Datasets:

lambdaofgod

/

pwc_repositories_with_dependencies

like 0

CapsuleEndoscope/VirtualCapsuleEndoscopy

['depth estimation', 'visual localization']

['VR-Caps: A Virtual Environment for Capsule Endoscopy']

Tasks/Disease Classification/classification_pytorch.py Tasks/Pose and Depth Estimation/poseError.py train_model save_model prepare_prediction_file test_model prepare_data prepare_submission_file prepare_model check_model_graph plot_confusion_matrix run_train translation_error rotation_error rotate_origin_only compute_ATE compute_RPE all_folds join print DataLoader data_root dataset val_fold len deepcopy add_scalars list save_model add_scalar print train step zero_grad tqdm eval num_epochs to double range enumerate state_dict resnet152 to in_features Linear train_model load prepare_data SGD prepare_model parameters eval ReduceLROnPlateau load_state_dict input CrossEntropyLoss join format save print Variable rand add_graph prepare_model summary to load print cpu prepare_data classification_report confusion_matrix close prepare_model set_printoptions eval load_state_dict classes plot_confusion_matrix input to load print prepare_data exit prepare_model eval load_state_dict classes input DataFrame head join print to_csv DataFrame column_stack arange add_figure tick_params xticks max yticks list ylabel colorbar imshow savefig range gcf format product astype print text xlabel len cos sin asarray mean sqrt append sum std range asarray inv translation_error mean rotation_error append std range len sqrt max min arccos

VR-Caps: A Virtual Environment for Capsule Endoscopy ===== <p align="center"> <img src='img/logo_turan.png'> </p> <p align="center"> <img src='img/Fig1.png' width=512/> </p> ## Overview We introduce a virtual active capsule endoscopy environment developed in Unity that provides a simulation platform to generate synthetic data as well as a test bed to develop and test algorithms. Using that environment, we perform various evaluations for common robotics and computer vision tasks of active capsule endoscopy such as classification, pose and depth estimation, area coverage, autonomous navigation, learning control of endoscopic capsule robot with magnetic field inside GI-tract organs, super-resolution, etc. The demonstration of our virtual environment is available on [YouTube](https://www.youtube.com/watch?v=UQ2u3CIUciA).

Carco-git/CW_Attack_on_MNIST

['adversarial attack']

['Towards Evaluating the Robustness of Neural Networks']

MNIST_Model.py MNIST_Model

# CW_Attack_on_MNIST Implemention of cw attack on pytorch with corresponding MNIST model ## MNIST model Based on [Towards Evaluating the Robustness of Neural Networks](https://arxiv.org/abs/1608.04644) TABLE1 Consist of four convolution layer, two pooling layers, tow FC layers and ReLU. Notice: softmax shouldn't be put into model. What we need is the result from FC.That's why I use crossEntropyLoss as loss function. ## Carlini and Wagner L2 L0,L2,L-inf be used for generating adversarial examples. In this repo, I choose L2 norm as the objective function. ``` def loss1_func(w,x,d,c):

CardiacModelling/fickleheart-method-tutorials

['gaussian processes']

['Considering discrepancy when calibrating a mechanistic electrophysiology model']

ion-channel-models/compare-posteriors-gpcovs.py ion-channel-models/plot-residual-noise.py ion-channel-models/fit-gp-v.py ion-channel-models/method/sparse_gp_custom_likelihood.py ion-channel-models/compare-pp.py ion-channel-models/protocol-time-series/staircase_to_zoom.py ion-channel-models/method/sparse_gp_custom_likelihood_new.py ion-channel-models/compare-evidence.py ion-channel-models/compare-posteriors-traces.py ion-channel-models/predict.py ion-channel-models/fit-gp-tv.py ion-channel-models/compare.py action-potential-models/method/model.py ion-channel-models/mcmc-gp-v.py ion-channel-models/mmt-model-files/model_C.py ion-channel-models/generate-data.py ion-channel-models/mcmc-gp.py ion-channel-models/mcmc-arma.py action-potential-models/fit.py ion-channel-models/compare-pp-gpcovs.py ion-channel-models/method/model.py action-potential-models/method/parametertransform.py action-potential-models/posterior.py ion-channel-models/compare-mean-error.py ion-channel-models/fit-gp-ov.py action-potential-models/method/default.py action-potential-models/test-model.py ion-channel-models/posterior-gp.py ion-channel-models/plot-iid-sigma.py ion-channel-models/protocol-time-series/ap_to_zoom.py ion-channel-models/posterior-arma.py ion-channel-models/method/priors.py action-potential-models/method/protocol.py ion-channel-models/method/parametertransform.py ion-channel-models/posterior-gp-tv.py ion-channel-models/protocol-time-series/sinewave_to_zoom.py ion-channel-models/fit-gp.py ion-channel-models/mcmc-gp-ov.py ion-channel-models/compare-error-mean.py ion-channel-models/posterior-gp-ov.py ion-channel-models/mcmc.py action-potential-models/generate-data.py ion-channel-models/mcmc-arma-invertible.py ion-channel-models/posterior.py action-potential-models/plot-differences.py ion-channel-models/mmt-model-files/model_A.py action-potential-models/mcmc.py action-potential-models/predict.py ion-channel-models/mmt-model-files/model_B.py setup.py ion-channel-models/method/armax_ode_tsa_likelihood.py ion-channel-models/compare-posteriors.py ion-channel-models/mcmc-gp-tv.py ion-channel-models/fit.py ion-channel-models/test-models.py ion-channel-models/compare-error-mean-tex.py f Model Timeout simulate_aps log_transform_from_model_param donothing log_transform_to_model_param hergblock_simulate row rmse row rmse row row rmse autocorrelation rmse rmse rmse rmse rmse DiscrepancyLogLikelihood Model Timeout log_transform_from_model_param donothing log_transform_to_model_param ModelALogPrior HalfNormalLogPrior InverseGammaLogPrior ArmaNormalLogPrior MultiPriori ModelBLogPrior ArmaNormalCentredLogPrior _create_theano_conditional_graph _create_theano_likelihood_graph_voltage stabilize _create_theano_likelihood_graph RbfKernel _create_theano_conditional_graph_voltage GpCovariance DiscrepancyLogLikelihood stabilize Matern12 _create_theano_likelihood_graph_voltage RbfKernel _create_theano_conditional_graph_voltage GpCovariance DiscrepancyLogLikelihood Matern32 get rhs load_model Number set_rhs Simulation set_protocol pre blocktrain Multiply log copy exp copy copy simulate zeros asarray enumerate strip min array enumerate text add_patch Rectangle copy abs max split mean set_xlim legend acorr stabilize pi log clip transpose cov_func solve_lower dscalar sum ones_like inf dvector square RbfKernel as_tensor_variable dot cholesky eye diag stabilize pi log clip transpose cov_func solve_lower dscalar sum ones_like inf dvector square RbfKernel as_tensor_variable dot cholesky eye diag ones_like inf dvector stabilize transpose cov_func square as_tensor_variable dot RbfKernel cholesky solve_lower eye solve_upper dscalar sum clip ones_like inf dvector stabilize transpose cov_func square as_tensor_variable dot RbfKernel cholesky solve_lower eye solve_upper dscalar sum clip Matern32 concatenate Matern12 reshape Matern32 concatenate Matern12 reshape

# Model calibration with discrepancy This repo contains the code for reproducing the results in the examples in the paper "*Considering discrepancy when calibrating a mechanistic electrophysiology model*" by Lei, Ghosh, Whittaker, Aboelkassem, Beattie, Cantwell, Delhaas, Houston, Novaes, Panfilov, Pathmanathan, Riabiz, dos Santos, Walmsley, Worden, Mirams, and Wilkinson. [doi:10.1098/rsta.2019.0349](https://doi.org/10.1098/rsta.2019.0349). ### Requirements The code requires Python (3.5+) and the following dependencies: [PINTS](https://github.com/pints-team/pints#installing-pints), [Myokit](http://myokit.org/install/), [Theano](http://deeplearning.net/software/theano/install.html), [StatsModels](https://www.statsmodels.org/stable/install.html), [Joblib](https://joblib.readthedocs.io/en/latest/installing.html).

Cartus/AGGCN

['relation extraction']

['Attention Guided Graph Convolutional Networks for Relation Extraction']

semeval/utils/helper.py utils/vocab.py PubMed/Binary/train.py semeval/model/trainer.py semeval/prepare_vocab.py PubMed/Tenary/utils/nary_scorer.py PubMed/Tenary/eval.py prepare_vocab.py semeval/data/loader.py PubMed/Tenary/utils/vocab.py PubMed/Binary/model/trainer.py train.py PubMed/Tenary/prepare_vocab.py semeval/eval.py semeval/model/aggcn.py utils/helper.py semeval/train.py PubMed/Binary/model/graph.py PubMed/Binary/utils/nary_scorer.py PubMed/Tenary/model/graph.py PubMed/Binary/eval.py PubMed/Tenary/data/loader.py semeval/utils/constant.py PubMed/Binary/utils/constant.py PubMed/Tenary/train.py PubMed/Tenary/utils/constant.py PubMed/Tenary/utils/helper.py semeval/utils/scorer.py PubMed/Binary/utils/vocab.py PubMed/Binary/data/loader.py PubMed/Binary/prepare_vocab.py model/trainer.py utils/scorer.py data/loader.py PubMed/Tenary/model/trainer.py eval.py PubMed/Binary/utils/helper.py semeval/utils/vocab.py model/tree.py PubMed/Tenary/model/aggcn.py PubMed/Binary/utils/torch_utils.py semeval/model/tree.py model/aggcn.py semeval/utils/torch_utils.py utils/constant.py utils/torch_utils.py PubMed/Binary/model/aggcn.py PubMed/Tenary/utils/torch_utils.py entity_masks count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention attention MultiGraphConvLayer AGGCN rnn_zero_state GraphConvLayer unpack_batch GCNTrainer Trainer tree_to_adj head_to_tree tree_to_dist Tree count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids read_file sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention MultiGraphConvLayer DCGCN attention GraphConvLayer tree_to_dist head_to_graph tree_to_adj Tree head_to_tree unpack_batch GCNTrainer Trainer FileLogger save_config print_config check_files ensure_dir check_dir load_config score load change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids read_file sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention MultiGraphConvLayer DCGCN attention GraphConvLayer tree_to_dist head_to_graph tree_to_adj Tree head_to_tree unpack_batch GCNTrainer Trainer FileLogger save_config print_config check_files ensure_dir check_dir load_config score load change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab entity_masks count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids sort_all word_dropout GCNClassifier pool GCNRelationModel batched_index_select clones MultiHeadAttention attention MultiGraphConvLayer AGGCN rnn_zero_state GraphConvLayer unpack_batch GCNTrainer Trainer tree_to_adj head_to_tree tree_to_dist Tree FileLogger save_config print_config check_files ensure_dir check_dir load_config score parse_arguments load MyAdagrad change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab FileLogger save_config print_config check_files ensure_dir check_dir load_config score parse_arguments load MyAdagrad change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab add_argument ArgumentParser vocab_dir save list data_dir glove_dir count_oov load_tokens parse_args build_vocab build_embedding format wv_dim wv_file lower ensure_dir items print min_freq load_glove_vocab len print format len sorted format print len Counter VOCAB_PREFIX entity_masks sum Counter values LongTensor fill_ max PAD_ID enumerate list masked_fill Variable zeros dropout size transpose matmul sqrt masked_fill softmax Variable squeeze cuda add_child tolist range len zeros T children ones dist str list DEPREL_TO_ID LABEL_TO_ID print POS_TO_ID exit shuffle get_positions word2id map_to_ids range enumerate len append head_to_tree range len print format exit print format exit print format makedirs print format print format print items enumerate len param_groups append range enumerate zero_ load_state_dict load uniform len set shape list expand parse_args add_argument ArgumentParser max list sorted format print write Counter mean append float keys range sum values

Attention Guided Graph Convolutional Networks for Relation Extraction ========== This paper/code introduces the Attention Guided Graph Convolutional graph convolutional networks (AGGCNs) over dependency trees for the large scale sentence-level relation extraction task (TACRED). You can find the paper [here](https://arxiv.org/pdf/1906.07510.pdf) See below for an overview of the model architecture:  ## Requirements Our model was trained on GPU Tesla P100-SXM2 of Nvidia DGX. - Python 3 (tested on 3.6.8)

Cartus/AGGCN_TACRED

['relation extraction']

['Attention Guided Graph Convolutional Networks for Relation Extraction']

semeval/utils/helper.py utils/vocab.py PubMed/Binary/train.py semeval/model/trainer.py semeval/prepare_vocab.py PubMed/Tenary/utils/nary_scorer.py PubMed/Tenary/eval.py prepare_vocab.py semeval/data/loader.py PubMed/Tenary/utils/vocab.py PubMed/Binary/model/trainer.py train.py PubMed/Tenary/prepare_vocab.py semeval/eval.py semeval/model/aggcn.py utils/helper.py semeval/train.py PubMed/Binary/model/graph.py PubMed/Binary/utils/nary_scorer.py PubMed/Tenary/model/graph.py PubMed/Binary/eval.py PubMed/Tenary/data/loader.py semeval/utils/constant.py PubMed/Binary/utils/constant.py PubMed/Tenary/train.py PubMed/Tenary/utils/constant.py PubMed/Tenary/utils/helper.py semeval/utils/scorer.py PubMed/Binary/utils/vocab.py PubMed/Binary/data/loader.py PubMed/Binary/prepare_vocab.py model/trainer.py utils/scorer.py data/loader.py PubMed/Tenary/model/trainer.py eval.py PubMed/Binary/utils/helper.py semeval/utils/vocab.py model/tree.py PubMed/Tenary/model/aggcn.py PubMed/Binary/utils/torch_utils.py semeval/model/tree.py model/aggcn.py semeval/utils/torch_utils.py utils/constant.py utils/torch_utils.py PubMed/Binary/model/aggcn.py PubMed/Tenary/utils/torch_utils.py entity_masks count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention attention MultiGraphConvLayer AGGCN rnn_zero_state GraphConvLayer unpack_batch GCNTrainer Trainer tree_to_adj head_to_tree tree_to_dist Tree count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids read_file sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention MultiGraphConvLayer DCGCN attention GraphConvLayer tree_to_dist head_to_graph tree_to_adj Tree head_to_tree unpack_batch GCNTrainer Trainer FileLogger save_config print_config check_files ensure_dir check_dir load_config score load change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids read_file sort_all word_dropout GCNClassifier pool GCNRelationModel clones MultiHeadAttention MultiGraphConvLayer DCGCN attention GraphConvLayer tree_to_dist head_to_graph tree_to_adj Tree head_to_tree unpack_batch GCNTrainer Trainer FileLogger save_config print_config check_files ensure_dir check_dir load_config score load change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab entity_masks count_oov load_tokens main build_vocab parse_args get_positions get_long_tensor DataLoader map_to_ids sort_all word_dropout GCNClassifier pool GCNRelationModel batched_index_select clones MultiHeadAttention attention MultiGraphConvLayer AGGCN rnn_zero_state GraphConvLayer unpack_batch GCNTrainer Trainer tree_to_adj head_to_tree tree_to_dist Tree FileLogger save_config print_config check_files ensure_dir check_dir load_config score parse_arguments load MyAdagrad change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab FileLogger save_config print_config check_files ensure_dir check_dir load_config score parse_arguments load MyAdagrad change_lr load_config keep_partial_grad set_cuda flatten_indices save get_optimizer build_embedding Vocab load_glove_vocab add_argument ArgumentParser vocab_dir save list data_dir glove_dir count_oov load_tokens parse_args build_vocab build_embedding format wv_dim wv_file lower ensure_dir items print min_freq load_glove_vocab len print format len sorted format print len Counter VOCAB_PREFIX entity_masks sum Counter values LongTensor fill_ max PAD_ID enumerate list masked_fill Variable zeros dropout size transpose matmul sqrt masked_fill softmax Variable squeeze cuda add_child tolist range len zeros T children ones dist str list DEPREL_TO_ID LABEL_TO_ID print POS_TO_ID exit shuffle get_positions word2id map_to_ids range enumerate len append head_to_tree range len print format exit print format exit print format makedirs print format print format print items enumerate len param_groups append range enumerate zero_ load_state_dict load uniform len set shape list expand parse_args add_argument ArgumentParser max list sorted format print write Counter mean append float keys range sum values

Attention Guided Graph Convolutional Networks for Relation Extraction ========== This paper/code introduces the Attention Guided Graph Convolutional graph convolutional networks (AGGCNs) over dependency trees for the large scale sentence-level relation extraction task (TACRED). You can find the paper [here](https://arxiv.org/pdf/1906.07510.pdf) See below for an overview of the model architecture:  ## Requirements Our model was trained on GPU Tesla P100-SXM2 of Nvidia DGX. - Python 3 (tested on 3.6.8)

Cassieyy/CLCINet_-MICCAI2019-

['medical image segmentation', 'lesion segmentation', 'semantic segmentation']

['CLCI-Net: Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke']

ConvLSTM.py CLCINet.py DoubleConv conv_lstm conv_1_init dilate_conv CLCInet concat_pool ConvLSTMCell ConvLSTM

# CLCINet_-MICCAI2019- CLCINet(Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke) implemented by pytorch. paper link:https://arxiv.org/abs/1907.07008 implemented by tensoflow(official):https://github.com/YH0517/CLCI_Net/blob/master/CLCI_Net.py

CedricTravelletti/MESLAS

['experimental design']

['Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling']

reporting/paper_figures/ebv_comparison/plot_eibv_heteroropic_onefig.py meslas/sensor.py meslas/covariance/cross_covariances.py meslas/tests/test_mvnorm.py examples/sample_and_plot.py reporting/paper_figures/intro_excursion_set/plot_excursion.py examples/full_example.py meslas/covariance/tests/test_heterotopic.py meslas/tests/test_cov_reduction.py doc_source/Tex/example_sample.py meslas/tests/test_geometry.py reporting/paper_figures/ebv_static_north/plot_eibv_static_north_onefig.py meslas/tests/test_discrete_sensor.py reporting/plot_variance_reduction.py reporting/paper_figures/ebv_comparison/plot_eibv_heteroropic.py meslas/tests/test_means.py meslas/external_dependencies/numpytorch.py reporting/make_animations/plotting_for_radar.py meslas/means.py meslas/geometry/tilings.py meslas/tests/test_sensor_eibv.py meslas/tests/test_sensor.py meslas/plotting_physical.py meslas/tests/test_coverage_fct.py meslas/tests/test_kriging.py reporting/paper_figures/intro_excursion_set/plot_excursion_onefig.py reporting/paper_figures/ebv_reduction_myopic/plotting_functions_onefig.py reporting/paper_figures/ebv_static_north/plot_eibv_static_north.py meslas/tests/test_discretization.py meslas/tests/test_vectors.py meslas/covariance/old.py meslas/geometry/grid.py meslas/excursion.py meslas/sensor_plotting.py meslas/vectors.py meslas/tests/test_triangular_plotting.py reporting/make_animations/animate_myopic_radar_from_groundtruth.py meslas/random_fields.py reporting/make_animations/plotting_functions.py reporting/make_animations/animate_myopic_radar.py reporting/paper_figures/ebv_reduction_myopic/plot_ebv_myopic_onefig.py meslas/inverse_random_fields.py reporting/paper_figures/ebv_reduction_myopic/plotting_functions.py meslas/covariance/heterotopic.py setup.py doc_source/source/conf.py reporting/paper_figures/intro_excursion_set/plot_excursion_from_ground_truth.py examples/kriging.py meslas/covariance/spatial_covariance_functions.py meslas/tests/test_sampling.py meslas/plotting.py generate_extensions coverage_fct_fixed_location InverseDiscreteGRF LinearMean ConstantMean plot_grid_values plot_grid_values_ax plot_grid_probas plot_grid_values plot_proba plot_krig_slice plot_2D_triangular_grid plot_2d_slice plot_grid_probas GRF DiscreteGRF Sensor DiscreteSensor Sensor DiscreteSensor GeneralizedMatrix GeneralizedVector _uniform_mixing_crosscov _parabolic_trend_crosscov ParabolicTrendCrossCov _linear_trend_crosscov LinearTrendCrossCov UniformMixing FactorStationaryCovariance FactorCovariance StationaryCovariance K_isotopic sample_isotopic k sample mu _matern32 Matern32 my_factor_cov my_crosscov my_matern32 mat2vec0 logit ravel_multi_index vec2matmul min_distrib vec2mat0 ____SHAPE____ ____NAN____ max_shape ____INDICES____ ____LINEAR_ALGEBRA____ inv_gaussian_pmf_mean_stdev ____NUMPY_COMPATIBILITY____ max_distrib ____GRADIENT____ attach_dim test_softmax_bias permute2st t categrnd ____AGGREGATE____ vmpdf_prad_pconc ____DISTRIBUTIONS_SAMPLING____ matsum normrnd circdiff rand interp1d repeat_to_shape log_normpdf repeat_dim vmpdf_a_given_b matmul0 permute2en prad2unitvec expand_batch var_distrib delta conv_t ____PERMUTE____ float kron aggregate append_dim shiftdim ____TYPE____ inv_gaussian_variance WrapTorch npys logistic repeat_all get_jacobian test_kron matvecmul0 lognormal_params2mean_stdev deg2rad unblock_diag mean_distrib lognorm_params_given_mean_stdev softmax_mask softmax_bias pconc2conc nansum crossvalincl expand_all prepend_to_ndim scatter_add bootstrap vec_on_dim ____CIRCULAR_STATS____ append_to_ndim sem_distrib prepend_dim ____STATS____ entropy inv_gaussian_cdf block_diag_irregular matmul2vec isnan nan2v unravel_index repeat_batch numpy kw_np2torch rad2deg lognorm_pmf tensor mvnpdf_log inv_gaussian_variance2lam inv_gaussian_mean_std2params block_diag expand_upto_dim mvnrnd vmpdf freeze sem inv_gaussian_pdf sumto1 std_distrib nanmean ____CROSS_VALIDATION____ onehotrnd create_triangular_grid SquareGrid get_isotopic_generalized_location IrregularGrid TriangularGrid create_square_grid generate_squares generate_unit_hexagons draw_tiling _scale_coordinates generate_triangles generate_unit_squares generate_hexagons generate_unit_triangles data_feed data_feed data_feed plot_grid_values plot_grid_values_ax plot_grid_probas plot_myopic_radar to_polar cart2pol data_feed OOMFormatter plot data_feed OOMFormatter plot plot_myopic_radar to_polar cart2pol plot_myopic_radar to_polar cart2pol data_feed OOMFormatter plot data_feed OOMFormatter plot data_feed OOMFormatter plot data_feed plot data_feed plot data_feed plot print Extension append stack multivariate_normal_cdf axis tick_params xticks yticks show str set_title colorbar imshow title scatter ylim ceil range set_xlim sqrt toggle_label ImageGrid xlim int isinstance set_yticks interpolate_to_image set_xticks isotopic figure numpy set_ylim len show colorbar imshow title interpolate_to_image scatter figure ylim xticks savefig xlim numpy yticks set_title isinstance make_axes_locatable append_axes set_xlim colorbar set imshow interpolate_to_image set_visible isotopic scatter tick_params numpy set_ylim axis xticks yticks show str set_title colorbar imshow title ceil range sqrt ImageGrid int set_yticks set_xticks figure numpy len axis xticks yticks show str set_title colorbar imshow title scatter ceil range sqrt toggle_label ImageGrid item int set_yticks set_xticks figure numpy len show colorbar imshow title xticks numpy yticks show tricontourf colorbar title scatter xticks numpy yticks meshgrid fill_ item fill_ reshape transpose item meshgrid sum fill_ reshape transpose item meshgrid einsum K mu MultivariateNormal tensor sqrt exp Tensor pop list keys exp arange tensor log cat parameters Size expand shape ndimension isnan clone isnan clone exp min clone float sum ones expand repeat append tensor max cat ndimension ones tensor dim append attach_dim amax zip attach_dim repeat cat zip append zeros max amax len dim ndimension ndimension ravel_multi_index LongTensor stack is_tensor npy tensor exp zeros transpose expand tensor long is_floating_point cat transpose expand_as tensor long is_floating_point var_distrib is_tensor sum clamp_min sumto1 reshape cumsum clone shape pad any nan2v sum prod min_distrib flip log2 show subplot plot xlabel print grid axis softmax_bias linspace tensor xticks npy enumerate yticks sqrt exp exp print isnan sqrt any cdf inv_gaussian_variance2lam inv_gaussian_pdf zeros_like inv_gaussian_cdf print isnan any expand_all cat sqrt log zeros_like lognorm_params_given_mean_stdev print isnan any cdf expand_all cat unsqueeze Uniform MultivariateNormal Normal rsample tensor MultivariateNormal eye len fun append randint range len expand_upto_dim backward squeeze requires_grad_ repeat eye net Size tensor unsqueeze print repeat_dim shape unsqueeze kron LongTensor Size zeros sum enumerate attach_dim unsqueeze dim cat Size zeros p2st tensor range long pi clamp double exp VonMisesFisher sumto1 sqrt zeros sum log_prob meshgrid stack linspace append generate_triangles reshape repeat_interleave long generator int range int range pi sin int range pi sin polygon coord_generator new save sample plot_grid_values_ax grid add_subplot linspace tensor show location argmin scatter savefig legend add_gridspec sum set_xlim close to_polar ListedColormap set_yticks repeat set_xticks figure color_palette numpy compute_exursion_prob sqrt arctan2 show ListedColormap plot_grid_values_ax set_yticks grid add_subplot set_xlim subplots_adjust close scatter set_xticks Normalize figure color_palette legend savefig

# MESLAS: Multivariate Excursion Set Learning by Adaptive Sampling The MESLAS package provides functionalities for simulation of multivariate gaussian random fields, as well as adaptive sampling startegies to learn excursion sets thereof. It originated as part of a collaboration between NTNU Trondheim and University of Bern that aimed at developing methods for identification of excursion sets of vector-valued random fields for applications in autonomous ocean sampling. More information may be found in the resulting [article](https://arxiv.org/abs/2007.03722). The package documentation and more details may be found on the [Project Website](https://cedrictravelletti.github.io/MESLAS/).

