicBlock(l3c, l3c) self.b1_3_2_down = down_conv(l3c, l4c) self.b1_3_3 = DoubleBasicBlock((l3c + l3c), l3c, nn.Sequential(conv1x1(in_planes=(l3c + l3c), out_planes=l3c), BatchNorm2d(l3c, momentum=BN_MOMENTUM))) self.b1_3_4_up = up_conv(l3c, l2c) self.b1_4_1 = DoubleBasicBlock(l4c, l4c) self.b1_4_2_down = down_conv(l4c, l5c) self.b1_4_2 = DoubleBasicBlock(l4c, l4c) self.b1_4_5 = DoubleBasicBlock(l4c, l4c) self.b1_4_6 = DoubleBasicBlock((l4c + l4c), l4c, nn.Sequential(conv1x1(in_planes=(l4c + l4c), out_planes=l4c), BatchNorm2d(l4c, momentum=BN_MOMENTUM))) self.b1_4_7_up = up_conv(l4c, l3c) self.b1_5_1 = DoubleBasicBlock(l5c, l5c) self.b1_5_4 = DoubleBasicBlock(l5c, l5c) self.b1_5_5_up = up_conv(l5c, l4c) for m in self.modules(): if isinstance(m, nn.Conv2d): nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') if (m.bias is not None): nn.init.constant_(m.bias, 0) elif isinstance(m, nn.BatchNorm2d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) elif isinstance(m, nn.Linear): nn.init.normal_(m.weight, std=0.001) if (m.bias is not None): nn.init.constant_(m.bias, 0) def forward(self, input1): x1_1 = self.b1_1_1(input1) x1_2 = self.b1_1_2_down(x1_1) x1_2 = self.b1_2_1(x1_2) x1_3 = self.b1_2_2_down(x1_2) x1_3 = self.b1_3_1(x1_3) x1_4_1 = self.b1_3_2_down(x1_3) x1_4_1 = self.b1_4_1(x1_4_1) x1_4_2 = self.b1_4_2(x1_4_1) x1_4_2 = self.b1_4_5(x1_4_2) x1_5_1 = self.b1_4_2_down(x1_4_1) x1_5_1 = self.b1_5_1(x1_5_1) x1_5_1 = self.b1_5_4(x1_5_1) x1_5_1 = self.b1_5_5_up(x1_5_1) x1_4_2 = torch.cat((x1_4_2, x1_5_1), dim=1) x1_4_2 = self.b1_4_6(x1_4_2) x1_4_2 = self.b1_4_7_up(x1_4_2) x1_3 = torch.cat((x1_3, x1_4_2), dim=1) x1_3 = self.b1_3_3(x1_3) x1_3 = self.b1_3_4_up(x1_3) x1_2 = torch.cat((x1_2, x1_3), dim=1) x1_2 = self.b1_2_3(x1_2) x1_2 = self.b1_2_4_up(x1_2) x1_1 = torch.cat((x1_1, x1_2), dim=1) x1_1 = self.b1_1_3(x1_1) x1_1 = self.b1_1_4(x1_1) return x1_1
def dump_fold_into_csv_CUB(lsamples, outpath, tag): msg = "'tag' must be an integer. Found {}.".format(tag) assert isinstance(tag, int), msg msg = "'tag' = {} is unknown. Please see constants.samples_tags = {}.".format(tag, constants.samples_tags) assert (tag in constants.samples_tags), msg assert (tag == constants.L) with open(outpath, 'w') as fcsv: filewriter = csv.writer(fcsv, delimiter=',', quotechar='|', quoting=csv.QUOTE_MINIMAL) for (img_path, mask_path, img_label, idcnt) in lsamples: filewriter.writerow([str(int(idcnt)), img_path, mask_path, img_label, tag])
def cifar_model_large(conv_layer, linear_layer, init_type, **kwargs): assert (init_type == 'kaiming_normal'), 'only supporting kaiming_normal init' model = nn.Sequential(conv_layer(3, 32, 3, stride=1, padding=1), nn.ReLU(), conv_layer(32, 32, 4, stride=2, padding=1), nn.ReLU(), conv_layer(32, 64, 3, stride=1, padding=1), nn.ReLU(), conv_layer(64, 64, 4, stride=2, padding=1), nn.ReLU(), Flatten(), linear_layer(((64 * 8) * 8), 512), nn.ReLU(), linear_layer(512, 512), nn.ReLU(), linear_layer(512, 10)) for m in model.modules(): if isinstance(m, nn.Conv2d): n = ((m.kernel_size[0] * m.kernel_size[1]) * m.out_channels) m.weight.data.normal_(0, math.sqrt((2.0 / n))) m.bias.data.zero_() return model