CellEight/Pytorch-Adaptive-Instance-Normalization

['style transfer']

['Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization']

train.py style.py AdaIN.py utils.py model.py AdaIN AdaINStyle debugHook styleHook loadImage debugHook AdaINLoss styleHook ImageDataset mu sigma preprocess Compose convert

# Pytorch-Adaptive-Instance-Normalization A Pytorch implementation of the 2017 Huang et. al. paper "Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization" [https://arxiv.org/abs/1703.06868](https://arxiv.org/abs/1703.06868) Written from scratch with essentially no reference to Xun Huangs implementation in lua/torch (can be found here: [https://github.com/xunhuang1995/AdaIN-style](https://github.com/xunhuang1995/AdaIN-style)) but I'm none the less incredbily greatful to Huang et. al. for writing such an outstandingly beautiful paper and making their method so clear and easy to implement!  ## Requirements To run this model please install the latest version of pytorch, torchvision and CUDA. ## Loading Pretrained Weights I have made a set of pretrained weights availabe on google drive if you don't want to train the model yourself. You can find them here [https://drive.google.com/file/d/1094pChApSOA7qJZn68kEdNxKIwPWRdHn/view?usp=sharing](https://drive.google.com/file/d/1094pChApSOA7qJZn68kEdNxKIwPWRdHn/view?usp=sharing). Once downloaded just place it into the root directory of the repo and you're good to go. ## Usage

Cerenaut/aha

['one shot learning']

['Unsupervised One-shot Learning of Both Specific Instances and Generalised Classes with a Hippocampal Architecture']

train_ltm.py aha/components/hopfieldlike_component.py aha/components/dg_sae.py aha/utils/generic_utils.py aha/utils/interest_filter.py plot_results.py aha/utils/recursive_component_harness.py aha/workflows/episodic_pattern_completion_workflow.py aha/workflows/dp_workflow.py aha/components/episodic_component.py plot_results_combined.py aha/datasets/omniglot_unseen_dataset.py aha/components/diff_plasticity_unit_tests.py aha/components/dg_stub.py aha/components/label_learner_fc.py aha/utils/few_shot_utils.py aha/workflows/episodic_workflow.py aha/datasets/omniglot_unseen_oneshot_dataset.py aha/datasets/omniglot_lake_dataset.py aha/workflows/episodic_few_shot_workflow.py aha/datasets/omniglot_lakelike_runs_dataset.py aha/components/recursive_sae_component.py aha/workflows/pattern_completion_workflow.py build_all_runs.py aha/datasets/omniglot_lake_runs_dataset.py aha/components/diff_plasticity_component.py build_all_runs2.py aha/components/dg_scae.py aha/datasets/omniglot_supervised_dataset.py aha/components/deep_autoencoder_component.py aha/samples/simple_diff_plasticity.py setup.py parse_results.py aha/samples/1-10-vanilla-rnn.py copy_files plot_mean_sd get_filenames concatenate_results build_xaxis compute_statistics load_and_parse_data main plot_mean_sd get_filenames concatenate_results build_xaxis chunks compute_statistics load_and_parse_data main BinaryDistribution main DeepAutoencoderComponent DGSAE DGSCAE DGStubComponent DifferentiablePlasticityComponent DifferentiablePlasticityUnitTests PCMode EpisodicComponent HopfieldlikeComponent unit_to_pc_sparse dg_to_pc pc_to_unit get_pc_topk_shift unit_to_pc_linear LabelLearnerFC RecursiveSAEComponent OmniglotLakelikeRunsDataset OmniglotLakeDataset OmniglotLakeRunsDataset OmniglotSupervisedDataset OmniglotUnseenDataset OmniglotUnseenOneShotDataset tf_build_stats_summaries tf_reduce_var plot generate_data variable_summaries get_summary_dir tfsumsc generate_inputs_and_target tfprint reset_random zero_if_less_than add_completion_summary_paper mod_hausdorff_distance compute_matching add_completion_summary create_and_add_comparison_image overlap_match normalize_minmax build_kernel_initializer overlap_sample_batch print_minmax overlap compute_overlap InterestFilter RecursiveComponentHarness DPWorkflow MockData EpisodicFewShotWorkflow EpisodicPatternCompletionWorkflow EpisodicWorkflow UseTrainForTest PatternCompletionWorkflow int str basename join copyfile zfill enumerate append float parse_csv enumerate len OrderedDict zip zeros enumerate items list namedtuple print len summary_stats OrderedDict mean std reject_outliers enumerate append round range append join walk print fill_between plot subplots set_visible tick_params max plot_mean_sd list set_title set_xlabel build_xaxis compute_statistics twinx savefig legend append concatenate_results plot set_xlim get_xlim load_and_parse_data keys join get_filenames print set_ylabel set_ylim len range len models perturb input_path xticks exp metric update num_seeds chunks zfill strftime call range to_float greater tf_set_min as_list expand_dims reduce_max tf_build_top_k_mask_op to_float tf_set_min greater reshape array roll choice now strftime reduce_mean square subplot list pause draw axis bar array range cla seed set_random_seed int permutation format ones debug shuffle copy shape append zeros range scalar cdist min mean subplots axis clip show str transpose imshow savefig arbitrary_image_summary flat format np_write_array_list_as_image square mean enumerate join switch_backend reshape text subplots_adjust len show join format subplots switch_backend reshape axis tight_layout imshow flat savefig range enumerate len int overlap_match format arange ones print mean accuracy_score ceil zeros float max enumerate list compute_truth_matrix_oneshot keys compute_truth_matrix_instance compute_accuracy compute_matrix len confusion_matrix zip append accuracy_score overlap enumerate append overlap zip enumerate

# Artificial Hippocampus Algorithm The codebase for the Artificial Hippocampus Algorithm (AHA) project. ## Dependencies - [PAGI Framework](https://github.com/ProjectAGI/pagi) >= 0.1 ## Getting Started Ensure that you have `pagi` installed and that its accessible via the command-line by running `pagi --help`. Clone this repository and install the package using `pip install -e .` ### Running Experiments You can run experiments using any of the existing definition files, or your own definition file, using the following command: `pagi run --experiment_def=definitions/FILENAME.json`

ChWick/ocropy

['optical character recognition']

['Improving OCR Accuracy on Early Printed Books using Deep Convolutional Networks']

OLD/ocrolib/ngraphs.py OLD/mlp.py OLD/test-feature-extractor.py ocrolib/psegutils.py ocrolib/__init__.py ocrolib/ligatures.py ocrolib/lstm.py ocrolib/lang.py OLD/lineproc.py ocrolib/utils.py OLD/wmodel.py ocrolib/chars.py ocrolib/extras/fgen.py ocrolib/extras/cairoextras.py ocrolib/sl.py OLD/distance.py ocrolib/lineest.py OLD/patrec.py ocrolib/toplevel.py ocrolib/morph.py OLD/lineseg.py ocrolib/hocr.py OLD/improc.py OLD/ocrolib/lattice.py ocrolib/default.py ocrolib/edist.py setup.py OLD/linerec.py OLD/mlinear.py ocrolib/exceptions.py OLD/ocrolib/h5utils.py ocrolib/common.py requote_fancy requote chist allsplitext midrange normalize_text ustrg2unicode warn write_image_gray RegionExtractor norm_max array2pil load_object binarize_range parallel_map save_object check_valid_class_label isbytearray plotgrid MovingStats caller quick_check_line_components iulib_page_iterator remove_noise base gt_explode read_line_segmentation read_image_gray project_text expand_args finddir rgb2int make_seg_black write_text_simple showrgb pil2array isintarray write_line_segmentation findfile int2rgb number_of_processors read_image_binary fvariant showgrid Record glob_all write_text unpickle_find_global set_params ocropus_find_file quick_check_page_components isintegerarray pad_by gt_implode read_text testset read_page_segmentation write_page_segmentation make_seg_white die warn_once obinfo isfloatarray write_image_binary getlocal xlevenshtein levenshtein Unimplemented FileNotFound BadClassLabel Warning Internal summary BadImage RecognitionError BadInput OcropusException footer header size_category common_ligatures LigatureTable scale_to_h CenterNormalizer hprime RangeError randu log_mul Softmax gfunc log_add ffunc Parallel rownorm make_target ctc_align_targets MLP1 check_nan ascii_codec translate_back Network Stacked forwardbackward SeqRecognizer MLP getstates_for_display Codec BIDILSTM sumouter Logreg add_training_info prepare_line hfunc forward_py normalize_nfkc gprime ocropus_codec sigmoid fprime forward_algorithm Reversed LSTM translate_back0 LSTM1 backward_py rg_closing select_regions keep_marked remove_marked r_erosion renumber_labels_ordered rg_erosion renumber_labels rg_dilation rb_closing find_objects check_binary renumber_by_xcenter r_opening propagate_labels r_dilation r_closing propagate_labels_simple rb_dilation pyargsort showlabels label all_neighbors correspondences rb_erosion rb_opening ordered_by_xcenter rg_opening spread_labels extract reading_order estimate_scale B read_gray record compute_lines read_binary rgbshow show_lines topsort blackout_images binary_objects extract_masked compute_boxmap pad_image find dims volume area yoverlap mbox dim1 extend_to is_slices xoverlaps center0 yoverlaps center_in aspect pad center1 width cut union dim center height intersect raster_FIXME bounds start dim0 box empty xoverlap_rel math ycenter shift raster xoverlap stop yoverlap_rel xcenter DATASET failfunc makeargcheck WHITESEG checks DARK inttuple PAGE SEGMENTATION ANONNEG ABINARY DATASET_VRANK PAGEEXTRA uintpair BOOL ARANGE uinttuple checktype RECTANGLE AINT disabled RANGE CheckWarning ABYTE TDATASET ANY DATASET_SIZE NUMBER BLACKSEG DATASET_VSIZE DATASET_VRANGE ARANK AFLOAT tracing CHANNELS LINE PATCH ALL trace1 unchanged method LIGHT deprecated strc replacedby CheckError sumprod sumouter PycairoContext create_cairo_font_face_for_file pango_render_gray pango_render_string pango_families gauss_distort cairo_render_string gauss_degrade cairo_render_gray cairo_render_at cdist ProtoDists c_order cdist_native_load extract shaped make_mask extract_centered stdsize dist classifier_normalize norm_max line_normalize csnormalize extract_centered_scaled symdist remove_noise center_maxsize pad_bin cut isotropic_rescale square extract_centered_scaled_barred pad_by pad_to deprecated cut_inefficient bbox latin_mask seg_geometry dewarp_line latin_filter avg seg_boxes rel_geo_normalize estimate_baseline rel_char_geom estimate_xheight read_lattice clean_line_image has_limited_gaps box_union extract_rsegs bestpath write_lattice best_correlation Segment number_of_vertical_strokes extract_seg shortest_path all_min_gaps non_noise_components extract_candidate_groups extract_char good_complexity extract_non_csegs extract_csegs max_boxgap number_of_holes check_line_image all_gaps CCSSegmentLine ccslineseg dplineseg2 ComboSegmentLine SimpleParams image_draw_line dplineseg1 seq2list contourcuts between image2contours centroid dptrack DPSegmentLine dpcuts logit logpred logreg_opt LinPcaClassifier pca lstsq_l2 LinKernelClassifier MLP LinClassifier logreg_fp logreg_gd dlogloss logreg_l2_fp lstsq_l1 sigmoid Err make2d logloss DiagGaussian Record nnet_native_load c_order MLP error test log_uniform finite AutoMLP kmeans pca_kmeans TrivialCostModel pca_kmeans0 coutputs_chunk pca Grad2Model predict_chunk distribution normalizer_normal HierarchicalSplitter cshow rselect datachunks vecsort LocalCmodel minsert normalizer_none knn Kmeans showgrid Grad0Model parallel_predict ModelWithExtractor Extractor2 Extractor1 Grad1Model PCA TrivialCmodel LogisticCmodel Err make2d Extractor0 PcaKmeans protosets Dataset method parallel_coutputs BadImage WhitespaceModel BadGroundTruth assign_array create_earray log_copy log table_copy Edge Lattice Lattice2 NGraphsCounts rsample safe_readlines ComboDict NGraphs NGraphsBackoff str sub str sub replacements str normalize sub upper normalize_text sub normalize_text normalize_text tobytes fromstring mean pil2array isfloatarray open print array2pil array save isfloatarray clip pil2array amax open print array2pil array save zeros list shape copy copy rgb2int make_seg_black pil2array open array2pil make_seg_white int2rgb save rgb2int make_seg_black pil2array open array2pil make_seg_white int2rgb save read_image_gray dtype zeros shape exec print ocropus_find_file get imap_unordered Pool fun search ocropus_find_file getlocal split search glob sorted join curdir get_config_var getenv dirname normpath getfile append currentframe exists pardir allsplitext items list hasattr copy setattr _getframe getframeinfo caller write exit caller write caller write length at chr range str hasattr amin amax subplot ginput reshape min gray imshow clf ion range len imshow transpose minimum int subplot str yticks xlabel ylabel gray sqrt imshow title xticks range len enumerate split append minimum list label sum range list min range join arange minimum_filter array split append empty full range len int shape affine_transform eye array T vstack amax any zip sigmoid tanh tanh hfunc gfunc ffunc dot range hprime list hfunc sumouter gprime reversed dot fprime sumprod range Logreg Stacked Logreg LSTM Softmax Stacked Softmax Parallel Reversed LSTM Stacked zeros enumerate append argmax range amax len arange reshape tile label maximum_position amax log_mul arange log_add copy append range len forward_algorithm subplot T exp amax ginput maximum dot imshow clf figure log forwardbackward set isinstance check_binary r_erosion check_binary r_dilation zeros uniform_filter shape zeros uniform_filter shape rb_erosion rb_dilation r_erosion r_dilation imshow where reshape distance_transform_edt label ravel in1d unique keep_marked array unique correspondences T label zeros amax correspondences T label zeros amax find_objects argsort label zeros len array unique roll sorted arange unique zeros ravel amax len find_objects argsort zeros array amax enumerate len find_objects range array bytearray print pad_by length label_components copy unpack_rgb textImageProbabilities at bounding_boxes rectarray range fill_rect intarray label find_objects median sorted area shape binary_objects zeros zeros sorted binary_objects shape record find_objects append enumerate ones shape array shape affine_transform shift eye extract mask where maximum_filter pad_image amax center plot ginput print imshow title clf x_overlaps zeros above enumerate left_of visit zeros range len ravel nonzero center plot bounds len add_patch imshow shape clf ylim Rectangle append xlim range cla mean imread mean imread clip print transpose shape imshow zeros abs array range list slice start stop range len list dtype intersect bounds transpose shift dims start pad empty isinstance __name__ type_ callable isinstance isinstance isinstance sum GRAYSCALE1 unique zeros zeros get_font_face CDLL ctx FORMAT_A8 cairo_ft_font_face_create_for_ft_face c_void_p cairo_set_font_face ImageSurface Context FORMAT_ARGB32 set_source_rgb create_cairo_font_face_for_file get_data select_font_face max show_text move_to range Context bytearray set_font_face fill set_font_size ImageSurface int rectangle array len CairoContext FORMAT_ARGB32 get_context ImageSurface create_layout Context set_font_description FORMAT_ARGB32 set_text set_source_rgb get_data set_size max CairoContext rotate SCALE move_to create_layout range Context bytearray zoom show_layout FontDescription fill get_pixel_extents ImageSurface int rectangle set_markup array len int gaussian_filter distance_transform_edt min binary_erosion mean shape prod sum max binary_dilation list randn transpose shape meshgrid array range gaussian_filter FONT_SLANT_NORMAL FORMAT_ARGB32 FONT_WEIGHT_BOLD set_source_rgb create_cairo_font_face_for_file get_data select_font_face show_text move_to Context bytearray set_font_face fill set_font_size ImageSurface FONT_SLANT_OBLIQUE rectangle FONT_SLANT_ITALIC FONT_WEIGHT_NORMAL array compile_and_load cpu_count in_dll cdist_native_load cpu_count zeros array len shape zeros int array affine_transform array diag affine_transform array diag amax dtype shift shape int shape dtype max zeros shape square max dtype zoom shape zeros float array thin show subplot abs make_mask shift min imshow center_of_mass sum max cla dist array find_objects diag affine_transform min diag bbox max dot sqrt shape eigh affine_transform sum array amax isotropic_rescale array amax csnormalize center_of_mass max scoreatpercentile plot print add_subplot sca polyfit add_patch imshow figure seg_boxes Rectangle gca array cla avg array clip log array range find_objects len arange amax grey_closing print argmin polyfit shape gaussian_filter arange shift polyval shape zeros estimate_baseline range amax sum grey_closing argmin imshow argmax gaussian_filter int grey_closing argmin maximum shape zeros argmax range gaussian_filter len latin_mask keep_marked maximum_filter maximum start stop range len union range len find_objects aspect append Segment range box_union len minimum distance_transform_edt len unique append amin range enumerate unique all_gaps label list sum range all_min_gaps extract_seg plot amax imshow sum gaussian_filter label binary_fill_holes number_of_holes number_of_vertical_strokes extract_seg extract_candidate_groups find_objects pad_by normalize correlate append img append best_correlation estimate_xheight remove_noise latin_filter noise_threshold amax sorted tuple write sp out bbox enumerate append heappop heappush len shortest_path out append max enumerate ones where roll shape zeros abs range len range shape between zeros clip len dpcuts maximum roll dptrack gaussian_filter find sum int subplot plot ginput dpcuts maximum where imshow clf figure centroid dptrack gaussian_filter find keep_marked maximum_filter label gaussian_filter spread_labels list h_next sort append array fromarray FindContours CV_CHAIN_APPROX_NONE CV_RETR_CCOMP CV_RETR_EXTERNAL array CreateMemStorage int linspace arange clf maximum_filter ion abs list ones cdist image_draw_line imshow shape append input range plot ginput grey_erosion circulant eval label center_of_mass gaussian_filter enumerate minimum print maximum figure zeros image2contours len T value randn abs print maximum add sigmoid shape dot Err sum range len print reshape sum logpred dot reshape T shape fmin_bfgs ravel randn T ones inv dot shape diag logit print dot sigmoid sum range lstsq T reshape dot shape eigh argsort sqrt diag classify take compile_and_load int in_dll getenv randn print AutoMLP classify train sum array clip int reshape size sqrt gray imshow ion reshape size ion list Counter zeros max values argmin list minsert cdist print minsert argsort sample amin len sorted maximum accumulate abs all cdist argmin average sample array range list kmeans print reshape min sample pca array len pca_kmeans cdist argsort append array range len enumerate amax add min array range len datachunks print range parallel_map len datachunks print range parallel_map len flush print createEArray len getNode createVLArray root atom range append split Int64Atom print createEArray Int32Atom root Float32Atom append getNodeAttr setNodeAttr dir _v_attrs int time setNodeAttr join Float32Atom Int64Atom accumulate rand readline range

ocropy ====== [](https://travis-ci.org/tmbdev/ocropy) [](https://circleci.com/gh/UB-Mannheim/ocropy.png) [](https://hub.docker.com/r/ubma/ocropy/) [](https://hub.docker.com/r/ubma/ocropy/) [](https://github.com/tmbdev/ocropy/blob/master/LICENSE) [](https://github.com/tmbdev/ocropy/wiki) [](https://gitter.im/tmbdev/ocropy?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) OCRopus is a collection of document analysis programs, not a turn-key OCR system. In order to apply it to your documents, you may need to do some image preprocessing,

ChaoFan96/GaitPart

['gait recognition']

['GaitPart: Temporal Part-Based Model for Gait Recognition']

layers.py TemporalFeatureAggregator BasicConv1d FocalConv2d clones

# GaitPart： Temporal Part-based Model for Gait Recognition ***！！！ IMPORTANT ！！！*** ***Check [OpenGait](https://github.com/ShiqiYu/OpenGait) to get the full codes of GaitPart！！！*** GaitPart is a CVPR2020 paper, and some of the source code is already opened while the rest will be released as soon as possible! ## Updates * Some of the source code is already opened, enjoy it ~ * My school e-mail '[email protected]' has been frozen, contact me by '[email protected]' if there is any question. * The paper is available [here](http://openaccess.thecvf.com/content_CVPR_2020/papers/Fan_GaitPart_Temporal_Part-Based_Model_for_Gait_Recognition_CVPR_2020_paper.pdf)！ ## Correction * *Clerical error* at Sec4.1 -> Training Details -> 3)In OU-MVLP, the value of p in each block have been set to 2, 2, 8, 8 but not 1, 1, 3, 3 (you know, 2=2^1, 8=2^3) in real practice.

ChaoYang93/GraduatePaper

['time series analysis', 'time series']

['Gated Res2Net for Multivariate Time Series Analysis']

GRes2Net.py FCRSN.py CRCN.py conv5 conv1x1 dilationblock2 ception2 adoptblock dilationblock1 dilation conv3x3 ception1 inceptionblock ception3 dilationblock3 conv1x1 conv3x1 Res2NetBottleneck Res2RNN conv5x1 SEModule conv3 GatedFCN GatedRes2NetBottleneck conv1 SEModule

# GraduatePaperA Codes for the model CRCN,FCRSN and GRes2Net. Trained model is available on: https://pan.baidu.com/s/1LaRqOQWUBZjsTaKG80tY9Q. For downloading, the extraction code is fj9a Thanks for reading.

Charlesthebird/BOXES-DroneDelivery-ML-Agents

['unity']

['Unity: A General Platform for Intelligent Agents']

gym-unity/tests/test_gym.py gym-unity/gym_unity/__init__.py gym-unity/gym_unity/envs/__init__.py gym-unity/setup.py gym-unity/gym_unity/envs/unity_env.py UnityGymException UnityEnv test_gym_wrapper test_multi_agent sample step MockCommunicator UnityEnv step MockCommunicator UnityEnv

<img src="docs/images/unity-wide.png" align="middle" width="3000"/> <img src="docs/images/image-banner.png" align="middle" width="3000"/> # Unity ML-Agents Toolkit (Beta) **The Unity Machine Learning Agents Toolkit** (ML-Agents) is an open-source Unity plugin that enables games and simulations to serve as environments for training intelligent agents. Agents can be trained using reinforcement learning, imitation learning, neuroevolution, or other machine learning methods through a simple-to-use Python API. We also provide implementations (based on TensorFlow) of state-of-the-art algorithms to enable game developers and hobbyists to easily train intelligent agents for 2D, 3D and VR/AR games. These trained agents can be

ChenXiao61/Img_augmentor

['data augmentation', 'image augmentation']

['Augmentor: An Image Augmentation Library for Machine Learning']

Augmentor/ImageSource.py tests/test_ground_truth_by_class.py tests/test_multi_threading.py tests/test_datapipeline.py tests/util_funcs.py tests/test_pipeline_add_operations.py tests/test_rotate.py tests/test_generators.py tests/test_user_operation_parameter_input.py docs/conf.py tests/test_load.py Augmentor/Pipeline.py tests/test_ground_truth_augmentation.py Augmentor/Operations.py tests/test_resize.py tests/test_torch_transform.py tests/test_gaussian.py Augmentor/__init__.py tests/test_random_color_brightness_contrast.py tests/test_distortion.py Augmentor/ImageUtilities.py setup.py ImageSource parse_user_parameter extract_paths_and_extensions scan_directory scan scan_directory_with_classes AugmentorImage scan_dataframe Rotate RandomColor Custom Zoom RandomBrightness RandomErasing CropRandom ZoomRandom Greyscale RotateRange Invert HistogramEqualisation CropPercentage Operation Skew Resize HSVShifting RotateStandard Crop ZoomGroundTruth Shear Flip GaussianDistortion RandomContrast Scale Distort BlackAndWhite DataPipeline Pipeline DataFramePipeline test_sample_with_masks test_sample_with_no_masks test_in_memory_distortions test_add_gaussian_to_pipeline test_create_gaussian_distortion_object test_image_generator_function test_generator_with_array_data test_generator test_generator_image_scan test_keras_generator_from_disk test_zoom_ground_truth_temporary_class_without_ground_truth_images test_loading_ground_truth_images test_zoom_ground_truth_temporary_class test_skew_ground_truth test_distort_gaussian_ground_truth test_rotate_ground_truth test_black_and_white_ground_truth test_crop_percentage_ground_truth test_flip_ground_truth test_random_erasing_ground_truth test_greyscale_operation_ground_truth test_scale_ground_truth test_distort_ground_truth test_hsv_shift_ground_truth create_temporary_data test_invert_operation_ground_truth test_shear_ground_truth test_rotate_ground_truth_multiple_passes test_crop_ground_truth test_zoom_ground_truth test_histogram_equalisation_ground_truth test_zoom_random_ground_truth test_flip_ground_truth_multiple_passes test_rotate_range_ground_truth test_rotate_standard_ground_truth destroy_temporary_data test_crop_random_ground_truth test_dataframe_initialise_with_ten_images test_initialise_with_no_parameters test_initialise_with_missing_folder test_initialise_with_ten_images test_class_image_scan test_initialise_with_nondefault_output_directory test_initialise_with_subfolders test_initialise_with_empty_folder test_simple_multi_threading_example test_all_operations_multi_thread test_multi_threading_override test_add_rotate_operation test_random_contrast_in_memory test_random_color_in_memory test_random_brightness_in_memory test_resize_save_to_disk test_resize_in_memory test_rotate_images_270 test_rotate_images_180 rotate_images test_rotate_images_custom_temp_files test_rotate_images_90 test_torch_transform test_user_param_parsing create_colour_temp_image create_greyscale_temp_image create_sub_folders Real isinstance dirname splitext join sorted basename isdir glob scan_directory AugmentorImage abspath append zeros list categories get_values codes len AugmentorImage Categorical abspath zip append zeros values enumerate glob join extend abspath join isdir glob scan_directory warn append BytesIO new mkdtemp write close getvalue NamedTemporaryFile rotate sample rmtree save DataPipeline append range flush len save DataPipeline fromarray list new mkdtemp getvalue NamedTemporaryFile rotate append range glob close zip sample flush join BytesIO print write rmtree zoom_random len create_colour_temp_image close Distort rmtree perform_operation create_greyscale_temp_image append open GaussianDistortion gaussian_distortion mkdtemp Pipeline save new mkdtemp getvalue NamedTemporaryFile rotate append next range close flip_left_right Pipeline flush image_generator BytesIO write rmtree flip_top_bottom len save new mkdtemp getvalue NamedTemporaryFile rotate append next range close flip_left_right Pipeline flush BytesIO write rmtree flip_top_bottom keras_generator randint len fromarray reshape astype keras_generator_from_array range rotate randint next Pipeline zeros len BytesIO new mkdtemp write close getvalue NamedTemporaryFile rmtree keras_generator save append augmentor_images next Pipeline range flush len fromarray join uint8 name rand mkdtemp close choice NamedTemporaryFile rmtree keras_generator save append randint next Pipeline range fromarray join uint8 augmentor_images rand mkdtemp rmtree abspath save append randint ground_truth Pipeline range fromarray join uint8 add_operation glob rand mkdtemp sample ZoomGroundTruth rmtree abspath save append randint Pipeline range fromarray join uint8 add_operation glob rand mkdtemp sample ZoomGroundTruth rmtree abspath save append randint ground_truth Pipeline range fromarray join uint8 rand mkdtemp abspath save append randint range rmtree fromarray join uint8 glob rand mkdtemp sample rmtree histogram_equalisation abspath save append randint ground_truth Pipeline range fromarray join uint8 greyscale glob rand mkdtemp sample rmtree histogram_equalisation abspath save append randint ground_truth Pipeline range fromarray join uint8 invert greyscale glob rand mkdtemp sample rmtree histogram_equalisation abspath save append randint ground_truth Pipeline range fromarray join uint8 greyscale glob rand mkdtemp sample rmtree abspath save append randint ground_truth Pipeline range black_and_white fromarray join uint8 skew greyscale glob rand mkdtemp sample rmtree abspath save append randint ground_truth Pipeline range fromarray join uint8 greyscale glob rand mkdtemp sample rotate_without_crop rmtree abspath save append randint ground_truth Pipeline range join rotate90 glob destroy_temporary_data ground_truth sample create_temporary_data Pipeline join rotate180 rotate90 glob rotate_random_90 destroy_temporary_data ground_truth rotate270 sample create_temporary_data Pipeline join glob rotate destroy_temporary_data ground_truth sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth sample flip_left_right create_temporary_data Pipeline join glob flip_random destroy_temporary_data flip_top_bottom ground_truth sample flip_left_right create_temporary_data Pipeline crop_by_size join glob destroy_temporary_data ground_truth sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth crop_centre sample create_temporary_data Pipeline crop_random join CropRandom add_operation glob destroy_temporary_data ground_truth sample create_temporary_data Pipeline shear join glob destroy_temporary_data ground_truth sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth scale sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth random_distortion sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth sample gaussian_distortion create_temporary_data Pipeline join zoom glob destroy_temporary_data ground_truth sample create_temporary_data Pipeline join glob destroy_temporary_data ground_truth sample zoom_random create_temporary_data Pipeline join add_operation glob sample destroy_temporary_data ground_truth HSVShifting create_temporary_data Pipeline join glob destroy_temporary_data ground_truth random_erasing sample create_temporary_data Pipeline Pipeline mkdtemp Pipeline mkdtemp Pipeline fromarray join uint8 list create_sub_folders isdir name glob rand mkdtemp close NamedTemporaryFile scan_directory_with_classes rmtree save append keys values BytesIO new len mkdtemp write close getvalue NamedTemporaryFile image_path rmtree save append Pipeline range flush open importorskip save DataFrame open DataFramePipeline new mkdtemp getvalue NamedTemporaryFile image_path append range close flush BytesIO write dict rmtree len fromarray join uint8 name rand mkdtemp len close choice NamedTemporaryFile rmtree mkdir save append randint range run join name glob new mkdtemp close sample NamedTemporaryFile rmtree open save resize append Pipeline range len save resize name new mkdtemp rotate NamedTemporaryFile append range glob close sample flip_left_right Pipeline join flip_random rmtree flip_top_bottom len join name glob new mkdtemp close sample NamedTemporaryFile rmtree open save resize append Pipeline range len Pipeline rotate create_colour_temp_image RandomColor close rmtree perform_operation create_greyscale_temp_image append open create_colour_temp_image close RandomContrast rmtree perform_operation create_greyscale_temp_image append open create_colour_temp_image close rmtree perform_operation create_greyscale_temp_image append RandomBrightness open name new mkdtemp close NamedTemporaryFile Resize perform_operation save join name glob new mkdtemp close sample NamedTemporaryFile rmtree open save resize append Pipeline range len join Rotate str new perform_operation save rotate_images rotate_images rotate_images Rotate name new mkdtemp close NamedTemporaryFile rmtree perform_operation save fromarray greyscale uint8 zoom rotate_random_90 Compose importorskip zeros Pipeline operations parse_user_parameter fromarray uint8 name rand mkdtemp NamedTemporaryFile save fromarray uint8 name rand mkdtemp NamedTemporaryFile save fromarray uint8 name rand mkdtemp NamedTemporaryFile save append range

 Augmentor is an image augmentation library in Python for machine learning. It aims to be a standalone library that is platform and framework independent, which is more convenient, allows for finer grained control over augmentation, and implements the most real-world relevant augmentation techniques. It employs a stochastic approach using building blocks that allow for operations to be pieced together in a pipeline. [](https://pypi.python.org/pypi/Augmentor) [](https://pypi.python.org/pypi/Augmentor) [](https://augmentor.readthedocs.io/en/master/?badge=master) [](https://travis-ci.org/mdbloice/Augmentor) [](LICENSE.md) [](http://www.repostatus.org/#active) [](https://mybinder.org/v2/gh/4QuantOSS/Augmentor/master) ## Installation

Cheneng/DPCNN

['sentiment analysis', 'text classification']

['Deep Pyramid Convolutional Neural Networks for Text Categorization']

data/__init__.py model/BasicModule.py model/DPCNN.py config.py data/dataset.py main.py Config TextDataset BasicModule DPCNN

# Deep Pyramid Convolutional Neural Networks for Text Categorization > This is a simple version of the paper *Deep Pyramid Convolutional Neural Networks for Text Categorization*.  You should rewrite the Dataset class in the data/dataset.py and put your data in '/data/train' or any other directory. run by ``` python main.py --lr=0.001 --epoch=20 --batch_size=64 --gpu=0 --seed=0 --label_num=2 ``` ## Evaluation

ChengBinJin/Adam-Analysis-TensorFlow

['stochastic optimization']

['Adam: A Method for Stochastic Optimization']

src/download.py src/dataset.py src/solver.py src/cifar10.py src/main.py src/cache.py src/mnist.py src/models.py src/utils.py src/tensorflow_utils.py cache ExpensiveClass expensive_function convert_numpy2pickle maybe_download_and_extract CIFAR10 load_cached one_hot_encoded DataSet _print_download_progress download maybe_download_and_extract save_model load_model test plot_loss init_logger main train MNIST init_logger optimizer_fn NeuralNetwork Logistic CNN Solver n_res_blocks flatten padding2d lrelu identity conv2d elu res_block batch_convert2int batch_norm dropout relu convert2int instance_norm xavier_init print_activations tanh norm linear upsampling2d max_pool sigmoid deconv2d show_all_variables plot_images release_handles init_logger make_folders CSVWriter print fn exists load max print cache float format write flush print join urlretrieve makedirs join urlretrieve print endswith extractall makedirs model batch_size is_train setLevel load_model addHandler setFormatter format gpu_index print_freq StreamHandler info random_seed INFO FileHandler join learning_rate print Formatter is_whiten epochs MNIST load_model info preprocessing gpu_index strftime test is_train init_logger CIFAR10 train make_folders CNN save_model batch_size set_random_seed plot_loss Saver reset_default_graph Logistic round Session str load_model range CSVWriter update random_batch format num_train close FileWriter release_handles init info random_seed flush int join evaluate print NeuralNetwork add_summary epochs Solver makedirs CNN str join format load_model evaluate print close Solver NeuralNetwork plot_loss Saver info reset_default_graph Logistic Session join sess format save info join restore format basename sess print get_checkpoint_state model_checkpoint_path info show str join asarray arange plot model transpose set title savefig legend append range print_activations as_list print_activations range res_block print_activations print_activations print_activations print_activations print_activations print_activations print_activations maximum print_activations sqrt as_list format name print info get_shape trainable_variables format print info prod enumerate show format subplots reshape set_xlabel set_yticks subplots_adjust imshow set_xticks flat enumerate join format makedirs getLogger removeHandler close

# Adam-Analysis-TensorFlow This repository is an evaluation of Kingma's ["ADAM: A Method for Stochastic Optimization"](https://arxiv.org/pdf/1412.6980.pdf) adam optimizer and others. <p align="center"> <img src="https://user-images.githubusercontent.com/37034031/57378858-98b32100-71e0-11e9-92c9-62c20e9a167e.png" width=700> </p> ## Requirements - tensorflow 1.13.1 - pickleshare 0.7.4 - numpy 1.15.2 - unicodecsv 0.14.1

ChienYiChi/kaggle-panda-challenge

['visual place recognition', 'image retrieval']

['NetVLAD: CNN architecture for weakly supervised place recognition']

src/pycls/core/net.py src/pycls/datasets/loader.py src/pycls/datasets/cifar10.py src/train.py src/model.py src/pycls/datasets/imagenet.py src/pycls/core/checkpoint.py src/pycls/core/benchmark.py src/pycls/models/resnet.py src/test.py src/pycls/core/plotting.py src/pycls/core/io.py src/preprocess.py src/pycls/models/anynet.py src/config.py src/pycls/models/regnet.py src/pycls/core/config.py src/generate_weights.py src/pycls/models/effnet.py src/pycls/core/logging.py src/pycls/core/builders.py src/modules.py src/utils.py src/engine.py src/pycls/core/trainer.py src/dataset.py src/pycls/core/timer.py src/pycls/datasets/transforms.py src/pycls/core/distributed.py src/pycls/core/optimizer.py src/pycls/core/meters.py blue_ratio_selection PANDADataset PANDADatasetTiles get_tiles_brs get_transforms get_tiles train_fn quadratic_weighted_kappa eval_fn OptimizedRounder compute_weights name_to_tile PANDADatasetTiles load_model_from_ckpt Resnext50Tiles AdaptiveConcatPool2d SEResNeXt EnetNetVLAD RegnetNetVLAD EnetV1 Regnet EfficientnetTiles Mish NetVLAD Resnext50 Resnext50wNetVLAD ResnetwNetVLAD MixNetVLAD EnetV2 AdaptiveConcatPool2d BasicHead EfficientModel AttentionHead ResnetModel QWKLoss GeM AttentionPool Flatten save_image_fix multiprocess_make_images_tiles crop_white save_crop_white train_val_split save_image images_to_tiles PandaDataset OptimizedRounder AdaptiveConcatPool2d BasicHead NetVLAD Model EfficientModel AttentionHead preprocess quadratic_weighted_kappa AttentionPool Flatten run seed_torch save_model GradualWarmupScheduler accuracy setup_logger save_dict_to_json compute_time_eval compute_time_full compute_time_train compute_time_loader get_loss_fun build_model register_model build_loss_fun register_loss_fun get_model get_checkpoint load_checkpoint get_last_checkpoint get_checkpoint_dir save_checkpoint has_checkpoint load_cfg load_cfg_fom_args assert_and_infer_cfg cache_cfg_urls dump_cfg is_master_proc init_process_group ChildException ErrorHandler multi_proc_run scaled_all_reduce destroy_process_group run cache_url download_url _progress_bar dump_log_data float_to_decimal sort_log_data _suppress_print get_log_files get_logger load_log_data setup_logging topk_errors time_string ScalarMeter TrainMeter gpu_mem_usage TestMeter drop_connect reset_bn_stats complexity_conv2d compute_precise_bn_stats set_flat_weights complexity init_weights get_flat_weights complexity_batchnorm2d complexity_maxpool2d set_lr lr_fun_steps lr_fun_cos construct_optimizer lr_fun_exp get_epoch_lr get_lr_fun get_plot_colors prepare_plot_data plot_error_curves_pyplot plot_error_curves_plotly Timer train_model test_model train_epoch setup_env test_epoch setup_model time_model Cifar10 ImageNet shuffle construct_test_loader construct_train_loader _construct_loader random_crop zero_pad random_sized_crop horizontal_flip center_crop color_norm scale lighting BasicTransform get_stem_fun ResStemCifar ResStemIN AnyStage get_block_fun SE ResBottleneckBlock SimpleStemIN ResBasicBlock BottleneckTransform VanillaBlock AnyNet AnyHead Swish StemIN SE MBConv EffStage EffNet EffHead adjust_ws_gs_comp RegNet quantize_float get_stages_from_blocks generate_regnet BasicTransform ResBlock get_trans_fun ResStemCifar ResStemIN ResHead ResNet BottleneckTransform ResStage shape pad reshape shape pad reshape Compose model zero_grad set_description accumulation_steps step append to predict apex concatenate item info quadratic_weighted_kappa enumerate backward add_scalar tqdm loss_fn coefficients train numpy fit num_class argmax round OptimizedRounder squeeze to sum range predict eval softmax info quadratic_weighted_kappa enumerate extend accuracy tqdm loss_fn numpy array add_scalar load eval EfficientModel load_state_dict to append str range concatenate nonzero join imwrite tqdm COLOR_RGB2BGR mkdir listdir MultiImage cvtColor join imwrite tqdm COLOR_RGB2BGR imread listdir cvtColor join crop_white imwrite tqdm mkdir listdir MultiImage int StratifiedKFold to_csv split head read_csv enumerate join str shape mkdir range imsave get_tiles get int str glob print cpu_count apply_async mkdir append Pool range len shape pad reshape save_model fold zero_grad num_epoch DataLoader DataParallel ckpt_path device DEBUG initialize OptimizedRounder multi_gpu RegnetNetVLAD step PANDADatasetTiles Adam eval_fn load_state_dict to head range SummaryWriter apex MODEL_PATH PANDADataset GradualWarmupScheduler setup_logger copy info CosineAnnealingLR tile_stats_csv load seed_torch train_fn index parameters fold_csv read_csv makedirs setFormatter addHandler min stderr StreamHandler info setLevel FileHandler module isinstance seed str manual_seed list logical_and astype range sum equal toc model synchronize WARMUP_ITER NUM_ITER eval reset tic Timer cuda range toc loss_fun model backward synchronize WARMUP_ITER NUM_ITER reset tic zip train cuda range toc min WARMUP_ITER shuffle NUM_ITER reset tic iter Timer next range len format compute_time_train compute_time_eval dump_log_data compute_time_loader info max format get_checkpoint_dir get_checkpoint_dir get_checkpoint_dir get_checkpoint save makedirs load load_state_dict cache_cfg_urls cache_url DOWNLOAD_CACHE WEIGHTS OUT_DIR join CFG_DEST merge_from_file join merge_from_file cfg_file add_argument exit merge_from_list print_help ArgumentParser opts parse_args set_device wait NUM_GPUS all_reduce mul_ append fun init_process_group Process pid join add_child SimpleQueue fun ErrorHandler start append randint range format replace download_url dirname info exists makedirs int format write float round flush Request urlopen strip int is_master_proc join stdout basicConfig OUT_DIR _suppress_print float_to_decimal dumps isinstance sorted format zip append keys zip int divmod topk t eq expand_as max max_memory_allocated ZERO_INIT_FINAL_GAMMA isinstance fill_ out_channels Conv2d normal_ zero_ BatchNorm2d Linear int batch_size model min NUM_GPUS islice scaled_all_reduce cuda enumerate NUM_SAMPLES_PRECISE len reset_running_stats BatchNorm2d modules isinstance IM_SIZE empty div_ mul_ bernoulli_ data numel copy_ parameters view_as USE_CUSTOM_WEIGHT_DECAY parameters LR_POLICY WARMUP_EPOCHS WARMUP_FACTOR param_groups to_rgb interp append format zip str list plot get_plot_colors Layout prepare_plot_data zip append Scatter enumerate len show plot xlabel len grid ylabel get_plot_colors title clf prepare_plot_data savefig legend enumerate is_master_proc seed format RNG_SEED manual_seed OUT_DIR makedirs dump_log_data info dump_cfg BENCHMARK setup_logging format build_model dump_log_data complexity DistributedDataParallel cuda info current_device topk_errors loss_fun model zero_grad iter_tic set_lr log_iter_stats iter_toc get_epoch_lr update_stats scaled_all_reduce size NUM_GPUS shuffle log_epoch_stats enumerate backward reset train step log_iter_stats topk_errors iter_toc model size NUM_GPUS update_stats log_epoch_stats eval reset scaled_all_reduce enumerate iter_tic construct_test_loader WEIGHTS compute_time_full save_checkpoint TestMeter cuda compute_precise_bn_stats get_last_checkpoint test_epoch range USE_PRECISE_STATS format construct_optimizer TrainMeter setup_env info MAX_EPOCH load_checkpoint train_epoch setup_model construct_train_loader len construct_test_loader format load_checkpoint WEIGHTS setup_env test_epoch setup_model info TestMeter len construct_test_loader setup_env setup_model compute_time_full construct_train_loader cuda join DataLoader format format isinstance sampler set_epoch type range randint zero_pad int float floor resize ceil int int sqrt uniform resize float round range normal repeat range sum list tolist zip arange divide power round log

ChienYiChi/kaggle-panda-challenge

ChingtingC/Code-Switching-Sentence-Generation-by-GAN

['data augmentation']

['Code-switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation']

train.py build_model.py tool/tag_pos.py generate.py utils.py tool/calculate_cs_rate.py main GAN train_for_n max_action translate_output2 make_trainable translate write_log evaluate_acc get_action str2bool translate_output EMBEDDING_POS generator HIDDEN_SIZE_G DROPOUT_RATE add_argument GAN NOISE_SIZE summary HIDDEN_SIZE_D discriminator ArgumentParser HIDDEN_SIZE_L parse_args EMBEDDING_SIZE MAX_SEQUENCE_LENGTH make_trainable random save_weights translate_output train_on_batch list pad_sequences append cut range predict normal concatenate write_log zeros reshape translate tqdm randint get_action len layers append argmax asarray transpose to_categorical choice append pad_sequences texts_to_sequences len append enumerate append enumerate append enumerate add_summary flush Summary add write enumerate

# Code switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation Ching-Ting Chang, Shun-Po Chuang, Hung-Yi Lee [Interspeech 2019](https://www.isca-speech.org/archive/Interspeech_2019/pdfs/3214.pdf) [arXiv:1811.02356](https://arxiv.org/abs/1811.02356) ## Abstract Code-switching is about dealing with alternative languages in speech or text. It is partially speaker-depend and domain-related, so completely explaining the phenomenon by linguistic rules is challenging. Compared to most monolingual tasks, insufficient data is an issue for code-switching. To mitigate the issue without expensive human annotation, we proposed an unsupervised method for code-switching data augmentation. By utilizing a generative adversarial network, we can generate intra-sentential code-switching sentences from monolingual sentences. We applied proposed method on two corpora, and the result shows that the generated code-switching sentences improve the performance of code-switching language models. ## Outline 1. Introduction 2. Methodology 3. Experimental setup

ChingtingC/Code-switching-Sentence-Generation-by-Generative-Adversarial-Networks-and-its-Application-to-Data-Au

['data augmentation']

['Code-switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation']

train.py build_model.py tool/tag_pos.py generate.py utils.py tool/calculate_cs_rate.py main GAN train_for_n max_action translate_output2 make_trainable translate write_log evaluate_acc get_action str2bool translate_output EMBEDDING_POS generator HIDDEN_SIZE_G DROPOUT_RATE add_argument GAN NOISE_SIZE summary HIDDEN_SIZE_D discriminator ArgumentParser HIDDEN_SIZE_L parse_args EMBEDDING_SIZE MAX_SEQUENCE_LENGTH make_trainable random save_weights translate_output train_on_batch list pad_sequences append cut range predict normal concatenate write_log zeros reshape translate tqdm randint get_action len layers append argmax asarray transpose to_categorical choice append pad_sequences texts_to_sequences len append enumerate append enumerate append enumerate add_summary flush Summary add write enumerate

# Code switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation Ching-Ting Chang, Shun-Po Chuang, Hung-Yi Lee [Interspeech 2019](https://www.isca-speech.org/archive/Interspeech_2019/pdfs/3214.pdf) [arXiv:1811.02356](https://arxiv.org/abs/1811.02356) ## Abstract Code-switching is about dealing with alternative languages in speech or text. It is partially speaker-depend and domain-related, so completely explaining the phenomenon by linguistic rules is challenging. Compared to most monolingual tasks, insufficient data is an issue for code-switching. To mitigate the issue without expensive human annotation, we proposed an unsupervised method for code-switching data augmentation. By utilizing a generative adversarial network, we can generate intra-sentential code-switching sentences from monolingual sentences. We applied proposed method on two corpora, and the result shows that the generated code-switching sentences improve the performance of code-switching language models. ## Outline 1. Introduction 2. Methodology 3. Experimental setup

ChrisLee63/person_search

['person search', 'person re identification', 'pedestrian detection']

['Joint Detection and Identification Feature Learning for Person Search']

lib/rpn/generate_anchors.py lib/models/network.py lib/utils/boxes.py lib/oim/unlabeled_matching_layer.py tools/demo.py lib/utils/config.py lib/oim/labeled_matching_layer.py lib/datasets/sampler.py lib/rpn/proposal_layer.py lib/rpn/anchor_target_layer.py lib/models/backbone.py tools/train_net.py lib/datasets/data_processing.py tools/test_net.py lib/datasets/psdb.py tools/_init_paths.py lib/utils/utils.py lib/rpn/proposal_target_layer.py lib/models/head.py lib/rpn/rpn_layer.py lib/utils/evaluate.py build_net_input img_preprocessing PSDB PSSampler Backbone Head Network LabeledMatching LabeledMatchingLayer UnlabeledMatchingLayer UnlabeledMatching AnchorTargetLayer generate_anchors scale_enum ratio_enum whctrs mkanchors ProposalLayer ProposalTargetLayer RPN clip_boxes bbox_transform bbox_overlaps bbox_transform_inv cfg_from_file merge_a_into_b evaluate_detections evaluate_search compute_iou smooth_l1_loss torch_rand_choice init_logger pickle unpickle parse_args visualize_result parse_args detect_and_exfeat exfeat parse_args add_path Tensor imread empty img_preprocessing resize MAX_SIZE min astype float32 transpose SCALE float max ANCHOR_SCALES ratio_enum ANCHOR_RATIOS Tensor cat unsqueeze_ cat sqrt round mkanchors whctrs mkanchors whctrs stack log exp unsqueeze clone clamp_ clamp view items list ndarray isinstance type array merge_a_into_b min max asarray format info roidb compute_iou average_precision_score any zip append zeros argmax range str list squeeze add append range image_index format asarray astype set mean zip info enumerate reshape min extend int32 probes zeros ravel len sum size abs where join basicConfig install DATA_DIR makedirs add_argument ArgumentParser show format subplots replace text axis tight_layout add_patch close imshow savefig Rectangle zip imread list num_images tqdm append inference imread numpy range image_path_at list tqdm append inference imread numpy range len insert

# Person Search ## :sparkles: News: We release the [source code](https://github.com/serend1p1ty/SeqNet.git) of the current state-of-the-art model [SeqNet(AAAI 2021)](https://arxiv.org/abs/2103.10148), which achieves :trophy: `94.8%` mAP on CUHK-SYSU. ## Introduction A pytorch implementation for CVPR 2017 "Joint Detection and Identification Feature Learning for Person Search". The code is based on the [offcial caffe version](https://github.com/ShuangLI59/person_search.git). You can find a better one achieving about `85%` mAP in `mmdetection` branch! **Note**: The implementaion of Faster R-CNN in `mmdetection` branch is better than that described in original paper. ## Highlights - **Simpler code**: After reduction and refactoring, the current version is simpler and easier to understand. - **Pure Pytorch code**: Numpy is not used, except for data loading.