def init(model_s, model_t, init_modules, criterion, train_loader, logger, opt): model_t.eval() model_s.eval() init_modules.train() if torch.cuda.is_available(): model_s.cuda() model_t.cuda() init_modules.cuda() cudnn.benchmark = True if ((opt.model_s in ['resnet8', 'resnet14', 'resnet20', 'resnet32', 'resnet44', 'resnet56', 'resnet110', 'resnet8x4', 'resnet32x4', 'wrn_16_1', 'wrn_16_2', 'wrn_40_1', 'wrn_40_2']) and (opt.distill == 'factor')): lr = 0.01 else: lr = opt.learning_rate optimizer = optim.SGD(init_modules.parameters(), lr=lr, momentum=opt.momentum, weight_decay=opt.weight_decay) batch_time = AverageMeter() data_time = AverageMeter() losses = AverageMeter() for epoch in range(1, (opt.init_epochs + 1)): batch_time.reset() data_time.reset() losses.reset() end = time.time() for (idx, data) in enumerate(train_loader): if (opt.distill in ['crd']): (input, target, index, contrast_idx) = data else: (input, target, index) = data data_time.update((time.time() - end)) input = input.float() if torch.cuda.is_available(): input = input.cuda() target = target.cuda() index = index.cuda() if (opt.distill in ['crd']): contrast_idx = contrast_idx.cuda() preact = (opt.distill == 'abound') (feat_s, _) = model_s(input, is_feat=True, preact=preact) with torch.no_grad(): (feat_t, _) = model_t(input, is_feat=True, preact=preact) feat_t = [f.detach() for f in feat_t] if (opt.distill == 'abound'): g_s = init_modules[0](feat_s[1:(- 1)]) g_t = feat_t[1:(- 1)] loss_group = criterion(g_s, g_t) loss = sum(loss_group) elif (opt.distill == 'factor'): f_t = feat_t[(- 2)] (_, f_t_rec) = init_modules[0](f_t) loss = criterion(f_t_rec, f_t) elif (opt.distill == 'fsp'): loss_group = criterion(feat_s[:(- 1)], feat_t[:(- 1)]) loss = sum(loss_group) else: raise NotImplemented('Not supported in init training: {}'.format(opt.distill)) losses.update(loss.item(), input.size(0)) optimizer.zero_grad() loss.backward() optimizer.step() batch_time.update((time.time() - end)) end = time.time() logger.add_scalar('init_train_loss', losses.avg, epoch) print('Epoch: [{0}/{1}]\tTime {batch_time.val:.3f} ({batch_time.avg:.3f})\tlosses: {losses.val:.3f} ({losses.avg:.3f})'.format(epoch, opt.init_epochs, batch_time=batch_time, losses=losses)) sys.stdout.flush()
class Bottleneck(nn.Module): expansion = 4 def __init__(self, inplanes, planes, stride=1, downsample=None, dilation=(1, 1), residual=True, BatchNorm=None): super(Bottleneck, self).__init__() self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) self.bn1 = BatchNorm(planes) self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=dilation[1], bias=False, dilation=dilation[1]) self.bn2 = BatchNorm(planes) self.conv3 = nn.Conv2d(planes, (planes * 4), kernel_size=1, bias=False) self.bn3 = BatchNorm((planes * 4)) self.relu = nn.ReLU(inplace=True) self.downsample = downsample self.stride = stride def forward(self, x): residual = x out = self.conv1(x) out = self.bn1(out) out = self.relu(out) out = self.conv2(out) out = self.bn2(out) out = self.relu(out) out = self.conv3(out) out = self.bn3(out) if (self.downsample is not None): residual = self.downsample(x) out += residual out = self.relu(out) return out
class Net(torch.nn.Module): def __init__(self, train_dataset): super(Net, self).__init__() self.conv1 = GATConv(train_dataset.num_features, 256, heads=4) self.lin1 = torch.nn.Linear(train_dataset.num_features, (4 * 256)) self.conv2 = GATConv((4 * 256), 256, heads=4) self.lin2 = torch.nn.Linear((4 * 256), (4 * 256)) self.conv3 = GATConv((4 * 256), train_dataset.num_classes, heads=6, concat=False) self.lin3 = torch.nn.Linear((4 * 256), train_dataset.num_classes) def forward(self, x, edge_index): x = F.elu((self.conv1(x, edge_index) + self.lin1(x))) x = F.elu((self.conv2(x, edge_index) + self.lin2(x))) x = (self.conv3(x, edge_index) + self.lin3(x)) return x