ChristianMarzahl/Exact

['whole slide images']

['EXACT: A collaboration toolset for algorithm-aided annotation of images with annotation version control']

exact/exact/images/migrations/0014_auto_20180629_2250.py exact/exact/annotations/api_views.py exact/exact/images/serializers.py exact/exact/annotations/migrations/0003_auto_20170826_1207.py exact/exact/annotations/models.py exact/exact/annotations/migrations/0010_auto_20170828_1628.py exact/exact/images/api_views.py exact/exact/annotations/migrations/0041_auto_20200403_1526.py exact/exact/datasets/admin.py exact/exact/annotations/migrations/0001_auto_20180508_1215.py exact/exact/annotations/migrations/0007_auto_20170826_1446.py exact/exact/annotations/migrations/0022_auto_20180425_1409.py exact/exact/base/management/commands/createdb.py exact/exact/tools/migrations/0003_auto_20171222_0302.py exact/exact/tagger_messages/migrations/0001_initial.py exact/exact/users/admin.py exact/exact/users/forms.py exact/exact/users/migrations/0003_auto_20210304_1530.py exact/exact/annotations/migrations/0004_auto_20170826_1211.py exact/exact/images/models.py exact/exact/tools/views.py exact/exact/images/migrations/0009_auto_20171122_1504.py exact/exact/images/migrations/0016_settag_test.py exact/exact/administration/migrations/0001_initial.py exact/exact/images/templates/images/download.py exact/exact/images/migrations/0023_image_image_type.py exact/exact/datasets/apps.py exact/exact/base/urls.py exact/exact/images/migrations/0010_imageset_public_collaboration.py exact/exact/users/urls.py exact/exact/annotations/migrations/0008_auto_20170826_1533.py exact/exact/annotations/migrations/0043_annotationversion.py exact/exact/datasets/forms.py exact/exact/tagger_messages/urls.py exact/exact/annotations/serializers.py exact/exact/annotations/migrations/0011_remove_export_export_type.py exact/exact/annotations/migrations/0032_annotationtype_imagefile.py exact/exact/administration/tests.py exact/exact/annotations/migrations/0001_initial.py exact/exact/users/migrations/0002_user_points.py exact/exact/tools/apps.py exact/exact/annotations/migrations/0034_auto_20191211_0951.py exact/exact/base/models.py exact/exact/base/tests.py exact/exact/images/migrations/0013_settag.py exact/exact/api.py exact/exact/annotations/migrations/0037_logimageaction_ip.py exact/exact/tools/forms.py exact/util/compare_files.py exact/exact/annotations/migrations/0038_annotation_unique_identifier.py exact/exact/images/migrations/0018_auto_20191203_1332.py exact/plugins/proc_EIPH.py exact/exact/datasets/urls.py exact/exact/images/migrations/0029_auto_20210304_1530.py exact/exact/images/apps.py exact/exact/images/migrations/0012_imageset_creator.py exact/exact/annotations/migrations/0016_auto_20171213_1115.py exact/exact/tagger_messages/views.py exact/exact/images/migrations/0005_auto_20170825_1129.py exact/exact/administration/apps.py exact/exact/annotations/fields.py exact/exact/annotations/migrations/0042_annotationmediafile.py exact/exact/users/apps.py exact/plugins/ExactServerPlugin.py exact/exact/users/models.py exact/exact/annotations/migrations/0025_points_trigger.py exact/exact/base/context_processors.py exact/exact/annotations/migrations/0002_auto_20180510_2310.py exact/exact/images/migrations/0006_auto_20170825_1148.py exact/exact/images/migrations/0022_auto_20191231_1419.py exact/exact/tools/admin.py exact/exact/annotations/migrations/0040_annotation_meta_data.py exact/exact/images/migrations/0003_auto_20170825_1012.py exact/exact/tagger_messages/management/commands/testdb.py exact/exact/annotations/migrations/0012_exportformat_not_in_image_format.py exact/exact/annotations/migrations/0013_exportformat_min_verifications.py exact/exact/images/migrations/0025_setversion.py exact/exact/users/management/commands/copyTeam.py exact/exact/tools/migrations/0002_tool_public.py exact/exact/images/migrations/0028_auto_20200611_0758.py exact/exact/annotations/migrations/0017_auto_20171220_1938.py exact/exact/users/tests.py exact/exact/annotations/migrations/0039_auto_20200218_1725.py exact/exact/annotations/migrations/0005_auto_20170826_1424.py exact/exact/annotations/migrations/0046_auto_20210304_1530.py exact/exact/images/management/commands/runzipdaemon.py exact/exact/tagger_messages/tests.py exact/exact/images/migrations/0015_imageset_pinned_by.py exact/exact/base/admin.py exact/exact/users/migrations/0001_initial.py exact/exact/annotations/apps.py exact/manage.py exact/exact/images/migrations/0021_screeningmode.py exact/exact/administration/api_views.py exact/exact/annotations/migrations/0036_logimageaction.py exact/exact/administration/models.py exact/exact/tools/models.py exact/exact/annotations/migrations/0028_annotationtype_color_code.py exact/exact/users/serializers.py exact/exact/annotations/views.py exact/exact/annotations/migrations/0009_auto_20170826_1535.py exact/exact/images/migrations/0017_imageset_zip_state.py exact/exact/base/views.py exact/exact/annotations/migrations/0031_auto_20191205_1039.py exact/exact/annotations/migrations/0035_auto_20191211_1255.py exact/exact/images/migrations/0027_setversion_file.py exact/exact/tagger_messages/admin.py exact/exact/annotations/forms.py exact/exact/images/migrations/0020_auto_20191209_1341.py exact/exact/annotations/admin.py exact/exact/annotations/migrations/0020_auto_20180417_1220.py exact/exact/images/forms.py exact/exact/images/migrations/0002_auto_20170822_1159.py exact/exact/annotations/migrations/0023_auto_20180428_1756.py exact/util/cellvizio.py exact/exact/images/migrations/0024_auto_20200403_1526.py exact/exact/tagger_messages/forms.py exact/exact/tools/tests.py exact/exact/annotations/migrations/0044_auto_20200501_1354.py exact/util/ISyntaxContainer.py exact/exact/administration/serializers.py exact/exact/annotations/migrations/0014_auto_20170907_1407.py exact/exact/images/urls.py exact/exact/tools/migrations/0001_initial.py exact/exact/tagger_messages/models.py exact/exact/images/views.py exact/exact/administration/admin.py exact/exact/annotations/migrations/0015_auto_20171129_1511.py exact/exact/annotations/migrations/export_format_conversion_20180504.py exact/exact/images/migrations/0001_initial.py exact/exact/tagger_messages/apps.py exact/exact/images/admin.py exact/exact/annotations/migrations/0029_auto_20191121_0936.py exact/exact/images/migrations/0004_auto_20170825_1114.py exact/exact/images/tests.py exact/exact/datasets/views.py exact/exact/datasets/tests.py exact/exact/annotations/migrations/0019_auto_20180314_0922.py exact/exact/images/migrations/0019_imageset_collaboration_type.py exact/exact/datasets/models.py exact/exact/urls.py exact/exact/tools/urls.py exact/exact/images/migrations/0007_auto_20170830_1626.py exact/exact/base/apps.py exact/exact/administration/forms.py exact/exact/annotations/migrations/0033_auto_20191211_0946.py exact/exact/images/migrations/0008_auto_20171120_1056.py exact/exact/annotations/migrations/0027_auto_20181114_1058.py exact/exact/annotations/migrations/0002_auto_20170822_1159.py exact/util/convert_slides.py exact/exact/users/views.py exact/exact/settings_base.py exact/exact/annotations/urls.py exact/exact/annotations/migrations/0018_auto_20171220_2020.py exact/util/slide_server.py exact/exact/annotations/tests.py exact/exact/annotations/migrations/0021_exportformat_vector_format.py exact/exact/administration/views.py exact/exact/administration/urls.py exact/exact/images/migrations/0011_auto_20180507_1830.py exact/plugins/pluginFinder.py exact/exact/annotations/migrations/0045_auto_20200611_0758.py exact/exact/wsgi.py exact/exact/users/management/commands/updatepoints.py exact/exact/images/migrations/0026_setversion_time.py exact/exact/annotations/migrations/0006_auto_20170826_1431.py exact/exact/annotations/migrations/0024_auto_20180429_0005.py exact/exact/annotations/migrations/0030_annotationtype_sort_order.py exact/exact/users/api_views.py exact/exact/annotations/migrations/0026_index_verification_time.py handler500 ProductViewset AdministrationConfig AnnotationTypeCreationForm AnnotationTypeEditForm ProductCreationForm ProductEditForm Product product_changed_handler ProductSerializer serialize_annotationType products product edit_annotation_type create_annotation_type edit_product create_product api_delete_annotation_type api_create_annotation_type migrate_bounding_box_to_0_polygon api_filter_products migrate_bounding_box_to_4_polygon annotation_type delete_annotation_type api_create_product annotation_types Migration VerificationAdmin AnnotationAdmin AnnotationTypeViewSet LogImageActionFilterSet AnnotationViewSet AnnotationMediaFileViewSet VerificationViewSet AnnotationVersionFilterSet LogImageActionViewSet AnnotationVersionViewSet AnnotationFilterSet AnnotationMediaFileFilterSet AnnotationsConfig NonStrippingTextField ExportFormatEditForm AnnotationMediafileForm ExportFormatCreationForm AnnotationType AnnotationMediaFile Annotation AnnotationVersion Export LogImageAction ExportFormat annotation_type_changed_handler Verification annotation_directory_path AnnotationQuerySet AnnotationSerializerFast AnnotationMediaFileSerializer VerificationSerializer AnnotationVersionSerializer AnnotationSerializer serialize_annotation LogImageActionSerializer AnnotationTypeSerializer AnnotationCreationTestCase load_annotation_types create_exportformat api_copy_annotation load_set_annotation_types api_update_annotation_mediafile load_filtered_set_annotations create_export delete_annotations api_delete_annotation_mediafile download_export update_annotation annotate_set delete_annotation export_format api_verify_annotation api_blurred_concealed_annotation apply_conditional edit_exportformat delete_exportformat verify load_annotations annotate manage_annotations api_delete_annotation create_annotation export_auth api_create_annotation_mediafile load_annotation load_set_annotations Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration create_uuid Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration convert_export_formats BaseConfig base_data report_time problem_report index Command DatasetsConfig DatasetForm MITOS_WSI_CMCDatasetForm create_miccai_eiph_dataset create_miccai_asthma_dataset create_eiph_dataset index create_miccai_mitotic_dataset create_mitos_wsi_cmc_dataset create_asthma_dataset ScreeningModeViewSet ImageFilterSet ImageRegistrationViewSet ImageSetFilterSet ImageViewSet SetTagViewSet ImageSetViewSet SetVersionViewSet ImagesConfig ImageSetEditForm ImageSetCreationFormWT ImageSetCreationForm LabelUploadForm CopyImageSetForm SetVersion SetTag ImageSet Image ImageRegistration registration_directory_path imageset_changed_handler image_changed_handler ScreeningMode version_directory_path SetVersionSerializer SetTagSerializer ScreeningModeSerializer ImageRegistrationSerializer serialize_imageset ImageSetSerializer ImageSerializer ImageRegistrationTestCase image_plugins delete_images_api product_image_set image_opened sync_annotation_map create_imageset_api api_index view_image_tile image_statistics set_free remove_image_set_product load_image_set view_image_navigator_overlay_tile view_imageset navigator_overlay_status dl_script list_images label_upload delete_imageset tag_image_set create_imageset view_image edit_imageset download_imageset_zip api_verify_image explore_imageset copy_image copy_images_to_imageset remove_image_set_tag upload_image download_image_api autocomplete_image_set_tag delete_images index toggle_pin_imageset image_closed create_annotation_map Command Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration set_test_tag Migration set_mpp_and_power Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration Migration download_imageset TaggerMessagesConfig TeamMessageCreationForm GlobalMessageCreationForm GlobalMessage Message TeamMessage delete_message read_all_annoucements send_global_message overview_all overview_global_active send_team_message read_all_messages overview_sent_active overview_global_hidden read_message overview_unread overview_sent_hidden Command Migration ToolsConfig file_size ToolUploadForm FileUploadForm ToolVote Tool create_tool delete_tool download_tool overview edit_tool tools_enabled Migration Migration Migration TeamMembershipViewset UserViewset TeamViewset UsersConfig TeamCreationForm UserRegistrationForm TeamMembership add_user_to_team_handler User Team TeamMembershipSerializer TeamSerializer UserSerializer user add_team_member grant_team_admin view_team create_team explore_user user_autocomplete leave_team revoke_team_admin api_filter_teams explore_team Command Command Migration Migration Migration ExactServerPlugin UpdatePolicy NavigationViewOverlayStatus ViewPolicy pluginEntry PluginFinder Plugin ReadableCellVizioMKTDataset circularMask fileinfo DeepZoomImageTiler DeepZoomStaticTiler TileWorker ISyntaxContainer TIFF PILBytesIO SlideFile SlideCache ProductSerializer DateTimeField ForeignKey TextField CharField AUTH_USER_MODEL ManyToManyField delete_pattern hasattr all filter get_object_or_404 POST error is_valid success ProductCreationForm _ exists error get_object_or_404 _ save success first filter AnnotationTypeCreationForm filter AnnotationTypeCreationForm AnnotationTypeEditForm get_object_or_404 count delete get_object_or_404 get create get_object_or_404 ProductSerializer save first get int filter ProductSerializer get int get_object_or_404 POST error is_valid AnnotationTypeCreationForm success _ has_perm error delete get_object_or_404 success _ name FileSystemStorage error url get_object_or_404 _ save success first vector delete get_object_or_404 filter save POLYGON vector delete get_object_or_404 filter save POLYGON BooleanFilter NumberFilter RangeFilter AnnotationSerializer BooleanFilter AnnotationVersionSerializer AnnotationTypeSerializer ChoiceFilter AnnotationMediaFileSerializer VerificationSerializer ChoiceFilter LogImageActionSerializer JSONField BooleanField ForeignKey UUIDField DateTimeField ForeignKey TextField ManyToManyField AUTH_USER_MODEL JSONField BooleanField ImageField ForeignKey IntegerField CharField BooleanField delete_pattern hasattr all DateTimeField TextField IntegerField ForeignKey AUTH_USER_MODEL AUTH_USER_MODEL DateTimeField BooleanField ForeignKey DateTimeField ForeignKey IntegerField NonStrippingTextField CharField BooleanField ManyToManyField GenericIPAddressField DateTimeField ForeignKey IntegerField AUTH_USER_MODEL now IntegerField FileField CharField ForeignKey BooleanField SerializerMethodField AnnotationTypeSerializer SerializerMethodField ImageSerializer is_authenticated user order_by get exclude hasattr timer get_object_or_404 set filter render info get_perms image_lock first count user has_perm delete get_object_or_404 user get has_perm str replace id get_object_or_404 export_format save Export format get_object_or_404 export_text filename HttpResponse get order_by Q get_object_or_404 filter get_page select_related bool date Paginator get user has_perm str Q print delete get_object_or_404 filter warning date count error filter get_object_or_404 select_related warning get_object_or_404 replace blurred image_format annotation_format concealed str list not_in_image range name_format not_in_image_format image_aggregation exclude replace apply_conditional vector_format base_format select_related vector items min_verifications filter len get POST ExportFormatCreationForm print is_valid add_error filter _ success exists POST error is_valid ExportFormatEditForm add_error get_object_or_404 success _ error delete get_object_or_404 success _ AnnotationMediaFile list AnnotationMediaFileSerializer get_object_or_404 save append first values get int delete AnnotationSerializer get_object_or_404 select_related bool save first id get_object_or_404 get int deleted get_object_or_404 save first get int Response get_object_or_404 filter bool get int order_by AnnotationSerializer get_object_or_404 filter bool order_by int AnnotationTypeSerializer get_object_or_404 get int order_by get_object_or_404 filter AnnotationTypeSerializer bool get int sorted list AnnotationSerializer get_object_or_404 filter select_related bool int AnnotationSerializer get_object_or_404 get int update get_object_or_404 bool user int get_object_or_404 verify serialize_annotation exists int get_object_or_404 uuid4 get_model all save str list items all replace save get_model is_authenticated filter SHOW_DEMO_DATASETS ModelChoiceField GenericIPAddressField CharField ChoiceField MultipleChoiceField DatasetForm MITOS_WSI_CMCDatasetForm POST MITOS_WSI_CMCDatasetForm is_valid get_object_or_404 mkdir POST is_valid get_object_or_404 DatasetForm mkdir POST is_valid get_object_or_404 DatasetForm mkdir POST is_valid get_object_or_404 DatasetForm mkdir str POST is_valid get_object_or_404 DatasetForm mkdir download str POST is_valid get_object_or_404 DatasetForm download BooleanFilter NumberFilter RangeFilter ChoiceFilter ImageSerializer ModelChoiceFilter ChoiceFilter ImageSetSerializer SetTagSerializer SetVersionSerializer ScreeningModeSerializer ImageRegistrationSerializer FileField BooleanField BooleanField ModelMultipleChoiceField BinaryField DateTimeField ForeignKey IntegerField CharField FloatField delete_pattern hasattr DateTimeField ForeignKey TextField IntegerField CharField AUTH_USER_MODEL BooleanField ManyToManyField delete_pattern hasattr CharField ManyToManyField FileField DateTimeField CharField ManyToManyField AUTH_USER_MODEL JSONField IntegerField ForeignKey ForeignKey IntegerField FileField JSONField FloatField order_by get str replace PAGE_SIZE filter get_page ImageSetViewSet GET Paginator split get filter order_by ImageSetCreationFormWT filter first TeamCreationForm Image extractall id root_path capitalize Path values str list seek name get_object_or_404 save_file append update rfind close digest ZipFile first join read remove sha512 sort chunks any get join hasattr get_dzi get_object_or_404 IMAGE_PATH path set get int user getPluginStatisticsElements all get_object_or_404 loads append filter_plugins int all get_object_or_404 updateNavigationViewOverlay getNavigationViewOverlayStatus filter_plugins get int join all get_tile search group getNavigationViewOverlay get_object_or_404 IMAGE_PATH path PILBytesIO getvalue save filter_plugins HttpResponse get int join hasattr get_tile search group timer get_object_or_404 IMAGE_PATH path PILBytesIO getvalue set save info get_object_or_404 get_object_or_404 get_object_or_404 get str format name FileResponse get_object_or_404 IMAGE_PATH path original_path Path getsize open remove delete original_path get_object_or_404 path exists get remove delete original_path get_object_or_404 path exists get user order_by ImageSetEditForm exclude Q get_unverified_ids get_object_or_404 filter warning annotate CopyImageSetForm annotate int get_object_or_404 get format name get_object_or_404 warning serialize_imageset format POST name is_valid add_error get_object_or_404 ImageSetCreationForm warning _ success exists data Image tiffsave delete default_height id IMAGE_PATH warning distinct save resize max min_x count str list default_width name max_y get_object_or_404 shape width ceil range order_by get format height new_from_memory sqrt max_x min_y first _ pop int vector join open_slide reshape values_list path zeros len format exclude name delete id get_object_or_404 filter warning annotation_type save last_editor _ ImageSetEditForm POST is_valid get_object_or_404 warning save _ delete root_path get_object_or_404 rmtree warning _ warning _ save get_object_or_404 user remove format name add get_object_or_404 save info api_copy_annotation id get_object_or_404 root_path symlink filter path save filename first exclude getlist copy_image id get_object_or_404 warning _ user list format validate_vector replace exists error Annotation get_object_or_404 filter loads warning save append success Verification _ split join format zip_path zip_name HttpResponse FileResponse get_object_or_404 IMAGE_PATH USE_NGINX_IMAGE_PROVISION getsize open user int Q get_object_or_404 verify filter get int data user get_unverified_ids get_object_or_404 get_verified_ids serialize_imageset int exists add get_object_or_404 ProductSerializer save first data int remove get_object_or_404 ProductSerializer save get int SetTag replace SetTagSerializer get_object_or_404 add lower save exists data int remove SetTagSerializer delete get_object_or_404 lower save list Q print extend lower filter get SetTag all add save get_model alias exists join all float IMAGE_PATH path OpenSlide save filename get_model alias get join format replace print getcwd post split append enumerate makedirs DateTimeField ForeignKey TextField CharField AUTH_USER_MODEL ManyToManyField BooleanField ForeignKey BooleanField error TeamMessageCreationForm POST error is_valid POST GlobalMessageCreationForm get user add get filter add get filter in_range add is_staff delete get MESSAGES_PER_PAGE TeamMessageCreationForm filter get_page Paginator user get MESSAGES_PER_PAGE TeamMessageCreationForm in_range get_messages_for_user filter get_page Paginator get MESSAGES_PER_PAGE TeamMessageCreationForm in_range filter get_page Paginator get MESSAGES_PER_PAGE TeamMessageCreationForm not_in_range filter get_page Paginator get user MESSAGES_PER_PAGE in_range get_page GlobalMessageCreationForm exists Paginator get user MESSAGES_PER_PAGE not_in_range get_page GlobalMessageCreationForm exists Paginator FileField FileField DateTimeField ForeignKey TextField CharField AUTH_USER_MODEL BooleanField AUTH_USER_MODEL DateTimeField BooleanField ForeignKey filter ToolUploadForm distinct ToolUploadForm format POST name is_valid error TOOLS_PATH instance id save success FILES makedirs join remove POST TOOLS_PATH name get_object_or_404 warning save filename FILES FileUploadForm user join has_perm remove TOOLS_PATH error delete get_object_or_404 filename exists user join has_perm TOOLS_PATH error get_object_or_404 filename exists UserSerializer TeamSerializer TeamMembershipSerializer IntegerField create hasattr filter ADD_USER_TO_TEAM save AUTH_USER_MODEL CharField ManyToManyField AUTH_USER_MODEL BooleanField ForeignKey is_valid POST TeamCreationForm user update has_perm format name get_object_or_404 warning update format name get_object_or_404 warning get str all filter get_page Paginator order_by str get filter get_page Paginator user int delete get_object_or_404 filter warning save _ get create format exists name get_object_or_404 warning info success first order_by ExportFormatEditForm get_object_or_404 admins filter append filter get_object_or_404 list Q print extend lower filter get int filter TeamSerializer fileinfo

# Exact [](https://pypi.python.org/pypi/EXCAT-Sync/) [](https://lbesson.mit-license.org/) This is a collaborative online tool for labeling image data.  ## Reference This paper describes the EXACT-Server in depth. Please cite if you use this tool in your research: Marzahl et al. [EXACT: A collaboration toolset for algorithm-aided annotation of almost everything](https://www.nature.com/articles/s41598-021-83827-4) ``` @Article{marzahl2021exact,

ChristophReich1996/Cell-DETR

['cell segmentation', 'instance segmentation', 'experimental design', 'semantic segmentation']

['Attention-Based Transformers for Instance Segmentation of Cells in Microstructures']

lossfunction.py pixel_adaptive_convolution/pac.py dataset.py pixel_adaptive_convolution/tools/flowlib.py misc.py validation_metric.py model_wrapper.py main.py pade_activation_unit/utils.py detr.py matcher.py pade_activation_unit/cuda/python_imp/Pade.py pade_activation_unit/cuda/setup.py pixel_adaptive_convolution/tools/plot_log.py bounding_box_head.py pixel_adaptive_convolution/paccrf.py augmentation.py pixel_adaptive_convolution/test_pac.py transformer.py pade_activation_unit/torchsummary.py backbone.py segmentation.py NoiseInjection Augmentation VerticalFlip ElasticDeformation ResNetBlock StandardBlock DenseNetBlock Backbone BoundingBoxHead CellInstanceSegmentation collate_function_cell_instance_segmentation CellDETR LovaszHingeLoss FocalLossMultiClass InstanceSegmentationLoss ClassificationLoss LovaszSoftmaxLoss SegmentationLoss BoundingBoxLoss BoundingBoxGIoULoss DiceLoss MultiClassSegmentationLoss FocalLoss HungarianMatcher absolute_bounding_box_to_relative normalize_0_1 plot_instance_segmentation_overlay_instances bounding_box_x0y0x1y1_to_xcycwh plot_image plot_instance_segmentation_instances plot_instance_segmentation_overlay_instances_bb_classes plot_instance_segmentation_overlay_bb_classes giou_for_matching iterable_to_device bounding_box_xcycwh_to_x0y0x1y1 Logger to_one_hot relative_bounding_box_to_absolute normalize plot_instance_segmentation_map_label giou plot_instance_segmentation_labels ModelWrapper FinalBlockReshaped MultiHeadAttention FinalBlock SegmentationHead ResPACFeaturePyramidBlock ResFeaturePyramidBlock Transformer TransformerDecoderLayer _get_clones TransformerDecoder build_transformer TransformerEncoder TransformerEncoderLayer _get_activation_fn F1 BoundingBoxGIoU Recall ClassificationAccuracy MeanAveragePrecision MIoU Accuracy IoU Precision InstancesAccuracy BoundingBoxIoU Dice CellIoU summary Swish PAU Swish_module activationfunc generate_cpp_kernels_module generate_cpp_module exec_act PADEACTIVATION_F_python PADEACTIVATION PADEACTIVATION_Function_based get_constants_for_inits test_v2 PADEACTIVATION_F_cpp PADEACTIVATION_F_abs_cpp PacPool2d GaussKernel2dFn PacPool2dFn _PacConvNd PacConv2dFn PacConvTranspose2dFn packernel2d PacConv2d PacConvTranspose2d np_gaussian_2d pacconv_transpose2d _neg_idx pacpool2d nd2col pacconv2d create_YXRGB PacCRFLoose create_position_feats _ceil_pad_factor PacCRF use_only_custom_impl use_only_native_impl repeat_impl_types PacConvTest _allclose _gradcheck make_color_wheel read_flow scale_image flow_error evaluate_flow flow_to_image write_flow compute_color show_flow smooth_plot parse_and_plot remove_common_prefix_suffix zeros long scatter_ unbind unbind bounding_box_xcycwh_to_x0y0x1y1 bounding_box_xcycwh_to_x0y0x1y1 gray2rgb subplots normalize_0_1 where set_visible show imshow savefig range close float enumerate set_size_inches text set_axis_off add_patch Rectangle bool numpy show set_size_inches gray2rgb subplots normalize_0_1 set_axis_off close where imshow set_visible savefig bool numpy range enumerate subplots set_visible show alltrue imshow savefig range close zip float enumerate set_size_inches text set_axis_off add_patch Rectangle zeros bool numpy array show set_size_inches subplots set_axis_off close alltrue imshow set_visible savefig zip zeros bool numpy array range show set_size_inches subplots normalize_0_1 set_axis_off close imshow set_visible save_image numpy detach show set_size_inches subplots gray2rgb normalize_0_1 float text set_axis_off close add_patch imshow set_visible savefig Rectangle bool numpy enumerate subplots set_visible show alltrue imshow savefig range format replace close zip enumerate set_size_inches reshape set_axis_off repeat bool numpy array clamp min max clamp min max to range len str remove format isinstance FloatTensor model print apply OrderedDict lower numpy abs prod config_cuda locals Template merge locals Template merge RandomState time format actv backward print synchronize range seed manual_seed randn exec_act device to reshape float exp arange conv_transpose2d im2col tuple contiguous unfold pad new_ones _pair len minimum sum view relu exp_ tuple contiguous apply shape conv2d new_ones pow startswith mul_ tensor nd2col _pair detach tuple apply nd2col einsum _pair conv_transpose2d tuple pacconv2d apply pad new_ones permute _pair nd2col apply shape sum _pair view from_numpy meshgrid tensor to array view device tensor create_position_feats cat tuple flow_error read_flow flow_to_image imshow show read_flow print reshape close float32 fromfile open tofile close dstack shape zeros array open mean sqrt eps print min sqrt repeat compute_color max uint8 arctan2 size astype pi logical_not isnan shape sqrt floor zeros make_color_wheel range zeros transpose floor arange min astype float32 array max int convolve plot repeat len all genfromtxt subplots tuple grid show sorted set_title smooth_plot set_xlabel savefig delaxes legend ceil append range flat tight_layout mean sqrt zip enumerate int switch_backend reshape min any len

# Cell-DETR: Attention-Based Transformers for Instance Segmentation of Cells in Microstructures [](https://arxiv.org/abs/2011.09763) [](https://github.com/ChristophReich1996/Cell-DETR/blob/master/LICENSE) **[Tim Prangemeier](https://www.bcs.tu-darmstadt.de/bcs_team/prangemeiertim.en.jsp), [Christoph Reich](https://github.com/ChristophReich1996) & [Heinz Koeppl](https://www.bcs.tu-darmstadt.de/bcs_team/koepplheinz.en.jsp)** This repository includes the **official** and **maintained** implementation of the paper **[Attention-Based Transformers for Instance Segmentation of Cells in Microstructures](https://www.bcs.tu-darmstadt.de/media/bcs/research_4/documents/Prangemeier2020c_BIBM_cell-detr_accepted.pdf)** [(BIBM 2020)](https://ieeebibm.org/BIBM2020/). <img src="images/test_plot_1_is.png" alt="drawing" width="150"/><img src="images/test_plot_4_is.png" alt="drawing" width="150"/><img src="images/test_plot_12_is.png" alt="drawing" width="150"/><img src="images/test_plot_29_is.png" alt="drawing" width="150"/><img src="images/test_plot_31_is.png" alt="drawing" width="150"/> <br/> ## Abstract *Detecting and segmenting object instances is a common task in biomedical applications. Examples range from detecting

ChunpingQiu/Human-settlement-extent-detection-from-Sentinel-2-images-via-fully-convolutional-neural-networks-

['semantic segmentation']

['A framework for large-scale mapping of human settlement extent from Sentinel-2 images via fully convolutional neural networks']

modelSelection.py dataGener.py train.py img2mapC.py dataPre.py sen2IS_net.py img2map.py DataGenerator traValFiles_spatial traValFiles_random img2mapC recall_m modelSelection f1_m dice_coef precision_m sen2IS_net_bn sen2IS_net_bn_core sen2IS_net_deep sen2IS_net_bn_core_2 unet sen2IS_net_wide str arange print loadmat int32 append sum glob sum loadmat print sen2IS_net_bn insert resnet50_pspnet build_model_resFCN sen2IS_net_deep fcn_8_resnet50 unet summary sen2IS_net_wide compile sum epsilon subtract floatx cast round argmax clip sum epsilon subtract floatx cast round argmax clip recall_m precision_m sum subtract flatten floatx cast argmax print print sen2IS_net_bn_core Input Model sen2IS_net_bn_core_2 sen2IS_net_bn_core Input Model sen2IS_net_bn_core_2 int arange print Model Input Model Input

# Use the trained model to predict HSE from S2 images ## Software Requirements environment.yml ## use demo - put the image data (s2) into folder: ./data/img/ - run python img2map.py - prediciton will be in folder ./data/pre/ ## reference data https://drive.google.com/drive/folders/1n2LGeGAv_O2cvxAJnSGNRUI4FMsm4psa?usp=sharing ## current status

ClancyZhou/P_Net_Anomaly_Detection

['anomaly detection']

['Encoding Structure-Texture Relation with P-Net for Anomaly Detection in Retinal Images']

utils/gan_loss.py utils/parser.py utils/utils.py train_structure_extraction_network.py train_P_Net.py networks/unet.py networks/unet_part.py networks/discriminator.py networks/P_Net_v1.py dataloader/resc_dataloader.py utils/visualizer.py MultiTestForFigures RunMyModel PNetModel SegTransferModel RunMyModel RandomEnhance OCT_ClsTrainSet SourceOCTdataset ChallengeOCTloader ChanllengeOCTdataset OCT_ClsDataloader SourceOCTloader OCT_ClsDataset BaseNetwork ResnetBlock Discriminator InpaintGenerator EdgeGenerator Strcutre_Extraction_Network Image_Reconstruction_Network UNet_4mp Seg_UNet_Rec_AE RecAE_4mp RecAE_5mp UNet_5mp main outconv up double_conv fusion_down up_wo_skip down inconv_single_conv down_single_conv inconv AdversarialLoss ParserArgs AverageMeter cuda_visible calc_confidence_interval l1_reg print_args adjust_lr main LastAvgMeter save_ckpt main Visualizer Contrast Brightness uniform randint Sharpness enhance param_groups pow index parameters join format save split output_root project makedirs format warn print format asarray interval len colour Visualizer

# P-Net for Anomaly Detection (Pytorch) This is the implementation of the paper: Kang Zhou, Yuting Xiao, Jianlong Yang, Jun Cheng, Wen Liu, Weixin Luo, Zaiwang Gu, Jiang Liu, Shenghua Gao. Encoding Structure-Texture Relation with P-Net for Anomaly Detection in Retinal Images. ECCV 2020. using PyTorch. **If you have any question, please feel free to contact us ({zhoukang, xiaoyt}@shanghaitech.edu.cn).** **The implementation on MvTec dataset could be found in https://github.com/YutingXiao/P-Net_Mvtec_AD** # Introduction  The motivation of leveraging structure information for anomaly detection. The normal medical images are highly structured, while the regular structure is broken in abnormal images. For example, the lesions (denoted by black bounding box and red arrow in (a) of diabetic retinopathy destroy the blood vessel and histology layer in retina. Thus, in the abnormal retinal fundus image and optical coherence tomography (OCT) image, the lesions (denoted by red color in (b) and (c)) broke the structure. Moreover, this phenomenon agrees with the cognition of doctors. Motivated by this clinical observation, we suggest utilizing the structure information in anomaly detection. # Method

ClorverCcy/GEDLoss_pytorch

['speech synthesis']

['A Spectral Energy Distance for Parallel Speech Synthesis']

spectral_ops.py torch_get_spectral_matrix torch_calc_spectrograms torch_hertz_to_mel torch_build_mel_basis torch_sum_spectral_dist torch_aligned_random_crop torch_matmul_real_with_complex torch_ged torch_mel_to_hertz shape torch_hertz_to_mel reshape view_as_complex torch_mel_to_hertz cos pi linspace sin float range matmul list torch_get_spectral_matrix torch_build_mel_basis torch_aligned_random_crop zip append range len sum torch_calc_spectrograms torch_sum_spectral_dist

# GEDLoss_pytorch a pytorch implementation of Google GEDLoss Full-text paper available on [arXiv](https://arxiv.org/abs/2008.01160). Origin code of TensorFlow edition at [GED_TTS](https://github.com/google-research/google-research/tree/68c738421186ce85339bfee16bf3ca2ea3ec16e4/ged_tts)

CoNexDat/mw2v

['word embeddings']

['Learning language variations in news corpora through differential embeddings']

NYT-TG/SimiLab.py tempName testClass

# mw2v Folders: Each of the folders corresponds to a different dataset. * NYT: files for New York Times newspaper dataset. Slices are years, from 1990 to 2016. * TG: files for The Guardian newspaper dataset. Slices are years, from 1999 to 2016. * NYT-TG: files for the two-source model, where one slice is the NYT and the other one is the TG, both during the 2010-2016 period. Files description: * Files starting with "delta_" and "mean_" contain the trained words embedding for each of the datasets. * Files starting with "slices_" contain the slices names for each dataset. * Files starting with "sampling_tables_" and "word_index_" are used to perform the negative/positive samplings during the parallel training of each slice embedding.

CodeAchieveDream/crnn_model

['optical character recognition', 'scene text recognition']

['An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition']

tolmdb.py imgaug_image.py alphabet.py dataset.py crnn_test.py recog.py params.py net/crnn.py utils.py crnn_train.py val trainBatch init_args training weights_init lmdbDataset alignCollate randomSequentialSampler resizeNormalize generate_image generate CRNN_MODEL createDataset writeCache init_args checkImageIsValid to_alphabet get_batch_label averager loadData prettyPrint assureRatio strLabelConverter oneHot compute_std_mean CRNN_VGG BidirectionalLSTM CRNN add_argument ArgumentParser normal_ __name__ fill_ data decode batchSize DataLoader IntTensor max crnn view add iter encode append next range averager loadData size eval zip float criterion print Variable min parameters len criterion backward Variable loadData size step zero_grad IntTensor encode next crnn save open str experiment add iter range state_dict val trainBatch format close print write parameters niter reset train len augment_image SomeOf generate frombuffer imdecode IMREAD_GRAYSCALE join list str replace print len close tqdm writeCache range open max fill_ size scatter_ long range str format print size type main size UpsamplingBilinear2d append astype float32 mean append zeros ravel std range

# Convolutional Recurrent Neural Network This software implements the Convolutional Recurrent Neural Network (CRNN), a combination of CNN, RNN and CTC loss for image-based sequence recognition tasks, such as scene text recognition and OCR. For details, please refer to our paper http://arxiv.org/abs/1507.05717. ### warp-CTC installation warp-CTC is a CTC code base of Baidu open source that can be applied to CPU and GPU efficiently and in parallel, and parallel processing of CTC algorithm. warp-CTC installation: ``` git clone https://github.com/SeanNaren/warp-ctc.git cd warp-ctc mkdir build; cd build cmake ..

Cold-Winter/Nattack

['adversarial attack']

['NATTACK: Learning the Distributions of Adversarial Examples for an Improved Black-Box Attack on Deep Neural Networks']

cascade_adv_training/evaluation.py cascade_adv_training/resnet.py robustnet/plot_accu.py inputtransformations/robustml_model.py cascade_adv_training/utils.py inputtransformations/robustml_model_origin.py lid/evaluation_bpda.py robustnet/main.py inputtransformations/quilt_preprocess.py sap/helpers.py cascade_adv_training/model.py guided-denoiser/helpers.py robustnet/models/vgg.py robustnet/models/resnet.py randomization/inceptionv3.py sap/evaluation.py inputtransformations/inceptionv3.py inputtransformations/re_l2_attack_clipimage.py robustnet/models/senet.py guided-denoiser/resnet.py robustnet/helpers.py sap/robustml_model.py pixel-deflection/utils.py guided-denoiser/re_li_attack_notanh.py wideresnet28/test_wres.py inputtransformations/utils.py therm-adv/evaluation.py guided-denoiser/resnext_features/resnext101_32x4d_features.py cascade_adv_training/cifar_load.py therm-adv/discretization_attacks.py randomization/re_li_attack299.py therm-adv/train.py guided-denoiser/re_li_attack.py robustnet/models/resnext.py robustnet/models/__init__.py guided-denoiser/resnext101.py robustnet/models/shufflenet.py sap/cifar10_input.py robustnet/models/preact_resnet.py lid/re_li_attack.py robustnet/models/densenet.py robustnet/models/googlenet.py therm-adv/discretization_utils.py cascade_adv_training/dataset.py randomization/provider.py wideresnet28/test.py guided-denoiser/resnext_features/resnext101_64x4d_features.py robustnet/re_li_attack.py therm-adv/helpers.py cascade_adv_training/re_li_attack_notanh.py cascade_adv_training/tf_rename_variables.py therm-adv/robustml_model.py therm-adv/re_li_attack_notanh.py sap/evaluation_bpda.py therm-adv/re_li_attack.py guided-denoiser/inception.py lid/detect_adv_samples.py randomization/imagenet_labels.py cascade_adv_training/provider.py robustnet/models/mobilenet.py guided-denoiser/provider.py pixel-deflection/main.py guided-denoiser/defense.py pixel-deflection/re_li_attack.py randomization/defense_batch.py robustnet/evaluation.py wideresnet28/evaluation.py randomization/robustml_attack.py cascade_adv_training/main.py sap/sap_model.py robustnet/main2.py robustnet/test.py guided-denoiser/resnext_features/__init__.py randomization/getperturb_li_attack32_tfmultigpu.py lid/evaluation.py inputtransformations/helpers.py pixel-deflection/methods.py randomization/re_li_attack32_tfmultigpu_notanh.py wideresnet28/discretization_utils.py randomization/re_li_attack.py inputtransformations/re_l2_attack.py randomization/re_li_attack32_tf.py randomization/robustml_model.py randomization/utils.py randomization/re_li_attack32.py cascade_adv_training/lenet.py wideresnet28/helpers.py pixel-deflection/helpers.py guided-denoiser/v3.py guided-denoiser/inres.py cascade_adv_training/re_li_attack.py robustnet/utils.py inputtransformations/defense.py lid/util.py guided-denoiser/inceptionresnetv2.py sap/cifar_model.py lid/attack_lid.py therm-adv/cifar_model.py guided-denoiser/resnext.py guided-denoiser/inceptionresnet.py robustnet/models/lenet.py sap/robustml_attack.py lid/cifar10_input.py cascade_adv_training/convert_pickle.py cascade_adv_training/resnet_layers.py sap/re_li_attack.py therm-adv/robustml_attack.py inputtransformations/robustml_attack.py randomization/re_li_attack32_multigpu.py wideresnet28/train.py cascade_adv_training/layers.py randomization/re_li_attack32_tfmultigpu.py confusion.py randomization/defense.py cascade_adv_training/helpers.py robustnet/models/dpn.py robustnet/attack.py wideresnet28/cifar10_input.py guided-denoiser/res152_wide.py lid/helpers.py robustnet/models/layer.py wideresnet28/cifar_model.py therm-adv/cifar10_input.py wideresnet28/discretization_attacks.py lid/extract_artifacts.py inputtransformations/imagenet_labels.py inputtransformations/re_l2_attack_clipimage_notanh.py therm-adv/evaluation_bpda.py guided-denoiser/dataset.py inputtransformations/re_li_attack.py robustnet/utils2.py randomization/helpers.py one_hot cifar10 _grayscale cifar100 _load_batch_cifar100 _load_batch_cifar10 maybe_download_and_extract DataSet main reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale fc gated_conv conv variable_on_cpu activation_summary variable_with_weight_decay lenet main train optimistic_restore save loss step_forward _add_loss_summaries _get_similarity_loss _get_pivot_loss inference train _per_image_standardization _sparse_cross_entropy_loss step_backward MNIST Provider ImageNet CIFAR10 noresnet resnet resnet_e10x2 resnet_e2 fc_layer residual_block residual_bottleneck gated_conv_layer gated_residual_block base_conv_layer noresidual_block conv_layer softmax_layer main main main rename get_step_adver_image get_batch plot_images plot_save_corr_grads copy_crop_images get_adver_image visualize_embeddings get_data plot_save_embeddings print_test_accuracy get_intensity plot_confusion_matrix data_mnist get_adver_grads trace_embeddings optimistic_restore plot_example_errors analyze_corr_grads Dataset find_images_and_targets natural_key main LeNormalize reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale Block17 Block8 Mixed_6a Null InceptionResnetV2 Denoise Bottleneck Mixed_5b Loss BasicConv2d Conv Net Block35 DenoiseLoss Mixed_7a get_net Block17 Block8 Mixed_6a InceptionResnetV2 Mixed_5b BasicConv2d Block35 Mixed_7a get_model MNIST Provider ImageNet CIFAR10 get_model Null ResNet Bottleneck Loss Conv Net DenoiseLoss Denoise get_net Null Bottleneck Loss Conv Net DenoiseLoss Denoise ResNeXt101_64x4d get_net get_model main softmax main softmax get_model Lambda LambdaBase LambdaReduce LambdaMap Lambda LambdaBase LambdaReduce LambdaMap defend_jpeg defend_tv bregman defend_crop make_defend_quilt defend_reduce reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale label_to_name _preprocess model _get_model main load_image main main main main main BPDA InputTransformations InputTransformations load_image optimistic_restore one_hot make_classify Attack DataSubset AugmentedCIFAR10Data CIFAR10Data AugmentedDataSubset main softmax main softmax reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale main softmax reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale process_image process_image_parallel get_arguments classify_images pixel_deflection denoiser main defend get_img rgb2ycc batches ycc2rgb get_imagenet_labels get_map defend defend_batch main reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale label_to_name _preprocess model _get_model MNIST Provider ImageNet CIFAR10 main main main main main main main main EOT Randomization load_image optimistic_restore one_hot make_classify attack weights_init main softmax weights_init reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale train test train_other accuracy weights_init main loss data read_f parse_line main softmax weights_init train_other accuracy weights_init main loss format_time init_params get_mean_and_std format_time init_params progress_bar get_mean_and_std DenseNet201 DenseNet161 DenseNet121 Transition DenseNet Bottleneck densenet_cifar DenseNet169 test_densenet DPN Bottleneck test DPN92 DPN26 GoogLeNet Inception Noise LeNet Block MobileNet test PreActBlock PreActResNet50 PreActResNet PreActResNet18 test PreActResNet152 PreActBottleneck PreActResNet101 PreActResNet34 ResNet ResNet18 ResNet34 Bottleneck ResNet101 test ResNet50 BasicBlock ResNet152 Block ResNeXt29_4x64d ResNeXt ResNeXt29_2x64d test_resnext ResNeXt29_32x4d ResNeXt29_8x64d PreActBlock SENet18 SENet test BasicBlock ShuffleNetG2 ShuffleNet Bottleneck test ShuffleBlock ShuffleNetG3 VGG DataSubset AugmentedCIFAR10Data CIFAR10Data AugmentedDataSubset Model main main reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale main softmax Attack main SAP SAPModel DataSubset AugmentedCIFAR10Data CIFAR10Data AugmentedDataSubset Model adv_dga adv_lspga discretize_range one_hot_to_thermometer unflatten_last random_convolution discretize_uniform get_centroids_by_percentile undiscretize_centroids undiscretize_uniform flatten_last discretize_centroids thermometer_to_one_hot main main reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale main main Attack main Thermometer DataSubset AugmentedCIFAR10Data CIFAR10Data AugmentedDataSubset Model adv_dga adv_lspga discretize_range one_hot_to_thermometer unflatten_last random_convolution discretize_uniform get_centroids_by_percentile undiscretize_centroids undiscretize_uniform flatten_last discretize_centroids thermometer_to_one_hot main reduce_max torch_arctanh l2_dist reduce_sum l2r_dist l1_dist l2_norm reduce_mean rescale l1_norm reduce_min tanh_rescale join urlretrieve print extractall stat makedirs array load join one_hot reshape transpose concatenate _grayscale append _load_batch_cifar10 range load join one_hot reshape transpose _grayscale _load_batch_cifar100 checkpoint_dir ArgumentParser resize Session clip run open str data_dir transpose copy_crop_images placeholder append inference parse_args sum optimistic_restore convert_to_tensor format get_checkpoint_state CIFAR10 ConfigProto listdir float load int time tanh print reshape add_argument float32 model_checkpoint_path list reversed sum dim range mul reduce reduce_sum list dim range reversed list dim range reversed reduce_sum abs reduce_sum get_variable multiply add_to_collection variable_on_cpu xavier_initializer l2_loss name zero_fraction sub histogram scalar matmul add constant_initializer variable_on_cpu variable_with_weight_decay conv2d constant_initializer variable_on_cpu bias_add variable_with_weight_decay add conv2d constant_initializer variable_on_cpu bias_add variable_with_weight_decay reshape lrn max_pool join makedirs sorted NewCheckpointReader list restore global_variables dict Saver zip get_variable_to_shape_map Exists DeleteRecursively pprint is_train MakeDirs __flags train restore_inplace train_dir reshape inference _sparse_cross_entropy_loss gradients embedding_at per_image_standard model_name resnet_n add_noise_inference _per_image_standardization add_to_collection add_to_collection add_to_collection batch_size pivot_loss_factor adver_batch_size min samebatch_loss_factor _get_similarity_loss similarity_loss_factor _get_pivot_loss adver_loss_factor _sparse_cross_entropy_loss name get_collection apply average ExponentialMovingAverage scalar int trainable_variables list batch_size name to_int32 print tuple get_collection _add_loss_summaries apply apply_gradients UPDATE_OPS histogram ExponentialMovingAverage zip piecewise_constant scalar reshape range reshape range reshape range reshape range constant_initializer variable_with_weight_decay matmul variable_on_cpu fc_layer softmax conv2d variable_with_weight_decay conv2d variable_with_weight_decay batch_norm conv2d variable_with_weight_decay batch_norm relu relu relu relu randn max log range mean softmax flush repeat zeros std get_checkpoint_state rename getopt exit print reshape astype float32 load_data int copy randint empty range normal ones randint abs clip percentile normal ones_like zeros_like where sign get_intensity float abs max_e clip len seed update normal int get_step_adver_image zeros_like concatenate ones min copy floor run randint float range max_e clip len batch_size per_pixel_rand floor max clip copy_crop_images get_adver_image get_intensity append max_e adver_batch_size empty_like cascade empty enumerate rand_crop sparsity adversarial zeros next_batch method adver_option copy_crop_images get_adver_image copy images len show format subplots reshape set_xlabel set_yticks subplots_adjust imshow set_xticks clip flat enumerate plot_images print sum confusion_matrix batch_size where get_data save argmax run str saved_data_dir sum train_dir format size plot_confusion_matrix float load print zeros plot_example_errors len copy_crop_images empty_like copy images run len get_major_ticks subplots grid show set_fontsize squeeze axvline ylabel set_linewidth ylim savefig legend train_dir semilogx get_lines annotate float xlabel get_texts fill_between items list str plot_save_corr_grads zip batch_size print reshape copy_crop_images map get_data mean get_adver_grads expand_dims argmax std enumerate len show get_major_ticks subplots set_fontsize plot xlabel grid ylabel set_linewidth ylim get_lines get_texts legend savefig fill_between train_dir str batch_size print reshape copy_crop_images get_data mean plot_save_embeddings zip run expand_dims argmax std enumerate len get_major_ticks add_subplot get_data argmax run show list set_fontsize copy_crop_images set_linewidth scatter savefig quiver legend append expand_dims next range train_dir get_position set_position get_lines enumerate print images cycle figure get_texts zeros array len join sorted list dict splitext zip append keys walk enumerate DataLoader input_dir cuda load_state_dict concatenate Compose get_model1 eval net get_model2 enumerate get_model3 get_model4 Variable numpy Dataset Net get_net argmax squeeze imagenet_path asarray ImageNet array append uint8 astype float32 fromarray uint8 BytesIO astype float32 save append inf shape sqrt zeros range uniform load reshape float32 placeholder matmul shape top_k to_float hasattr inception_v3_arg_scope inception_v3 default_image_size network_fn _preprocess default_image_size _get_model argmax optimistic_restore join load_image extend randint shape save sample astype float32 repeat resize open InputTransformations outlogits epsilon evaluate BPDA targeted defense zeros model perturb cifar_path set_session load_weights get_model savez add_argument ArgumentParser preprocess_input join format decode_predictions print stack zip predict get_img window deflections get_map pixel_deflection zeros sigma format batches glob directory batch_size cpu_count print classify_images append shape range zeros pixel_deflection window denoiser sigma ResNet50 deflections defend preprocess_input get_session dot T array float array astype load_img Normalize load open pad set_shape random_uniform crop_and_resize to_float pad set_shape tile random_uniform crop_and_resize range batchlogits_modify batchlogits dump multigpu_npoplogits_modify Randomization npoplogits multigpu_npoplogits npoplogits_modify multigpu_npoplogits_modify_notanh batchlogits_modify_notanh EOT data view FloatTensor Variable backward step zero_grad Adam range scatter_ zero_ sum cuda net len normal_ __name__ fill_ modelIn DataParallel VGG apply CrossEntropyLoss noise data criterion backward progress_bar zero_grad step max net enumerate len data criterion print progress_bar eval mkdir save max net enumerate len data sum cuda iter cuda iter loss time format backward print exit Adam SGD RMSprop zero_grad parameters loss_f accuracy range iter step net train_other modelOut state_dict epoch lr method float strip append parse_line open show plot read_f ylim legend zip array enumerate print DataLoader div_ zeros range len normal constant isinstance kaiming_normal Conv2d bias modules BatchNorm2d weight Linear int int time join format_time write append range flush len randn Variable print densenet_cifar net randn Variable DPN92 MobileNet size PreActResNet18 ResNet18 randn Variable print size ResNeXt29_2x64d net SENet18 ShuffleNetG2 SAP Attack discretize_fn clip_by_value linspace thermometer_to_one_hot discretize_fn softmax_cross_entropy_with_logits discretize_range one_hot gradients model cumsum projection_fn one_hot_to_thermometer unflatten_last clip_by_value random_uniform stop_gradient flatten_last argmax range thermometer_to_one_hot softmax_cross_entropy_with_logits discretize_range one_hot zeros_like model cumsum projection_fn gradients one_hot_to_thermometer unflatten_last range sign shape softmax stop_gradient flatten_last argmax random_normal as_list reshape as_list int reshape append one_hot to_int32 one_hot_to_thermometer clip_by_value flatten_last lin_space map_fn to_float one_hot argmin one_hot_to_thermometer stack flatten_last squared_difference to_float argmax unflatten_last reduce_sum multiply unflatten_last reduce_sum thermometer_to_one_hot cumsum flatten_last unflatten_last one_hot to_int32 unflatten_last reduce_sum flatten_last conv2d flatten_last unflatten_last Thermometer discretize_uniform

# NATTACK: A STRONG AND UNIVERSAL GAUSSIAN BLACK-BOX ADVERSARIAL ATTACK Data and model can be found here: [data and model](https://knightsucfedu39751-my.sharepoint.com/:f:/g/personal/liyandong_knights_ucf_edu/EnmkFaQkvwdDq0xcqIbbEfYBAhSkQK16ONPgjMJncbCwmg). Please download the data&&model and unzip them to './cifar-data' and './all_models' Below is Table 1 from our paper, where we show the robustness of each accepted defense to the adversarial examples we can construct: | Defense | Dataset | Distance | Success rate | |---|---|---|---| | ADV-TRAIN [Madry et al. (2018)](https://arxiv.org/abs/1706.06083) | CIFAR | 0.031 (linf) | 47.9% |

ColumbiaDVMM/Transform_Covariant_Detector

['image retrieval']

['Learning discriminative and transformation covariant local feature detectors.', 'Learning Discriminative and Transformation Covariant Local Feature Detectors']

tensorflow/patch_network_point_test.py tensorflow/patch_cnn.py eval/external/vlfeat-0.9.18/docsrc/wikidoc.py eval/external/vlfeat-0.9.18/docsrc/webdoc.py tensorflow/patch_network_train_point.py eval/external/vlfeat-0.9.18/docsrc/doxytag.py tensorflow/patch_reader.py eval/external/vlfeat-0.9.18/docsrc/mdoc.py eval/external/vlfeat-0.9.18/docsrc/formatter.py Doxytag Terminal Lexer B PL L lex Formatter DL BL E extract towiki depth_first breadCrumb MFile Node runcmd xscan wikidoc usage PatchCNN read_image_from_name SiameseDataSet Doxytag Terminal Lexer B PL L lex Formatter DL BL E extract towiki depth_first breadCrumb MFile Node runcmd xscan wikidoc usage PatchCNN read_image_from_name SiameseDataSet bullet indent inner_content PL group match DL BL len pid Popen waitpid children group lstrip match startswith append open join addMFile addChildNode print sort MFile Node match listdir __next__ prev runcmd join wikidoc print print insert print readlines close len writelines append range exists open float transpose repeat process_file resize imread flip open