def test(args, model, device, test_loader, logger): model.eval() with torch.no_grad(): for (data, target) in test_loader: start = time() (data, target) = (data.to(device), target.to(device)) predictions = model(data) loss = F.cross_entropy(predictions, target) pred = predictions.max(1, keepdim=True)[1] accuracy = pred.eq(target.view_as(pred)).double().mean() stats = {'val.loss': loss.item(), 'val.accuracy': accuracy.item()} if logger: logger.update_average(stats) if (logger.avg_count['val.loss'] > 3): logger.update_average({'val.time': (time() - start)}) else: print(stats) if logger: logger.print(prefix='val')
class Adjective_Rate(object): def __init__(self, sentence_objs): self.sentence_objs = sentence_objs def handle(self): (tot_num_adjs, tot_num_words) = (0, 0) for so in self.sentence_objs: tot_num_adjs += so.pos_tag_counter.get_pos_tag_count(ADJECTIVE) tot_num_words += so.num_words() return (tot_num_adjs / tot_num_words)
def clear_dir(directory): if (not os.path.isdir(directory)): raise Exception(('%s is not a directory' % directory)) if (type(directory) != str): raise Exception(('string type required for directory: %s' % directory)) if (directory in ['..', '.', '', '/', './', '../', '*']): raise Exception('trying to delete current directory, probably bad idea?!') for f in os.listdir(directory): path = os.path.join(directory, f) try: if os.path.isfile(path): os.remove(path) elif os.path.isdir(path): shutil.rmtree(path) except Exception as e: print(e)
def get_config(num_targets): if ((num_targets == 0) or (not isinstance(num_targets, int))): raise ValueError(f'num_targets is {num_targets}, but must be a positive integer') screen = sprite.Sprite(x=0.5, y=0.5, shape='square', scale=2.0, c0=0.6, c1=0.7, c2=0.7) target_factor_distrib = distribs.Product([distribs.Continuous('c0', 0.0, 1.0)], shape='circle', scale=0.085, c1=1.0, c2=1.0) cover_factors = dict(mass=0.0, shape='circle', scale=0.1, c0=0.0, c1=0.0, c2=0.5, opacity=0) def state_initializer(): sprite_positions = (0.5 + (0.35 * _get_polygon(num_targets, 0.7))) target_factors = [target_factor_distrib.sample() for _ in range(num_targets)] targets = [sprite.Sprite(x=pos[0], y=pos[1], **factors) for (pos, factors) in zip(sprite_positions, target_factors)] covers = [sprite.Sprite(x=pos[0], y=pos[1], **cover_factors) for pos in sprite_positions] for (i, s) in enumerate(covers): if (i == 0): s.metadata = {'prey': True} else: s.metadata = {'prey': False} cue_factors = copy.deepcopy(target_factors[0]) cue_factors['scale'] = (0.7 * target_factors[0]['scale']) cue = sprite.Sprite(x=0.5, y=0.501, opacity=0, mass=np.inf, **cue_factors) agent = sprite.Sprite(x=0.5, y=0.5, shape='circle', scale=0.1, c0=0.4, c1=0.0, c2=1.0, mass=np.inf) annulus_verts = shapes.annulus_vertices(0.34, 0.36) annulus = sprite.Sprite(x=0.5, y=0.5, shape=annulus_verts, scale=1.0, c0=0.0, c1=0.0, c2=0.3) state = collections.OrderedDict([('annulus', [annulus]), ('targets', targets), ('covers', covers), ('agent', [agent]), ('cue', [cue]), ('screen', [screen])]) return state drag = (physics_lib.Drag(coeff_friction=0.25), ['agent', 'cue']) tether_covers = physics_lib.TetherZippedLayers(('targets', 'covers'), anchor=np.array([0.5, 0.5])) physics = physics_lib.Physics(drag, updates_per_env_step=1, corrective_physics=[tether_covers]) contact_task = tasks.ContactReward(reward_fn=(lambda _, s: (1 if s.metadata['prey'] else (- 1))), layers_0='agent', layers_1='covers') def _should_reset(state, meta_state): should_reset = ((state['covers'][0].opacity == 0) and (meta_state['phase'] == 'response')) return should_reset reset_task = tasks.Reset(condition=_should_reset, steps_after_condition=15) task = tasks.CompositeTask(contact_task, reset_task, timeout_steps=800) action_space = action_spaces.Joystick(scaling_factor=0.01, action_layers=['agent', 'cue']) _polygon_modifier = observers.polygon_modifiers.FirstPersonAgent(agent_layer='agent') observer = observers.PILRenderer(image_size=(64, 64), anti_aliasing=1, color_to_rgb='hsv_to_rgb', polygon_modifier=_polygon_modifier) def _make_opaque(s): s.opacity = 255 def _make_transparent(s): s.opacity = 0 screen_phase = gr.Phase(duration=1, name='screen') disappear_screen = gr.ModifySprites('screen', _make_transparent) visible_phase = gr.Phase(one_time_rules=disappear_screen, duration=2, name='visible') def _move(s): s.velocity = np.random.uniform((- 0.25), 0.25, size=(2,)) cover_targets = gr.ModifySprites('covers', _make_opaque) begin_motion = BeginMotion(angle_vel_range=(0.1, 0.3)) motion_phase = gr.Phase(one_time_rules=[cover_targets, begin_motion], duration=100, name='motion') def _stop(s): s.angle_vel = 0.0 s.velocity = np.zeros(2) def _unglue(s): s.mass = 1.0 appear_cue = gr.ModifySprites('cue', _make_opaque) stop_targets = gr.ModifySprites(('targets', 'covers'), _stop) unglue_agent = gr.ModifySprites(('agent', 'cue'), _unglue) make_targets_discoverable = gr.ModifyOnContact(layers_0='agent', layers_1='covers', modifier_1=_make_transparent) response_phase = gr.Phase(one_time_rules=[appear_cue, stop_targets, unglue_agent], continual_rules=make_targets_discoverable, name='response') phase_sequence = gr.PhaseSequence(screen_phase, visible_phase, motion_phase, response_phase, meta_state_phase_name_key='phase') config = {'state_initializer': state_initializer, 'physics': physics, 'task': task, 'action_space': action_space, 'observers': {'image': observer}, 'game_rules': (phase_sequence,), 'meta_state_initializer': (lambda : {'phase': ''})} return config
def plot_alignment_to_numpy(alignment, info=None): global MATPLOTLIB_FLAG if (not MATPLOTLIB_FLAG): import matplotlib matplotlib.use('Agg') MATPLOTLIB_FLAG = True mpl_logger = logging.getLogger('matplotlib') mpl_logger.setLevel(logging.WARNING) import matplotlib.pylab as plt import numpy as np (fig, ax) = plt.subplots(figsize=(6, 4)) im = ax.imshow(alignment, aspect='auto', origin='lower', interpolation='none') fig.colorbar(im, ax=ax) xlabel = 'Decoder timestep' if (info is not None): xlabel += ('\n\n' + info) plt.xlabel(xlabel) plt.ylabel('Encoder timestep') plt.tight_layout() fig.canvas.draw() data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='') data = data.reshape((fig.canvas.get_width_height()[::(- 1)] + (3,))) plt.close() return data
def parse_dict_args(**kwargs): def to_cmdline_kwarg(key, value): if (len(key) == 1): key = '-{}'.format(key) else: key = '--{}'.format(re.sub('_', '-', key)) value = str(value) return (key, value) kwargs_pairs = (to_cmdline_kwarg(key, value) for (key, value) in kwargs.items()) cmdline_args = list(sum(kwargs_pairs, ())) LOG.info('Using these command line args: %s', ' '.join(cmdline_args)) return create_parser().parse_args(cmdline_args)