## Learning Discriminative and Transformation Covariant Local Feature Detectors This code is the training and evaluation code for our CVPR 2017 paper. It includes the implement of a translation covariant local feature detector. The affine covariant model will be added in the future. @inproceedings{zhang2017learning, title={Learning Discriminative and Transformation Covariant Local Feature Detectors}, author={Zhang, Xu and Yu, Felix X. and Karaman, Svebor and Chang, Shih-Fu}, booktitle={CVPR}, year={2017} } The code is tested on Ubuntu 14.04 ### Requirement

ConstantinSeibold/SGL

['multiple instance learning']

['Self-Guided Multiple Instance Learning for Weakly Supervised Disease Classification and Localization in Chest Radiographs']

MNIST-Bags_Experiments/mnist_exp.py MNIST-Bags_Experiments/bags_nets.py MNIST-Bags_Experiments/functions_.py MNIST-Bags_Experiments/mnist_bags.py F5 C1 C3 F4 LeNet5 C2 sgl cust_bce mean weight max MnistBags init_func get_loaders run_max run_mmm run_mean run_bil run_sgl run view mean cuda abs max view cust_bce sigmoid pooling weight cuda detach DataLoader MnistBags time get_loaders clear_output format print roc_curve Adam apply parameters eval BCELoss train BCEWithLogitsLoss cuda range cat auc time get_loaders clear_output format print roc_curve Adam apply parameters eval BCELoss train BCEWithLogitsLoss cuda range cat auc time get_loaders clear_output format print roc_curve Adam apply parameters eval BCELoss train BCEWithLogitsLoss cuda range cat auc time get_loaders clear_output format print roc_curve Adam apply parameters eval BCELoss train BCEWithLogitsLoss cuda range cat auc time get_loaders clear_output format print roc_curve Adam apply parameters eval BCELoss train BCEWithLogitsLoss cuda range cat auc run_iteration

# Self-guiding Loss for Multiple Instance Learning  The Self-Guiding Loss is a novel multiple-instance learning loss which integrates artificial supervision based on the networks predictions into its formulation in an online step. The SGL can be seen as an extension to the standard MIL-setting. This repository contains the loss comparison studies on the MNIST-Bags dataset of the ACCV 2020 paper [**Self-Guided Multiple Instance Learning for Weakly Supervised Thoracic Disease Classification and Localization in Chest Radiographs**](https://arxiv.org/pdf/2010.00127.pdf). > [**Self-Guided Multiple Instance Learning for Weakly Supervised Thoracic Disease Classification and Localization in Chest Radiographs**](https://arxiv.org/pdf/2010.00127.pdf)<br> > Constantin Seibold, Jens Kleesiek, Heinz-Peter Schlemmer, Rainer Stiefelhagen<br> > > > **Abstract:** *Due to the high complexity of medical images and the scarcity of trained personnel, most large-scale radiological datasets are lacking fine-grained annotations and are often only described on image-level. These shortcomings hinder the deployment of automated diagnosis systems, which require human-interpretable justification for their decision process. In this paper, we address the problem of weakly supervised identification and localization of abnormalities in chest radiographs in a multiple-instance learning setting. To that end, we introduce a novel loss function for training convolutional neural networks increasing the localization confidence and assisting the overall disease identification. The loss leverages both image- and patch-level predictions to generate auxiliary supervision and enables specific training at patch-level. Rather than forming strictly binary from the predictions as done in previous loss formulations, we create targets in a more customized manner. This way, the loss accounts for possible misclassification of less certain instances. We show that the supervision provided within the proposed learning scheme leads to better performance and more precise predictions on prevalent datasets for multiple-instance learning as well as on the NIH ChestX-Ray14 benchmark for disease recognition than previously used losses.* ## Contents Available material to our paper can be found here:

ContextScout/ned-graphs

['entity disambiguation']

['Named Entity Disambiguation using Deep Learning on Graphs']

wikidata_entity_linking_with_attentive_gcn/wikidata_query/sentence_processor.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/utils.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/reformat_wikidata.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/graphs.py wikidata_entity_linking_with_gcn_only/wikidata_query/train.py wikidata_entity_linking_with_gcn_only/wikidata_query/read_data.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/train.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/utils.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/graphs.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/read_data.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/gcn_qa_model.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/wikidata_items.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/wikidata_items.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/wikidata_items.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/wikidata_items.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/read_data.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/test.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/sentence_processor.py wikidata_entity_linking_with_gcn_only/wikidata_query/test.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/gcn_qa_model.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/read_data.py wikidata_entity_linking_with_gcn_only/wikidata_query/graphs.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/test.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/test.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/gcn_qa_model.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/read_data.py wikidata_entity_linking_with_gcn_only/wikidata_query/wikidata_items.py wikidata_entity_linking_with_gcn_only/wikidata_query/gcn_qa_model.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/train.py wikidata_entity_linking_with_attentive_rnn_triplets/wikidata_query/utils.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/graphs.py wikidata_entity_linking_with_gcn_only/wikidata_query/sentence_processor.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/train.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/gcn_qa_model.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/graphs.py wikidata_entity_linking_with_attentive_gcn/wikidata_query/utils.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/sentence_processor.py wikidata_entity_linking_with_text_and_centroid/wikidata_query/test.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/sentence_processor.py wikidata_entity_linking_with_gcn_only/wikidata_query/utils.py wikidata_entity_linking_with_rnn_triplets/wikidata_query/train.py compute_new_adjacency_matrix GCN_layer_fw GCN_QA get_triplets_for_word_1_hop get_triplets_for_word_2_hops get_graph_from_wikidata_id get_wikidata_id_of_item_different_from_given_one_with_boundaries get_json_data infer_vector_from_vector_nodes get_json_data_many_wrong_ids get_wikidata_id_from_wikipedia_id get_data convert_text_into_vector_sequence create_text_item_graph_dict get_item_mask_for_words get_data_and_write_json get_wikidata_id_of_item_different_from_given_one get_bw_graph get_adjacency_matrices_and_vectors_given_triplets get_prediction_from_models find_position_of_best_match test erase_edges_with_mask get_vector_list_from_sentence get_answers_and_questions_from_json train find_position_of_best_match get_vector_list_from_sentence get_answers_and_questions_from_json get_edge_name_with_signature infer_vector_from_word low_case get_vectors_from_nodes_in_graph add_triplets_to_graph_bw get_chunks capitalize infer_vector_from_doc get_node_name_with_signature plot_graph get_types_from_nodes_in_graph get_words bin_data_into_buckets WikidataItems GCN_QA get_triplets_for_word_1_hop get_triplets_for_word_2_hops get_graph_from_wikidata_id get_wikidata_id_of_item_different_from_given_one_with_boundaries get_json_data infer_vector_from_vector_nodes get_json_data_many_wrong_ids get_wikidata_id_from_wikipedia_id get_data convert_text_into_vector_sequence create_text_item_graph_dict get_item_mask_for_words get_data_and_write_json get_wikidata_id_of_item_different_from_given_one get_adjacency_matrices_and_vectors_given_triplets create_vectors_from_triplets get_prediction_from_models find_position_of_best_match test erase_edges_with_mask get_vector_list_from_sentence get_answers_and_questions_from_json train find_position_of_best_match get_vector_list_from_sentence get_answers_and_questions_from_json get_edge_name_with_signature infer_vector_from_word low_case get_vectors_from_nodes_in_graph add_triplets_to_graph_bw get_chunks capitalize infer_vector_from_doc get_node_name_with_signature plot_graph get_types_from_nodes_in_graph get_words bin_data_into_buckets WikidataItems GCN_layer_bw GCN_layer_fw GCN_QA get_triplets_for_word_1_hop get_triplets_for_word_2_hops get_graph_from_wikidata_id get_wikidata_id_of_item_different_from_given_one_with_boundaries get_json_data infer_vector_from_vector_nodes get_json_data_many_wrong_ids get_wikidata_id_from_wikipedia_id get_data convert_text_into_vector_sequence create_text_item_graph_dict get_item_mask_for_words get_data_and_write_json get_wikidata_id_of_item_different_from_given_one get_bw_graph get_adjacency_matrices_and_vectors_given_triplets get_prediction_from_models find_position_of_best_match test erase_edges_with_mask get_vector_list_from_sentence get_answers_and_questions_from_json train find_position_of_best_match get_vector_list_from_sentence get_answers_and_questions_from_json get_edge_name_with_signature infer_vector_from_word low_case get_vectors_from_nodes_in_graph add_triplets_to_graph_bw get_chunks capitalize infer_vector_from_doc get_node_name_with_signature plot_graph get_types_from_nodes_in_graph get_words bin_data_into_buckets WikidataItems GCN_QA get_triplets_for_word_1_hop get_triplets_for_word_2_hops get_graph_from_wikidata_id get_wikidata_id_of_item_different_from_given_one_with_boundaries get_json_data infer_vector_from_vector_nodes get_json_data_many_wrong_ids get_wikidata_id_from_wikipedia_id get_data convert_text_into_vector_sequence create_text_item_graph_dict get_item_mask_for_words get_data_and_write_json get_wikidata_id_of_item_different_from_given_one get_adjacency_matrices_and_vectors_given_triplets create_vectors_from_triplets get_prediction_from_models find_position_of_best_match test erase_edges_with_mask get_vector_list_from_sentence get_answers_and_questions_from_json train find_position_of_best_match get_vector_list_from_sentence get_answers_and_questions_from_json get_edge_name_with_signature infer_vector_from_word low_case get_vectors_from_nodes_in_graph add_triplets_to_graph_bw get_chunks capitalize infer_vector_from_doc get_node_name_with_signature plot_graph get_types_from_nodes_in_graph get_words bin_data_into_buckets WikidataItems GCN_layer_bw GCN_layer_fw GCN_QA get_triplets_for_word_1_hop get_triplets_for_word_2_hops get_graph_from_wikidata_id get_wikidata_id_of_item_different_from_given_one_with_boundaries get_json_data infer_vector_from_vector_nodes get_json_data_many_wrong_ids get_wikidata_id_from_wikipedia_id get_data convert_text_into_vector_sequence create_text_item_graph_dict get_item_mask_for_words get_data_and_write_json get_wikidata_id_of_item_different_from_given_one get_bw_graph get_adjacency_matrices_and_vectors_given_triplets get_prediction_from_models find_position_of_best_match test erase_edges_with_mask get_vector_list_from_sentence get_answers_and_questions_from_json train find_position_of_best_match get_vector_list_from_sentence get_answers_and_questions_from_json get_edge_name_with_signature infer_vector_from_word low_case get_vectors_from_nodes_in_graph add_triplets_to_graph_bw get_chunks capitalize infer_vector_from_doc get_node_name_with_signature plot_graph get_types_from_nodes_in_graph get_words bin_data_into_buckets WikidataItems Variable multiply transpose map_fn random_uniform compute_new_adjacency_matrix relu Variable transpose matmul map_fn random_uniform translate_from_url append json translate_from_url append json str sorted list json set translate_from_url append append infer_vector_from_word get_words append lower get_words zeros norm convert_text_into_vector_sequence infer_vector_from_vector_nodes get_item_mask_for_words get_graph_from_wikidata_id append create_text_item_graph_dict append create_text_item_graph_dict list reverse_lookup set list reverse_lookup set add_triplets_to_graph_bw DiGraph list get_vectors_from_nodes_in_graph nodes index get_bw_graph get_types_from_nodes_in_graph array to_numpy_matrix append read loads norm enumerate append infer_vector_from_word get_words enumerate list len append range predict enumerate print predict str sorted write save append range bin_data_into_buckets to_unicode tokenize TweetTokenizer zeros zeros norm get_words replace infer_vector_from_doc nodes append append zeros nodes split lower lower add_edge get_node_name_with_signature get_edge_name_with_signature add_node draw_networkx_edge_labels show shell_layout draw_networkx list get_chunks append keys len join getLogger create_vectors_from_triplets relu Variable transpose matmul map_fn random_uniform sorted

Code and Dataset for Named Entity Disambiguation using Deep Learning on Graphs ============================================================================== This repository contains the code and dataset for the paper "Named Entity Disambiguation using Deep Learning on Graphs". The full paper can be found [here](https://arxiv.org/pdf/1810.09164.pdf). Installation ------------ The main requirements are installed with: ```bash virtualenv --python=/usr/bin/python3 .env source .env/bin/activate pip install -r requirements.txt

Coolgiserz/NLP_starter

['text classification', 'stochastic optimization']

['Optimization Methods for Large-Scale Machine Learning', 'Convex Optimization: Algorithms and Complexity']

CodePratices/MachineLearning/pos_tagging_hmm_project/main.py CodePratices/MachineLearning/pos_tagging_hmm_project/models/hmm.py CodePratices/MachineLearning/pos_tagging_hmm_project/utils/corpus.py test HMMPOSTagger CorpusHelper print decode

# README ## 学习资料 关于人工智能、机器学习、自然语言处理等领域的学习资料。 ### 课程 #### 人工智能 - [Artificial Intelligence: A Modern Approach](http://aima.cs.berkeley.edu) #### 机器学习 - 台大李宏毅机器学习课程： [中文课程！台大李宏毅机器学习公开课2019版上线](https://zhuanlan.zhihu.com/p/59655414) [[课程地址]](http://speech.ee.ntu.edu.tw/~tlkagk/courses_ML19.html) - [CS224W: Machine Learning with Graphs](http://web.stanford.edu/class/cs224w/)

CrazySummerday/ctpn.pytorch

['scene text detection']

['Detecting Text in Natural Image with Connectionist Text Proposal Network']

train.py ctpn/dataset.py ctpn/config.py ctpn/utils.py ctpn/ctpn.py predict.py get_text_boxes save_checkpoint weights_init RPN_REGR_Loss CTPN_Model RPN_CLS_Loss basic_conv VOCDataset readxml ICDARDataset nms clip_bbox TextProposalConnectorOriented Graph bbox_transfrom filter_bbox compute_iou TextLineCfg cal_rpn TextProposalGraphBuilder transform_bbox gen_anchor int astype float32 copy IMAGE_MEAN resize to max join checkpoints_dir format print save normal_ __name__ fill_ int list parse text iter append float round list arange reshape hstack dstack stack meshgrid array range append len minimum maximum tile zeros enumerate log transpose exp minimum maximum argmax sum bbox_transfrom compute_iou fill empty gen_anchor append maximum minimum

# ctpn.pytorch Pytorch implementation of CTPN (Detecting Text in Natural Image with Connectionist Text Proposal Network) # Paper https://arxiv.org/pdf/1609.03605.pdf # train training dataset: ICDAR2013 and ICDAR2017. If you want to train your own dataset, you need to change the 'img_dir' and 'label_dir' in file *ctpn/config.py*, then run ``` python train.py ```

Crespo-dong/caffe_ocr

['optical character recognition', 'scene text recognition']

['An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition']

examples/pycaffe/layers/pascal_multilabel_datalayers.py examples/finetune_flickr_style/assemble_data.py tools/extra/resize_and_crop_images.py tools/extra/summarize.py src/caffe/test/test_data/generate_sample_data.py examples/pycaffe/tools.py tools/extra/parse_log.py 3rdparty/include/google/protobuf/arena_nc_test.py examples/pycaffe/caffenet.py tools/extra/extract_seconds.py examples/web_demo/exifutil.py examples/pycaffe/layers/pyloss.py examples/web_demo/app.py ArenaNcTest download_image make_net max_pool caffenet conv_relu fc_relu CaffeSolver SimpleTransformer print_info check_params PascalMultilabelDataLayerSync load_pascal_annotation BatchLoader EuclideanLossLayer start_tornado start_from_terminal embed_image_html classify_upload index allowed_file ImagenetClassifier classify_url open_oriented_im apply_orientation get_start_time extract_seconds extract_datetime_from_line get_log_created_year write_csv parse_log fix_initial_nan_learning_rate save_csv_files main parse_args parse_line_for_net_output ResizeCropImagesMapper PILResizeCrop OpenCVResizeCrop print_table printed_len summarize_net main read_net format_param imread urlretrieve Convolution InnerProduct Data SoftmaxWithLoss LRN Accuracy max_pool InnerProduct conv_relu fc_relu Dropout join list getElementsByTagName get_data_from_tag csr_matrix dict zip zeros float range enumerate len print format get read info load_image classify_image StringIO join replace info secure_filename save filename open_oriented_im classify_image fromarray replace astype save resize StringIO items list listen HTTPServer format print start WSGIContainer update start_tornado add_option OptionParser debug port parse_args ImagenetClassifier forward run hasattr _getexif astype float32 tile apply_orientation open transpose int rfind datetime split getctime year strip extract_datetime_from_line get_start_time total_seconds strip write get_log_created_year close extract_datetime_from_line open float get_log_created_year compile fix_initial_nan_learning_rate search group OrderedDict append float join basename write_csv print excel add_argument ArgumentParser parse_log save_csv_files output_dir logfile_path parse_args NetParameter decay_mult format name lr_mult append print zip len get join str format convolution_param list setdefault param kernel_size map set top bottom append type module layer enumerate print_table add_argument ArgumentParser filename summarize_net read_net

# 简介 caffe_ocr是一个对现有主流ocr算法研究实验性的项目，目前实现了CNN+BLSTM+CTC的识别架构，并在数据准备、网络设计、调参等方面进行了诸多的实验。代码包含了对lstm、warp-ctc、multi-label等的适配和修改，还有基于inception、restnet、densenet的网络结构。代码是针对windows平台的，linux平台下只需要合并相关的修改到caffe代码中即可。 ## caffe代码修改  1. data layer增加了对multi-label的支持<br>  2. lstm使用的是junhyukoh实现的lstm版本（lstm_layer_Junhyuk.cpp/cu），原版不支持变长输入的识别。输入的shape由(TxN)xH改为TxNxH以适应ctc的输入结构。<br>  3. WarpCTCLossLayer去掉了对sequence indicators依赖（训练时CNN输出的结构是固定的），简化了网络结构（不需要sequence indicator layer）。<br>  4. densenet修改了对Reshape没有正确响应的bug，实现了对变长输入预测的支持。<br>  5. 增加transpose_layer、reverse_layer，实现对CNN feature map与lstm输入shape的适配<br> ## 编译   1. 安装opencv,boost,cuda,其它依赖库在3rdparty下（包含debug版的lib:http://pan.baidu.com/s/1nvIFojJ）<br>

Cysu/person_reid

['person re identification']

['Learning Deep Feature Representations with Domain Guided Dropout for Person Re-identification']

data/format_3dpes.py tools/make_lists_id_training.py data/format_prid.py utils/__init__.py data/format_shinpuhkan.py data/format_cuhk03.py utils/core.py data/format_ilids.py tools/save_joint_impact_score.py tools/convert_lmdb_to_numpy.py tools/save_individual_impact_score.py utils/cmc.py eval/metric_learning.py data/format_viper.py data/format_cuhk01.py tools/merge_lists_single_task.py tools/compute_impact_score.py main main main _load main main main main _learn_pca _eval_cmc _get_train_data _get_test_data _learn_metric main main parse_prototxt main main _get_list _save main main load_domain_impact main _cmc_core cmc read_kv write_list read_json write_kv read_list write_json mkdir_if_missing pickle unpickle join list defaultdict format permutation sorted glob len shuffle copy input_dir output_dir append write_json keys enumerate mkdir_if_missing cuhk01_dir range join loadmat _load squeeze sort choice set cuhk03_dir zip imsave int items ilids_dir prid_dir str basename shinpuhkan_dir viper_dir asarray unique load join asarray unique PCA fit arange inv choice dot eye meshgrid sum len pairwise_distances cmc _learn_pca _get_train_data _eval_cmc result_dir print _get_test_data save _learn_metric transform pca method name len range layer enumerate data model forward max log parse_prototxt set_device num_iters normalize TEST Net weights output set_mode_gpu zeros layer asarray input_lmdb output_npy Datum open append int write_list read_json val_ratio extend dataset_dir _get_list _save read_kv write_list map dataset_dirs write_kv db_dirs values load isdir output_lmdb rmtree input_npy load join list glob SerializeToString extend Datum len load_domain_impact image_list_file read_list impact_dir argsort shape asarray arange isinstance size choice shape unique zeros range enumerate makedirs read_list list write_list zip

# Domain Guided Dropout for Person Re-id This project aims at learning generic person re-identification (re-id) deep features from multiple datasets with domain guided dropout. Mainly based on our CVPR 2016 paper [Learning Deep Feature Representations with Domain Guided Dropout for Person Re-identification](http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Xiao_Learning_Deep_Feature_CVPR_2016_paper.pdf). ## Installation We have integrated our self-brewed caffe into `external/caffe`, which provides batch-normalization and multi-gpu parallel training. Please clone this project with the command: git clone --recursive https://github.com/Cysu/dgd_person_reid.git Apart from the official installation [prerequisites](http://caffe.berkeleyvision.org/installation.html), we have several other dependencies: cudnn-v4, openmpi, and 1.55 <= boost < 1.60. You may install them manually or by a package manager (a tip for installing boost 1.55 on Ubuntu 14.04: `sudo apt-get autoremove libboost1.54*` then `sudo apt-get install libboost1.55-all-dev`). Then configure the `Makefile.config` and compile the caffe. To use multi-GPU for training, please uncomment the MPI parallel block in the `Makefile.config` and set the `MPI_INCLUDE` and `MPI_LIB` properly. Please find more details of using the caffe [here](https://github.com/Cysu/caffe/tree/domain-guided-dropout). cd external/caffe cp Makefile.config.example Makefile.config # Configure the libraries properly

D-Roberts/lq_backprop

['graph learning']

['QR and LQ Decomposition Matrix Backpropagation Algorithms for Square, Wide, and Deep -- Real or Complex -- Matrices and Their Software Implementation']

test_lq_op_grad.py lq_op_grad.py lq LqGrad numeric_q test theoretic_q _extra_feeds _test_LQ_op numeric_l theoretic_l qr adjoint _LqGradSquareAndWideMatrices matmul update seed eps astype Session seed convert_to_tensor reshape astype range shape lq ravel prod Session seed convert_to_tensor reshape astype range shape lq ravel prod Session seed eps astype shape Session convert_to_tensor lq astype Session _test_LQ_op

# LQ Matrix Backpropagation Algorithm Implementation TensorFlow implementation of differentiable LQ matrix decomposition for square, wide and deep tensors. # To Use Requirements: tf >v1; Python > 3.6. Recommended: install Anaconda. Create a tensorflow environment. ``` # tf cpu only; v2 by default at this time. conda create -n tf tensorflow conda activate tf

D3-AI/Orion

['anomaly detection']

['Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding']

orion/benchmark.py orion/primitives/intervals.py orion/data.py tests/test_analysis.py orion/primitives/estimators.py orion/evaluation/point.py tests/test_benchmark.py orion/db/schema.py tutorials/tulog/utils.py orion/__main__.py orion/primitives/azure_anomaly_detector.py orion/core.py tests/test_functional.py tests/evaluation/test_common.py orion/functional.py tests/primitives/test_timeseries_anomalies.py docs/conf.py tests/evaluation/test_point.py tests/evaluation/test_contextual.py orion/evaluation/__init__.py orion/db/base.py orion/progress.py orion/utils.py orion/evaluation/contextual.py tests/primitives/test_timeseries_errors.py tests/primitives/test_timeseries_preprocessing.py orion/results.py orion/primitives/timeseries_errors.py orion/runner.py orion/db/__init__.py tasks.py orion/primitives/timeseries_anomalies.py tutorials/tulog/model.py orion/__init__.py orion/db/explorer.py tests/test_core.py orion/evaluation/utils.py orion/primitives/detectors.py orion/evaluation/common.py tests/test_data.py setup.py orion/analysis.py tests/evaluation/test_utils.py orion/primitives/timeseries_preprocessing.py orion/primitives/tadgan.py minimum install_minimum readme remove_readonly tutorials lint rmdir pytest analyze _load_pipeline _run_pipeline get_available_templates _build_events_df _detrend_signal _get_pipeline_hyperparameter _load_signal _sort_leaderboard _run_on_dask _parse_confusion_matrix benchmark _evaluate_signal main _run_job Orion load_csv load_signal format_csv load_anomalies download_demo download _load_data evaluate_pipeline _load_orion fit_pipeline _load_dict detect_anomalies TqdmLogger progress get_summary_page write_results add_sheet get_f1_scores start_signalrun get_outputs_spec start_datarun process_pipeline_output logging_setup _reset _run _evaluate logging_setup _add_pipeline _process _list _add_comment main get_parser _add_dataset restore_dots OrionMeta _merge_meta PipelineField Status OrionDocument remove_dots walk key_has_dollar OrionDBExplorer EventInteraction Experiment Signal Annotation Template Pipeline Datarun Signalrun Event Dataset _overlap _recall _weighted_segment _any_overlap _accuracy _precision _f1_score _pad contextual_f1_score contextual_accuracy contextual_precision _overlap_segment contextual_confusion_matrix _contextual_partition contextual_recall point_f1_score _point_partition point_recall point_precision point_confusion_matrix point_accuracy from_list_points_labels from_pandas_contextual from_list_points_timestamps from_pandas_points from_pandas_points_labels split_sequence _convert_date detect_anomalies _convert_anomalies_to_contextual ThresholdDetector MeanEstimator build_anomaly_intervals _compute_critic_score RandomWeightedAverage score_anomalies TadGAN find_anomalies _find_sequences _get_max_errors _find_threshold _fixed_threshold _compute_scores _merge_sequences _prune_anomalies count_above deltas _find_window_sequences z_cost reconstruction_errors _area_error _point_wise_error _dtw_error regression_errors slice_array_by_dims fillna tadgan_pipline test__build_events_df_events test__build_events_df_empty test__load_pipeline_tadgan tadgan_hyperparameters test__detrend_signal_no_trend TestBenchmark test__load_signal_test_split_float test__detrend_signal_trend test__get_pipeline_hyperparameter_does_not_exist test__sort_leaderboard_rank test__load_signal_test_split_true test__get_pipeline_hyperparameter test__sort_leaderboard_no_rank test__load_signal_test_split_false test__get_pipeline_hyperparameter_dataset test__sort_leaderboard_rank_does_not_exist TestOrion test_load_signal_test_size test_load_signal_nasa_signal_name_multivariate test_download_cached test_download_new test_load_signal_filename test_load_signal_nasa_signal_name TestLoadOrion TestLoadDict TestDetectAnomalies TestEvaluatePipeline TestLoadData TestFitPipeline test__overlap_false test__any_overlap_true expected observed test__overlap_true test__any_overlap_false test__overlap_segment test_contextual_recall test_contextual_accuracy test_contextual_confusion_matrix_overlap_points test__contextual_partition expected_point test_contextual_confusion_matrix_points test_contextual_accuracy_overlap test_contextual_f1_score_nan test_contextual_confusion_matrix test_contextual_confusion_matrix_overlap observed_point test_contextual_f1_score test__overlap_segment_points expected observed test_contextual_precision test_point_precision test_point_f1_score test_point_confusion_matrix test_point_f1_score_nan test_point_recall expected observed test_point_accuracy test__point_partiton assert_list_tuples test_from_list_points_timestamps test_from_pandas_points test_from_list_points_labels test_from_pandas_contextual_severity test_from_pandas_points_error test_from_pandas_contextual test_from_pandas_points_labels_error test_from_pandas_contextual_error test_from_pandas_points_labels PruneAnomaliesTest FindSequencesTest FindAnomaliesTest GetMaxErrorsTest MergeSequencesTest test__area_error RegressionErrorsTest test__point_wise_error ReconstructionErrorsTest test__dtw_error test_slice_dim_error test_fillna_multi_dimensional test_slice_dim_identity test_fillna_ffill_bfill test_slice_dim_axis_zero test_slice_dim_axis_one test_fillna_series series multidimensional signal test_fillna_ffill test_fillna_bfill test_fillna_single_dimension plot_rws convert_date_single unroll_ts plot convert_date plot_error plot_ts run strip sub startswith append run install_minimum run chdir getcwd copy rmtree Path run makedirs glob run run chmod func S_IWRITE rmtree append list basename isinstance load set_hyperparameters MLPipeline debug len predict fit list DataFrame astype _run_pipeline _load_pipeline load_signal isinstance detrend deepcopy join exists lower list zip len pop columns reset_index insert sort_values exception _detrend_signal analyze _load_pipeline total_seconds _load_signal load_anomalies _parse_confusion_matrix utcnow info seed str from_records insert to_csv _evaluate_signal info persist progress delayed concat ProcessPoolExecutor Path values list map append range get _get_pipeline_hyperparameter _run_on_dask info items isinstance to_csv tqdm dict callable makedirs join benchmark tuple join format makedirs to_csv dirname startswith info read_csv exists split download info makedirs dict values read_csv load_csv format_csv isfile download round len concatenate min loads download DataFrame max DataFrame isinstance isinstance _load_dict isinstance _load_data DEFAULT_PIPELINE _load_dict save Orion fit _load_data _load_orion fit _load_data _load_orion LogProgressBar futures_of from_tuples list columns reset_index insert index mean droplevel zip std values pop T reset_index DataFrame sort_values apply dict mean sum to_dict columns set_column isinstance merge_range write MultiIndex copy droplevel to_excel append max enumerate len add_format ExcelWriter get_f1_scores get_summary_page add_sheet save get_output_names append dict items list zip load name end STATUS_SUCCESS id add_signalrun start get_outputs_spec info pipeline process_pipeline_output fit start_signalrun end STATUS_SUCCESS id add_datarun signals start info setFormatter getLogger addHandler StreamHandler Formatter DEBUG setLevel FileHandler format drop_database print eval input user name add_dataset load_signal satellite location value_column start signal stop timestamp_column name add_pipeline user path user text add_comment event to_string format model print limit delete output to_csv getattr empty head method user analyze format print id dataset pipeline user add_pipeline analyze format basename print add_dataset paths pipeline format all tabulate print output to_csv benchmark pipeline add_argument add_mutually_exclusive_group add_parser ArgumentParser set_defaults add_subparsers function _evaluate OrionDBExplorer verbose logging_setup parse_args logfile get_parser database items list isinstance dict transform deepcopy list items isinstance setdefault extend DateTimeField StringField items list StringField IntField ReferenceField StringField PipelineField StringField PipelineField ReferenceField StringField ListField ReferenceField StringField IntField list ReferenceField DateTimeField StringField ListField freeze IntField ReferenceField StringField DateTimeField IntField ReferenceField StringField FloatField IntField ReferenceField StringField ReferenceField StringField _overlap _partition cm cm cm _precision _recall remove _overlap copy update list sorted _any_overlap add set append list itertuples _pad min max int min union set max list min max tolist all list sorted append len list tolist tolist arange len list sorted append len append list len is_anomaly list extend Point timezone Request _convert_date CognitiveServicesCredentials entire_detect zip AnomalyDetectorClient range append len list tuple extend mean zip append enumerate asarray logical_and absolute mean quantile std values list asarray clip reshape multiply tolist min extend _compute_critic_score zscore reconstruction_errors append median max range Series shift sum len count_above deltas inf mean fmin std range mean std all Series tolist index flatten argwhere append max fillna values len append sort_values all mean list std append append sorted average _find_sequences _get_max_errors _find_threshold _fixed_threshold _compute_scores _prune_anomalies list astype extend _merge_sequences mean append _find_window_sequences int len abs trapz apply list flatten pad dtw append len asarray _point_wise_error trunc isinstance min extend _dtw_error _area_error append median max range values len shape isinstance len ndarray isinstance Series copy DataFrame _load_pipeline array _build_events_df DataFrame array _build_events_df assert_frame_equal extend list get_hyperparameters items DataFrame _sort_leaderboard assert_frame_equal DataFrame _sort_leaderboard assert_frame_equal list reset_index assert_frame_equal _sort_leaderboard DataFrame range _detrend_signal DataFrame assert_frame_equal _detrend_signal DataFrame copy assert_frame_equal _get_pipeline_hyperparameter _get_pipeline_hyperparameter _get_pipeline_hyperparameter _load_signal Mock _load_signal DataFrame assert_frame_equal _load_signal Mock assert_called_once_with join download assert_called_once_with join download return_value load_signal assert_frame_equal assert_called_once_with DataFrame load_signal assert_frame_equal assert_not_called assert_called_once_with DataFrame load_signal assert_frame_equal assert_not_called assert_called_once_with DataFrame list load_signal assert_frame_equal DataFrame range _any_overlap _any_overlap list itertuples _contextual_partition assert_array_equal array len list itertuples _overlap_segment assert_array_equal array list itertuples _overlap_segment assert_array_equal array assert_array_equal contextual_confusion_matrix array assert_array_equal contextual_confusion_matrix array assert_array_equal contextual_confusion_matrix array assert_array_equal contextual_confusion_matrix array float contextual_accuracy contextual_precision float float contextual_recall float contextual_f1_score DataFrame contextual_f1_score assert_array_equal list _point_partition array assert_array_equal point_confusion_matrix array float point_accuracy point_precision float point_recall float point_f1_score float point_f1_score DataFrame zip assert_list_tuples DataFrame from_pandas_contextual assert_list_tuples DataFrame from_pandas_contextual DataFrame assert_list_tuples from_list_points_timestamps from_pandas_points assert_list_tuples DataFrame DataFrame assert_list_tuples DataFrame from_pandas_points_labels DataFrame assert_list_tuples from_list_points_labels assert_array_equal _point_wise_error array _area_error array assert_allclose array _dtw_error assert_allclose reshape array signal assert_array_equal slice_array_by_dims multidimensional assert_array_equal slice_array_by_dims multidimensional assert_array_equal slice_array_by_dims signal multidimensional assert_array_equal series array fillna flatten assert_array_equal array fillna assert_array_equal reshape fillna signal assert_array_equal reshape fillna signal assert_array_equal reshape fillna signal assert_array_equal multidimensional array fillna list asarray min append median max range append fromtimestamp list show list plot xlabel yticks add_subplot ylabel xlim title figure legend xticks range len update subplot list show plot set_xticklabels yticks axis GridSpec figure xlim range len set_yticklabels convert_date add_subplot DayLocator set_minor_locator set_major_formatter xticks DataFrame yticks show list set_major_locator itertuples ylabel title DateFormatter axvspan convert_date_single MonthLocator xlim enumerate isinstance xlabel figure len show int list plot set_yticklabels add_subplot tight_layout title ylim figure ceil range len

<p align="left"> <img width=15% src="https://dai.lids.mit.edu/wp-content/uploads/2018/06/Logo_DAI_highres.png" alt=“DAI-Lab” /> <i>An open source project from Data to AI Lab at MIT.</i> </p> <p align="left"> <img width=20% src="https://dai.lids.mit.edu/wp-content/uploads/2018/08/orion.png" alt=“Orion” /> </p> [](https://pypi.org/search/?c=Development+Status+%3A%3A+2+-+Pre-Alpha) [](https://badge.fury.io/py/orion-ml) [](https://pypi.python.org/pypi/orion-ml)

DCSong/CRNN-DenseNet

['optical character recognition', 'scene text recognition']

['An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition']

main_v2.py Config.py infer.py model_v2.py data_generator/generator.py data_generator/transform.py main.py infer_v2.py model.py dictionary.py cv_imread infer Idx2Word decode cv_imread infer Idx2Word decode valid decode_trueLabel minStringDistance calculate_accuracy ImageDataSet collate_fn ctc_greedy_decoder train valid decode_trueLabel minStringDistance calculate_accuracy ImageDataSet collate_fn ctc_greedy_decoder train VGG_16 BLSTM CRNN DenseNet Transition22 Transition21 DenseNet_BLSTM_CTC_MODEL BLSTM random_font sto_choice_from_info_str random_font_size darken_func random_word_color create_an_image random_x_y main append fromfile imdecode argmax numpy tolist join format Idx2Word model print log_softmax eval unsqueeze ctc_greedy_decoder to list LongTensor extend stack zip transform range len append numpy tolist range append argmax numpy tolist zeros min range len minStringDistance zip len eval randint len tolist array choice open listdir crop choice filter choice randint int randint listdir choice random_font str truetype sto_choice_from_info_str Draw random_font_size text size random_word_color darken_func write create_an_image random_x_y save

# CRNN-DenseNet a DenseNet-CRNN implement with PyTorch An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition https://arxiv.org/abs/1507.05717 ## 概况 CRNN是一种经典的用于文本/场景文本识别架构 这里我们用CRNN来进行印刷体文本行的识别 提供了CRNN的两种实现：原始版本的VGG实现和DenseNet作为CNN架构的实现(后缀v2) 文件夹结构和其他说明请见 **源代码说明.txt** <a href="url"><img src="https://github.com/DCSong/CRNN-DenseNet/blob/master/README_imgs/CRNN.jpg" align="middle" height=40% width=40% ></a>

DENG-MIT/CRNN

['time series']

['Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network']

HyChem/gen_data_pyrolysis.py

# CRNN (Chemical Reaction Neural Network) CRNN is an interpretable neural network architecture for autonomously inference chemical reaction pathways in various chemical systems. It is designed based on the following two fundamental physics laws: the Law of Mass Action and Arrhenius Law. It is also possible to incorporate other physics laws to adapt CRNN to a specific domain. <p align="center"> <img src="./assets/CRNN_TOC.png" width="500"> </p> You can find the common questions regarding CRNN in the [FAQs](https://github.com/DENG-MIT/CRNN/wiki/FAQs). # Structure of this repo This repo provides the case studies presented in the original CRNN paper as well as ongoing preliminary results on other systems. Currently, we are actively working on the following systems: * [Biomass pyrolysis kinetics](https://github.com/DENG-MIT/CRNN-Pyrolysis) * [Cell signaling pathways for quantitative modeling drug effects](https://github.com/jiweiqi/CellBox.jl)

DFKI-NLP/REval

['relation extraction']

['Probing Linguistic Features of Sentence-Level Representations in Neural Relation Extraction']

reval/probing_tasks/sent_length.py reval/probing_tasks/probing_task_base.py reval/engine.py reval/__init__.py probing_task_evaluation.py reval/probing_tasks/pos_tag_argument_position.py reval/probing_tasks/entity_exists_between_head_tail.py reval/probing_tasks/sdp_tree_depth.py reval/datasets.py reval.py reval/probing_tasks/__init__.py reval/probing_tasks/entity_type_count_between_head_tail.py reval/dataset_utils.py reval/dependency_graph_utils.py reval/probing_tasks/argument_type.py reval/probing_tasks/argument_order.py reval/probing_task_example.py reval/probing_tasks/argument_grammatical_role.py reval/probing_tasks/tree_depth.py reval/probing_tasks/entity_distance.py run_evaluation _get_parser generate_all_from_semeval generate_all_from_tacred generate load_tacred_dataset load_jsonl_dataset save_probing_task_dataset train_val_split dep_heads_to_tree tree_to_adjacency_list Tree RE ProbingTaskExample find_common_head generate_task_examples generate generate_task_examples generate generate_task_examples generate generate_task_examples generate generate_task_examples generate length_in_bucket generate_task_examples generate generate_task_examples generate generate generate_task_examples in_bucket generate generate_task_examples generate generate_task_examples in_bucket generate get_probing_task_generator EntityTypeCountBetweenHeadTailEval ArgumentOrderEval SDPTreeDepthEval TreeDepthEval PosTagArgPositionEval EntityExistsBetweenHeadTailEval ArgumentTypeEval ArgumentGrammaticalRoleEval REPROBINGEval LengthEval EntityDistanceEval add_argument ArgumentParser eval RE load_predictor info pop seed get get_probing_task_generator save_probing_task_dataset Counter info dataset_loader probing_task_generator partial tacred_generate semeval_generate partial dirname makedirs train_test_split Counter info int list intersection_update add_child len reversed add set difference append range enumerate children popleft extend deque append add set ProbingTaskExample dep_heads_to_tree find_common_head append len extend generate_task_examples train_val_split update absolute_entity_dist_in_bucket enumerate min set max length_in_bucket get

# REval ## Table of Contents * [Introduction](#introduction) * [Overview](#-overview) * [Requirements](#-requirements) * [Installation](#-installation) * [Probing](#-probing) * [Usage](#-usage) * [Citation](#-citation) * [License](#-license)

DFKI-NLP/RelEx

['relation extraction']

['Probing Linguistic Features of Sentence-Level Representations in Neural Relation Extraction']

relex/modules/seq2vec_encoders/bag_of_embeddings_encoder.py relex/predictors/__init__.py relex/modules/nn.py relex/dataset_readers/tacred.py relex/modules/seq2vec_encoders/utils.py tests/models/basic_relation_classifier_test.py relex/modules/offset_embedders/__init__.py relex/metrics/f1_measure.py relex/dataset_readers/__init__.py relex/models/relation_classification/basic_relation_classifier.py scripts/probing_task_evaluation.py relex/modules/offset_embedders/relative_offset_embedder.py relex/__init__.py relex/modules/offset_embedders/offset_embedder.py relex/modules/offset_embedders/sine_offset_embedder.py relex/evaluation/semeval2010_task8_evaluation.py scripts/evaluation_multi.py tests/dataset_readers/semeval2010_task8_test.py relex/dataset_readers/semeval2010_task8.py scripts/evaluation.py tests/dataset_readers/tacred_test.py scripts/summary.py relex/models/__init__.py relex/modules/seq2vec_encoders/gat.py relex/modules/seq2vec_encoders/gcn.py tests/predictors/relation_classifier_predictor_test.py relex/models/model_utils.py relex/modules/seq2vec_encoders/seq2seq_pool_encoder.py relex/predictors/predictor_utils.py scripts/probing_task_evaluation_parallel.py relex/dataset_readers/dataset_reader_utils.py relex/evaluation/tacred_evaluation.py setup.py relex/predictors/relation_classification/relation_classifier_predictor.py relex/modules/offset_embedders/entity_only_offset_embedder.py relex/metrics/__init__.py relex/modules/__init__.py dep_heads_to_tree tree_to_adjacency_list parse_adjacency_indices Tree SemEval2010Task8DatasetReader normalize_glove TacredDatasetReader prec_rec_f1_from_report _write_id_label_file evaluate score evaluate F1Measure batched_predict_instances batched_predict_json BasicRelationClassifier WordDropout EntityOnlyOffsetEmbedder OffsetEmbedder RelativeOffsetEmbedder position_encoding_init SineOffsetEmbedder BagOfEmbeddings GraphAttentionLayer GAT GraphConvolution GCN Seq2SeqPoolEncoder scoped_pool pool PoolingScope load_predictor RelationClassifierPredictor main _get_parser evaluate_multi main _get_parser run_evaluation _get_parser runner run_evaluation_parallel _get_parser experiment_summary main _get_parser TestSemEval2010Task8DatasetReader TestTacredDatasetReader BasicRelationClassifierTest RelationClassifierPredictorTest dep_heads_to_tree len int list intersection_update add_child len reversed add set difference append range enumerate children popleft extend deque append float search group stdout read _write_id_label_file name print batched_predict_instances load_predictor run max list sorted format print write Counter float sum keys range values len predict_batch_json range extend len predict_batch_instance range extend len array cos sin size unsqueeze repeat append long join load_archive add_argument ArgumentParser print parse_args evaluate _get_parser join scorer dict info append walk partial info evaluate_multi experiment_dir eval RE load_predictor info get str run_evaluation put join put Manager Queue append range walk len append join walk scorer experiment_summary

# RelEx A simple framework for Relation Extraction built on AllenNLP. --- ## 🔭&nbsp; Overview | Path | Description | |------------------------- |------------------------------ | | [configs/](configs/) | This directory contains model configurations for relation classification. | | [scripts/](scripts/) | This directory contains scripts, e.g., for evaluating a model with a dataset-specific scorer.| ## ✅&nbsp; Requirements RelEx is tested with:

DFKI-NLP/TRE

['relation extraction', 'unsupervised pre training']

['Improving Relation Extraction by Pre-trained Language Representations']

logging_utils.py analysis_util.py train_utils.py utils.py datasets/__init__.py text_utils.py model_pytorch.py opt.py dataset_converter.py datasets/semeval_2010_task8.py relation_extraction.py loss.py read_experiment_logs experiments_to_dataframe evaluate_semeval2010_task8 read_log_file add_official_scorer_metrics load_experiments_df read_config_file main DatasetConverter ResultLogger MultipleChoiceLossCompute ClassificationLossCompute ClfHead Block load_openai_pretrained_model MLP gelu LayerNorm swish dotdict TransformerModel WordDropout MultipleChoiceHead DoubleHeadModel LMHead Conv1D Attention SimilarityHead warmup_cosine warmup_constant warmup_linear OpenAIAdam evaluate _print_classification_details _print_labeled_confusion_matrix _print_undirected_classifcation_scores _remove_label_direction _get_max_label_length train run_epoch get_pairs LabelEncoder text_standardize Dictionary TextEncoder iter_data iter_predict load_model persist_model iter_apply predict ResultLogger np_softmax _np_init np_init make_path stsb_label_encoding remove_none flatten identity_init _identity_init SemEval2010Task8 list_experiment_dirs read_log_file basename read_config_file join list evaluate_semeval2010_task8 items stdout group search float run append items list extend read_experiment_logs reset_index experiments_to_dataframe subsample dataset_dir output_dir DatasetConverter dataset run load int zip concatenate print cumsum fullmatch from_numpy getattr split open get __delitem__ __setitem__ find zip print confusion_matrix array2string splitlines _get_max_label_length len get list format print set dict mean _remove_label_direction _get_max_label_length append range len print _print_labeled_confusion_matrix classification_report _print_undirected_classifcation_scores get_items model log_test_predictions compute_loss_fct accuracy_score argmax max log get_items iter_data discard _print_classification_details iter_apply get_idx_for_item to predict format set eval f1_score enumerate int log_test_pr_curve print extend log_dev_predictions cpu train len get_base_dir LabelEncoder DataParallel device ResultLogger dotdict max seed list fetch log_dev_labels MASKED_ENTITY_TOKENS ClassificationLossCompute device_count transformer encode encoder to run_epoch CrossEntropyLoss range max_length manual_seed_all format persist_model DoubleHeadModel manual_seed items join load_openai_pretrained_model print min OpenAIAdam parameters transform TextEncoder len log_test_predictions device ResultLogger dotdict accuracy_score list sorted load_model evaluate_semeval2010_task8 NamedTemporaryFile get_idx_for_item dirname encode precision_recall_fscore_support n_ctx to append predict close set realpath zip items remove join _load_from_jsonl log_test_pr_curve print add set sub replace list min len stderr tqdm devnull range open concatenate iter_predict pred_fn make_path join save_to_file join load_from_file astype float32 floor range enumerate exp max dirname makedirs reshape eye

# Improving Relation Extraction by Pre-trained Language Representations This repository contains the code of our paper: [Improving Relation Extraction by Pre-trained Language Representations.](https://openreview.net/forum?id=BJgrxbqp67) Christoph Alt*, Marc Hübner*, Leonhard Hennig We fine-tune the pre-trained OpenAI GPT [1] to the task of relation extraction and show that it achieves state-of-the-art results on SemEval 2010 Task 8 and TACRED relation extraction datasets. Our code depends on huggingface's PyTorch reimplementation of the OpenAI GPT [2] - so thanks to them.

DFKI-NLP/tacrev

['relation extraction', 'unsupervised pre training']

['TACRED Revisited: A Thorough Evaluation of the TACRED Relation Extraction Task']

tacrev/analysis/plotting.py tests/readers/evaluation_results_test.py tests/readers/tacred_test.py tests/__init__.py tacrev/analysis/errudite/utils.py tacrev/writers/writer_utils.py scripts/apply_tacred_patch.py tacrev/analysis/errudite/attributes.py tacrev/readers/tacred.py tacrev/readers/__init__.py tacrev/analysis/errudite/prim_funcs.py tacrev/writers/webanno_v3.py tacrev/analysis/__init__.py tacrev/readers/evaluation_results.py tacrev/analysis/errudite/__init__.py tacrev/definitions.py main read_patch write_tacred read_tacred plot_model_confusion_matrix true_pred_labels_from_dataframe add_annotation_labels_to_df SPAN_DISTANCE HAS_INVERSE_RELATION LABEL_CONTAINS ARG_TYPES_FINE HEAD_SPAN HEAD_TYPE COARSE_SUBJ_OBJ_TYPE NUM_DISTRACTOR_BETWEEN_ARGUMENTS TAIL_SPAN ARG_TYPES_COARSE_FIRST ARG_TYPES_COARSE_SECOND ARG_TYPES_COARSE GET_ENTITY_SPANS INVERSE_LABEL COUNT_SAME_ENTITY_IN_CONTEXT TAIL_TYPE COUNT_ENTITY_IN_CONTEXT strip_end classification_report_from_instances group_info usefulness_score set_predictions_from_df classification_report_from_group load_evaluation_results load_tacred save_as_tsv doc_to_webanno_v3 _highlight_arguments results_as_dataframe documents_as_dataframe _mark_arguments test_load_tacred_results test_load_tacred patch_file update read_patch dataset_file info add_argument Counter set most_common output_file ArgumentParser read_tacred write_tacred parse_args len list remove sorted defaultdict iterrows set figure append most_common DataFrame heatmap values update remove list set apply values rename fillna read_excel head tail head_type tail_type append groupby list len GET_ENTITY_SPANS tail GET_ENTITY_SPANS head remove GET_ENTITY_SPANS set strip_end items list compute_perform set_entries append PredefinedLabel append label remove set get get_instance_list print eval_stats join list defaultdict items endswith EvaluationResult append walk join list defaultdict tokens items tags relations len append head enumerate append DataFrame tokens set_index DataFrame set_index join load_evaluation_results load_tacred print join load_tacred

# TACRED Revisited: A Thorough Evaluation of the TACRED Relation Extraction Task [[Paper](https://arxiv.org/abs/2004.14855)] ## Table of Contents * [Overview](#-overview) * [Requirements](#-requirements) * [Installation](#-installation) * [Patch TACRED](#-patch-the-original-tacred) * [Experiments](#-experiments) * [Citation](#-citation) * [License](#-license) ## 🔭&nbsp; Overview

DIAGNijmegen/neural-odes-segmentation

['medical image segmentation', 'semantic segmentation']

['Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands']

model_utils.py metrics.py dataloader.py models.py train_utils.py augmentations.py inference_utils.py RandomRotationWithMask ElasticTransformations GLaSDataLoader postprocess split_objects remove_small_object inference_image hole_filling_per_object evaluate_image grow_to_fill_borders resize_image crop_result resize_to_size pad_image F1score ObjectDice ObjectHausdorff Hausdorff Dice ConvResUNet LevelBlock ConvResFunc ConvODEFunc Unet ConvODEUNet ConvBlock ODEBlock Conv2dTime Swish get_nonlinearity plot_losses evaluate_image resize_image crop_result array pad_image resize_to_size split_objects remove_small_object hole_filling_per_object grow_to_fill_borders label array eval resize pad max max copy regionprops range maximum_filter binary_fill_holes unique fromarray uint8 size astype resize array uint8 argmin astype delete where any mode unique zeros sum range Hausdorff len uint8 transpose fit astype delete where unique kneighbors max Inf len uint8 astype delete where logical_and flatten any mode unique zeros sum range len uint8 astype delete where shape any mode unique zeros sum range len logical_and show subplots arange plot suptitle cpu transpose imshow legend append numpy nfe len

# Neural Ordinary Differential Equations for Semantic Segmentation of Individual Colon Glands *Accepted to Medical Imaging meets NeurIPS workshop at NeurIPS 2019* Automated medical image segmentation plays a key role in quantitative research and diagnostics. Convolutional neural networks based on the U-Net architecture are the state-of-the-art. A key disadvantage is the hard-coding of the receptive field size, which requires architecture optimization for each segmentation task. Furthermore, increasing the receptive field results in an increasing number of weights. Recently, Neural Ordinary Differential Equations (NODE) have been proposed, a new type of continuous depth deep neural network. This framework allows for a dynamic receptive field at a constant memory cost and a smaller amount of parameters.

DIAGNijmegen/pathology-streaming-pipeline

['whole slide images', 'multiple instance learning']

['Detection of prostate cancer in whole-slide images through end-to-end training with image-level labels']

streaming/train.py streaming/train_remote.py streaming/torch_utils/streaming_trainer.py streaming/experiment_options.py streaming/tissue_dataset.py streaming/torch_utils/checkpointed_trainer.py streaming/torch_utils/utils.py streaming/torch_utils/diagnostics.py streaming/torch_utils/samplers.py streaming/torch_utils/scnn.py streaming/trim_tissue.py streaming/torch_utils/trainer.py ExperimentOptions TissueDataset Experiment DataLoaderX RemoteExperimentOptions initialize_wandb RemoteExperiment fix_seed create_parser TissueExtractor search_empty_regions empty_regions_2d zero_runs extract_tissue CheckpointedTrainer CheckpointedTrainerOptions CheckpointedMultiClassTrainer log_params_norm check_params_distributed DistributedWeightedRandomSampler OrderedDistributedSampler Lost Box StreamingCNN backward_amp_decorator IOShape forward_amp_decorator Sides StreamingConv2dF StreamingConv2d _ntuple StreamingCheckpointedTrainer CheckpointedStreamingMultiClassTrainer StreamingTrainerOptions TrainerOptions Trainer count_parameters progress_bar format_time call watch init seed manual_seed add_argument ArgumentParser reshape diff abs concatenate append sum zero_runs search_empty_regions with_suffix TissueExtractor extract_file stem Path print grad parameters all_gather next range int time join format_time write append range flush len int

Whole-slide classification pipeline &mdash; end-to-end ====== This repository will give an overview on how to use [streaming](https://github.com/DIAGNijmegen/StreamingCNN) to train whole slides to single labels. Streaming is an implementation of convolutions using tiling and gradient checkpointing to save memory.  Papers until now about this method (please consider citing when using this code): - _Application on prostate data, paper:_ H. Pinckaers, W. Bulten, J. Van der Laak and G. Litjens, "Detection of prostate cancer in whole-slide images through end-to-end training with image-level labels," in IEEE Transactions on Medical Imaging, doi: [10.1109/TMI.2021.3066295](https://ieeexplore.ieee.org/document/9380553) - Open Access. - _Methods paper:_ H. Pinckaers, B. van Ginneken and G. Litjens, "Streaming convolutional neural networks for end-to-end learning with multi-megapixel images," in IEEE Transactions on Pattern Analysis and Machine Intelligence, doi: [10.1109/TPAMI.2020.3019563](https://ieeexplore.ieee.org/abstract/document/9178453) - [older preprint](http://arxiv.org/abs/1911.04432) Other resources:

DIAL-RPI/PIPO-FAN

['semantic segmentation']

['Multi-organ Segmentation over Partially Labeled Datasets with Multi-scale Feature Abstraction']

pipo_fan/segment_sf_partial.py pipo_fan/dice.py pipo_fan/model/concave_dps_w.py pipo_fan/resample.py pipo_fan/dataset/dataset_liverCT_2D.py pipo_fan/dataset/dataset_muor_2D.py pipo_fan/model/concave_dps.py pipo_fan/model/denseu_net.py pipo_fan/train_concave0.py pipo_fan/model/resu_net.py pipo_fan/model/unet.py pipo_fan/train_sf_partial.py compute_dice ResampleBySize_view SimpleITKAsNibabel compute_dice SimpleITKAsNibabelHeader make_affine load_network load_image construct_volume extract_volume Nifti_from_numpy validate LoCeLoss2d CrossEntropyLoss2d compute_length AverageMeter LovaszLoss2d adjust_learning_rate save_checkpoint HybridLoss2d DiceLoss FocalLoss2d dice_similarity train validate visualize_val1 CrossEntropyLoss2d visualize_train1 AverageMeter dice_similarity_u adjust_learning_rate save_checkpoint visualize_val dice_similarity train visualize_train ToTensor Clip RandomVerticalFlip RandomCrop Normalize RandomHorizontalFlip get_composed_transform LiverCTDataset ToTensor Clip RandomVerticalFlip RandomCrop Normalize RandomHorizontalFlip get_composed_transform LiverCTDataset outconv ResUNet up one_conv double_conv down res_conv inconv ResUNet attention outconv DenseUNet up_out up_in upblock _DenseLayer _DenseBlock _Transition outconv ResUNet up one_conv double_conv down inconv outconv up double_conv UNet down inconv sum SetOutputSpacing asarray GetSpacing Execute GetSize print ResampleImageFilter sitkLinear tolist WriteImage AffineTransform SetTransform ReadImage SetSize SetReferenceImage SetInterpolator load load ResUNet format print load_state_dict isfile cpu cuda concatenate transpose matmul array diag append range ones shape div cuda append zeros sum max range cat len float sum update data time format criterion model backward print AverageMeter size zero_grad dice_similarity float step cuda enumerate len update data time format criterion model print size AverageMeter eval dice_similarity float cuda enumerate len param_groups copyfile join format save view squeeze mean conv2d unsqueeze float sum size view dice_similarity clone shape zeros range join transpose numpy imsave join numpy imsave join transpose numpy imsave join numpy imsave dice_similarity_u cat long clamp clone clone dice_similarity_u long Compose

# Multi-organ Segmentation over Partially Labeled Datasets with Multi-scale Feature Abstraction ## Introduction In this paper, we propose a novel network architecture for unified multi-scale feature abstraction, which incorporates multi-scale features in a hierarchical fashion at various depths for image segmentation. The 2D network shows very competitive performance compared with other 3D networks in liver CT image segmentation with a single step. We further develop a unified segmentation strategy to train the three separate datasets together and do multi-organ segmentation with these partial datasets. It gives the segmentation network more robustness and accuracy. We'll test the method further in future work. For more details, please refer to our [IEEE TMI paper](https://doi.org/10.1109/TMI.2020.3001036) or the pre-print version available on [arXiv](https://arxiv.org/pdf/2001.00208.pdf). ## Instruction - set up your environment by anaconda, (**python3.7, torch 1.3.0**) - conda install -c simpleitk simpleitk - conda install -c conda-forge nibabel

DIVA-DIA/Text-Line-Segmentation-Method-for-Medieval-Manuscripts

['denoising', 'semantic segmentation']

['Labeling, Cutting, Grouping: an Efficient Text Line Segmentation Method for Medieval Manuscripts']

src/line_segmentation/preprocessing/preprocess.py src/line_segmentation/utils/XMLhandler.py src/line_segmentation/utils/graph_util.py src/pixel_segmentation/evaluation/evaluate_algorithm.py src/line_segmentation/evaluation/overall_score_singel_eval.py src/line_segmentation/evaluation/overall_score.py src/line_segmentation/utils/util.py src/line_segmentation/line_segmentation.py src/line_segmentation/utils/graph_logger.py src/line_segmentation/utils/unused_but_keep_them.py src/line_segmentation/evaluation/evaluator.py src/line_segmentation/polygon_manager.py src/pixel_segmentation/evaluation/apply_postprocess.py src/optimization/optimizationLoop.py src/line_segmentation/bin_algorithm.py src/line_segmentation/evaluation/grid_search_eval.py src/line_segmentation/preprocessing/energy_map.py src/line_segmentation/seamcarving_algorithm.py src/line_segmentation/evaluation/evaluate_algorithm.py src/line_segmentation/preprocessing/load_image.py src/optimization/sigoptManager.py compute_avg_pairwise_distance split_into_bins_and_index merge_small_bins draw_bins check_for_anomaly count_seams_below pairwise majority_voting extract_textline init_logger find_cc_from_centroid get_polygons_from_lines find_graph_node polygon_to_string draw_polygons measure_energy draw_seams_red draw_seams post_process_seams get_seams non_zero_runs horizontal_seam get_score get_file_list evaluate compute_for_all check_extension get_score get_file_list evaluate compute_for_all check_extension _get_lines write_stats check_extension get_file_list _get_lines write_stats create_distance_matrix create_projection_profile smooth find_cc_centroids_areas create_energy_map get_connected_components create_heat_map_visualization detect_outliers prepare_energy prepare_image load_image preprocess wipe_outside_textarea remove_small_components remove_big_components GraphLogger get_neighbouring_seams_index find_intersected_edges cut_graph_with_seams createTINgraph get_edge_node_coordinates chunks graph_to_point_lists detect_small_graphs merge_small_graphs asymetric_distance print_graph_on_img blow_up_image blur_image save_img create_folder_structure calculate_asymmetric_distance dict_to_string writePAGEfile prettify read_max_textline_from_file evaluate_metric delete_with_pattern print_with_pattern retrieve_id_by_name apply_preprocess get_score evaluate compute_for_all time asarray list format split_into_bins_and_index merge_small_bins mean info append count_seams_below zeros int enumerate expand_dims min digitize unique append pairwise pairwise next tee str int LINE_AA addWeighted draw_seams_red reshape putText save_img FONT_HERSHEY_SIMPLEX copy flatten stack zip fill zeros circle post_process_seams prepare_image draw_seams_red majority_voting shape init_logger load_image format create_folder_structure get_seams create_energy_map preprocess writePAGEfile create_heat_map_visualization info join time draw_seams get_polygons_from_lines save_img polygon_to_string join setFormatter getLogger addHandler StreamHandler Formatter setLevel INFO FileHandler fillPoly print_graph_on_img find_cc_from_centroid list ROTATE_90_CLOCKWISE ones rotate shape array append asarray createTINgraph set find_graph_node zeros enumerate minimum_spanning_tree filter2D centroid coords asarray print polylines array ROTATE_90_CLOCKWISE rotate int time format join info append enumerate append argmin range list polylines int32 expand_dims array range enumerate len list polylines int32 expand_dims array range enumerate len extend copy horizontal_seam prepare_energy array range measure_energy list non_zero_runs range enumerate reshape diff abs concatenate join sorted sort append check_extension walk findall str extract_textline str format replace print Popen starmap get_file_list Pool write_stats list len append format close unlink mean zip listdir join time isdir print extend dict rmtree repeat isfile makedirs list array join get_file_list time format applyColorMap copy stack array info COLORMAP_JET max maxsize minimum max time asarray format ones reshape slice min astype info zeros array time asarray create_distance_matrix format ones reshape transpose min find_cc_centroids_areas create_projection_profile filter2D mean blur_image info zeros max sum max min ones sum eval convolve time asarray format get_connected_components mean info detect_outliers abs std bbox len label regionprops time format mean info std imread asarray stack open time format ROTATE_90_COUNTERCLOCKWISE where rotate info time wipe_outside_textarea format remove_big_components remove_small_components info ones fillPoly stack filter2D mean label regionprops mean label regionprops sorted time format info nsimplex Graph Delaunay add set add_weighted_edges_from set_node_attributes range len edges line get_edge_node_coordinates copy asarray get_node_attributes connected_component_subgraphs time format find_intersected_edges asarray list is_connected merge_small_graphs tolist detect_small_graphs remove_edges_from info int min ceil range len max get_neighbouring_seams_index sort min intersects chunks unique zip append LineString array get_node_attributes pop asarray hstack delete unique add_weighted_edges_from append time asarray format info time format insert copy shape info append array range len imwrite ones waitKey filter2D imshow zeros destroyAllWindows join format mkdir imshow waitKey imwrite destroyAllWindows append join list items SubElement Element prettify write close strftime set open enumerate findall parse max getroot tostring parseString delete append id data str sorted fetch print len save_img where mkdir zip load_image

**&rarr; Link to paper: [https://arxiv.org/abs/1906.11894](https://arxiv.org/abs/1906.11894)** # Text Line Segmentation Method for Medieval Manuscripts Image and Text Segmentation pipeline for the paper ["Labeling, Cutting, Grouping: an Efficient Text Line Segmentation Method for Medieval Manuscripts"](https://arxiv.org/abs/1906.11894), published at the 15th IAPR International Conference on Document Analysis and Recognition (ICDAR) in 2019. ## Getting started In order to get the pipeline up and running it is only necessary to clone the latest version of the repository: ``` shell git clone https://github.com/DIVA-DIA/Text-Line-Segmentation-Method-for-Medieval-Manuscripts ``` Run the install the conda environment: ``` shell

DLTK/DLTK

['semantic segmentation']

['DLTK: State of the Art Reference Implementations for Deep Learning on Medical Images']

dltk/core/upsample.py tests/test_sliding_window_segmentation.py dltk/io/abstract_reader.py dltk/networks/autoencoder/convolutional_autoencoder.py dltk/networks/segmentation/fcn.py data/IXI_Guys/download_IXI_Guys.py dltk/version.py dltk/core/residual_unit.py examples/applications/IXI_HH_superresolution/train.py dltk/core/losses.py data/IXI_HH/download_IXI_HH.py examples/applications/IXI_HH_DCGAN/train.py dltk/core/metrics.py examples/applications/MRBrainS13_tissue_segmentation/reader.py examples/applications/IXI_HH_DCGAN/reader.py dltk/utils.py examples/applications/IXI_HH_sex_classification_resnet/train.py examples/applications/IXI_HH_sex_classification_resnet/reader.py dltk/io/augmentation.py examples/applications/IXI_HH_superresolution/reader.py dltk/io/preprocessing.py dltk/networks/segmentation/unet.py tests/test_activations.py docs/source/conf.py examples/applications/IXI_HH_age_regression_resnet/reader.py examples/applications/IXI_HH_representation_learning_cae/reader.py examples/applications/IXI_HH_age_regression_resnet/train.py dltk/networks/segmentation/deepmedic.py dltk/networks/gan/dcgan.py examples/applications/IXI_HH_sex_classification_resnet/deploy.py examples/applications/IXI_HH_representation_learning_cae/train.py dltk/networks/super_resolution/simple_super_resolution.py examples/applications/MRBrainS13_tissue_segmentation/train.py setup.py examples/applications/MRBrainS13_tissue_segmentation/deploy.py dltk/networks/regression_classification/resnet.py dltk/core/activations.py examples/applications/IXI_HH_age_regression_resnet/deploy.py resample_image reslice_image resample_image reslice_image SlidingWindow sliding_window_segmentation_inference leaky_relu prelu sparse_balanced_crossentropy dice_loss crossentropy dice abs_vol_difference vanilla_residual_unit_3d get_linear_upsampling_kernel linear_upsample_3d Reader IteratorInitializerHook extract_class_balanced_example_array extract_random_example_array add_gaussian_noise elastic_transform add_gaussian_offset flip normalise_one_one whitening resize_image_with_crop_or_pad normalise_zero_one convolutional_autoencoder_3d dcgan_discriminator_3d dcgan_generator_3d resnet_3d deepmedic_3d crop_central_block upscore_layer_3d residual_fcn_3d asymmetric_residual_unet_3d residual_unet_3d upsample_and_concat simple_super_resolution_3d predict read_fn train model_fn read_fn train read_fn train model_fn predict read_fn train model_fn read_fn train model_fn predict read_fn train model_fn test_leaky_relu test_sw_inference SetOutputSpacing GetPixelIDValue SetDefaultPixelValue GetSpacing GetOrigin Transform ResampleImageFilter GetSize SetOutputDirection SetOutputOrigin GetDirection SetTransform sitkBSpline SetSize sitkNearestNeighbor SetInterpolator ResampleImageFilter SetInterpolator sitkBSpline SetReferenceImage sitkNearestNeighbor as_list list SlidingWindow len shape zip append next array range run get_variable as_list constant one_hot reshape reduce_sum bincount reduce_mean softmax cast int32 stop_gradient tiny log len one_hot is_nan reshape boolean_mask reduce_sum logical_not reduce_mean softmax equal zeros sum range sum zeros float abs range exp mean sum log amax pool_op max_pooling3d list zeros_like tuple constant_initializer zeros float abs array range len as_list list format tuple shape get_linear_upsampling_kernel set_shape info conv3d_transpose len ndarray isinstance random range len normal ndim list broadcast_arrays reshape rand ndim map shape zeros array range gaussian_filter len int list isinstance concatenate tuple astype enumerate floor argwhere zip ceil round array range append len ndarray range isinstance len mean astype float32 std max min astype float32 normalise_zero_one int slice floor range len as_list get_shape dense format list reshape conv3d reversed conv_op info range conv3d_transpose len get_shape format conv3d info range len as_list get_shape dense format reshape conv3d sigmoid cast int32 info range len get_shape format conv3d info append relu6 range len as_list slice range len get_shape concat _build_normal_pathway crop_central_block info append _build_subsampled_pathways range len batch_normalization conv3d linear_upsample_3d get_shape format conv3d info append range len linear_upsample_3d get_shape format conv3d info append range len get_shape format conv3d info append range len get_shape format conv3d info range conv3d_transpose len time format from_saved_model print read_fn extract_random_example_array mean as_matrix append abs run join str format astype float32 extract_random_example_array ReadImage GetArrayFromImage whitening _augment float range get_global_step get_collection mean_absolute_error AdamOptimizer resnet_3d UPDATE_OPS root_mean_squared_error mean_squared_error seed join format evaluate print SummaryAtEndHook get_inputs Estimator set_random_seed StepCounterHook export_savedmodel Reader as_matrix run_validation model_path range normal normalise_one_one zeros_like TRAINABLE_VARIABLES MonitoredTrainingSession run get_collection assign_add cast ones_like get_or_create_global_step train_input_fn minimize greater float32 accuracy mean_squared_error scalar convolutional_autoencoder_3d argmax int int32 one_hot reshape accuracy precision softmax_cross_entropy linear_upsample_3d image average_pooling3d simple_super_resolution_3d join CopyInformation astype WriteImage nanmean GetImageFromArray int32 model_path expand_dims extract_class_balanced_example_array len MomentumOptimizer float32 sparse_softmax_cross_entropy_with_logits residual_unet_3d reduce_mean py_func leaky_relu constant ones float32 placeholder

## Deep Learning Toolkit (DLTK) for Medical Imaging [](https://gitter.im/DLTK/DLTK?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge) [](https://coveralls.io/github/DLTK/DLTK?branch=dev) [](https://travis-ci.org/DLTK/DLTK)  DLTK is a neural networks toolkit written in python, on top of [TensorFlow](https://github.com/tensorflow/tensorflow). It is developed to enable fast prototyping with a low entry threshold and ensure reproducibility in image analysis applications, with a particular focus on medical imaging. Its goal is to provide the community with state of the art methods and models and to accelerate research in this exciting field. ### Documentation The DLTK API can be found [here](https://dltk.github.io/) ### Referencing and citing DLTK If you use DLTK in your work please refer to this citation for the current version:

DNNToolBox/Net-Trim-v1

['network pruning']

['Fast Convex Pruning of Deep Neural Networks']

main.py NetTrimSolver.py GeneralSoftmaxModel.py GeneralSoftmaxModel train_neural_network NetTrimSolver net_trim_solver_np get_weights initialize format compute_signals print train len labels GeneralSoftmaxModel images savemat zip compute_accuracy range read_data_sets count T reshape transpose maximum matmul solve_triangular cholesky append zeros kron range

# Net-Trim (first version) This page contains the first version of Net-Trim, which addresses the **regularized form** of the Net-Trim convex program discussed in the [NIPS (2017) paper](https://papers.nips.cc/paper/6910-net-trim-convex-pruning-of-deep-neural-networks-with-performance-guarantee). See *main.py* for an example of network training followed by a pruning. The *Additional Material* folder contains the NIPS paper, presentation and supplementary files. ## New Version with Manual Available Starting June 20th, 2018, a new version of Net-Trim addressing the **original constrained form** of the convex program is available. The code, along with a comprehensive manual and the new version of the paper are available at the following links: * [**Latest Net-Trim Website and Manual**](https://dnntoolbox.github.io/Net-Trim/) * [**Latest Code Github Page**](https://github.com/DNNToolBox/Net-Trim) * [**2018 Journal Preprint**](https://arxiv.org/pdf/1806.06457.pdf)

DTaoo/Discriminative-Sounding-Objects-Localization

['object localization']

['Discriminative Sounding Objects Localization via Self-supervised Audiovisual Matching']

music-exp/training_stage_two.py compared model/Sound-of-Pixels/dataset/music.py compared model/Sound-of-Pixels/create_index_files.py music-exp/eval_duet.py music-exp/data/syn_dataset.py compared model/Sound-of-Pixels/utils.py music-exp/model/location_model.py music-exp/training_stage_one.py compared model/Sound-of-Pixels/dataset/video_transforms.py audioset-instrument/scripts/cut_video.py audioset-instrument/scripts/generate_test.py audioset-instrument/model/location_model.py compared model/Sound-of-Pixels/models/vision_net.py music-exp/data/cut_audios.py compared model/Sound-of-Pixels/dataset/base.py compared model/attention_net.py music-exp/training_stage_two_duet.py audioset-instrument/scripts/cut_audio.py audioset-instrument/training_stage_two.py compared model/visualization_attention.py compared model/Sound-of-Pixels/models/__init__.py music-exp/data/cut_videos.py music-exp/eval.py music-exp/test.py audioset-instrument/scripts/conver_to_audio.py compared model/Sound-of-Pixels/models/synthesizer_net.py music-exp/model/base_model2.py compared model/Sound-of-Pixels/arguments.py music-exp/test_stage_two_duet.py music-exp/tools.py music-exp/match_cluster.py compared model/Sound-of-Pixels/main.py compared model/Sound-of-Pixels/dataset/__init__.py music-exp/model/base_model.py music-exp/test_stage_two.py compared model/location_attention_stage_one.py compared model/Sound-of-Pixels/viz.py audioset-instrument/training_stage_one.py compared model/Sound-of-Pixels/models/audio_net.py compared model/Sound-of-Pixels/models/criterion.py compared model/location_dmc_stage_one.py audioset-instrument/model/base_model.py music-exp/data/MUSIC_dataset.py audioset-instrument/data/audioset.py compared model/dmc_model.py location_model_eva extract_feature location_model_train class_model_val feature_clustering batch_organize class_model_train location_dilation main eva_metric main batch_organize eva_metric location_model_train Audioset_AV_Classify augment_image Audioset_Dataset conv1x1 ResNet Bottleneck conv3x3 _resnet resnet18 BasicBlock Location_Net_stage_two Location_Net_stage_one find_file_using_prefix find_file_using_prefix audio_extract video2frame_update find_file_using_prefix video2frame find_file_using_prefix Attention_Net DMC_NET Cluster_layer location_model_eva location_model_train eva_metric2 batch_organize ContrastiveLoss main eva_metric location_model_eva location_model_train eva_metric2 batch_organize ContrastiveLoss main eva_metric location_model_eva location_model_train batch_organize returnCAM main eva_metric ArgParser find_recursive calc_metrics create_optimizer evaluate NetWrapper adjust_learning_rate output_visuals main train checkpoint kill_proc combine_video_audio save_audio AverageMeter magnitude2heatmap save_video warpgrid VideoWriter istft_reconstruction run_proc_timeout recover_rgb makedirs HTMLVisualizer plot_loss_metrics BaseDataset MUSICMixDataset Stack CenterCrop ToTensor Resize RandomCrop Normalize RandomHorizontalFlip UnetBlock Unet L1Loss L2Loss BCELoss BaseLoss InnerProd Bias ResnetDilated ResnetFC Resnet ModelBuilder activate visualize_model batch_organize returnCAM main eva_metric visualize_model batch_organize returnCAM main eva_metric visualize visualize_model batch_organize returnCAM main eva_metric cal_ciou visualize location_model_eva extract_feature location_model_train class_model_val feature_clustering batch_organize location_model_test class_model_train location_dilation main eva_metric visualize location_model_eva location_model_train batch_organize main eva_metric location_model_eva location_model_train batch_organize main eva_metric audio_extract video2frame_update video2frame MUSIC_Dataset MUSIC_AV_Classify augment_image MUSIC_Dataset augment_image conv1x1 ResNet Bottleneck conv3x3 _resnet resnet18 BasicBlock conv1x1 ResNet Bottleneck conv3x3 _resnet resnet18 BasicBlock Location_Net_stage_two Location_Net_stage_one zeros range sum uint8 range zeros_like time criterion model print batch_organize backward zero_grad localtime numpy asctime train step eva_metric enumerate len eval dataset len print eval dataset len normalized_mutual_info_score len labels_ append normalize range predict fit a_class_layer model zero_grad DataParallel v_class_layer list Adam normal_ to range zero_ class_iter enumerate Linear criterion backward print parameters train step print eval location_model_train DataLoader DataParallel ArgumentParser device save list extract_feature data_dir exit Adam Audioset_Dataset class_model_train load_state_dict resnet18 parse_args to append CrossEntropyLoss range state_dict Location_Net_stage_one location_model_eva epoch feature_clustering mean BCELoss load join Audioset_AV_Classify class_model_val print add_argument parameters use_pretrain ckpt_file zeros_like criterion_class cuda criterion_location detach eval softmax tile argsort Location_Net_stage_two FLIP_LEFT_RIGHT transpose Brightness random Color enhance print ResNet load load_state_dict listdir load join int power_to_db format stft set_trace astype mkdir melspectrogram range len get VideoCapture read format int tofile CAP_PROP_FPS get VideoCapture int read format tofile mkdir CAP_PROP_FPS CAP_PROP_FRAME_COUNT int range eva_metric2 Attention_Net ContrastiveLoss cuda MUSIC_Dataset data_list_dir DMC_NET uint8 min max append join filter walk update asarray grid_sample bss_eval_sources size AverageMeter binary_mask astype float32 mean log_freq num_mix device istft_reconstruction to numpy range device istft_reconstruction combine_video_audio save_video log_freq num_mix audRate append to imsave range format asarray grid_sample size astype vis binary_mask join uint8 magnitude2heatmap float32 write numpy makedirs HTMLVisualizer ckpt set_grad_enabled forward calc_metrics plot_loss_metrics num_mix append range update format vis mean eval add_header item enumerate join print AverageMeter write_html average output_visuals add_rows makedirs set_grad_enabled zero_grad forward append update format lr_sound synchronize lr_frame perf_counter mean lr_synthesizer item epoch_iters enumerate backward print AverageMeter average step format ckpt print save state_dict param_groups list_val build_synthesizer batch_size num_epoch adjust_learning_rate ModelBuilder MUSICMixDataset format create_optimizer build_frame build_criterion epoch_iters checkpoint evaluate NetWrapper build_sound train list_train len power astype float32 linspace meshgrid zeros log print rmtree isdir uint8 add_ astype zip uint8 applyColorMap astype log10 COLORMAP_JET complex exp astype istft print kill communicate start Timer print run_proc_timeout format Popen add_frame shape VideoWriter range release write_wav join plot close savefig figure legend eval dataset cuda len visualize_model load dump open uint8 applyColorMap astype range array resize COLORMAP_JET numpy max imsave load max print len resize append zeros auc sum array range open str clip dump len eval dataset open argmax cuda cluster numpy open MUSIC_AV_Classify fill_ location_model_test mask normalize dump concatenate cluster criterion_category view cat log_softmax contiguous repeat cuda KLDivLoss weight print print print

# **Discriminative Sounding Objects Localization** Code for our NeurIPS 2020 paper [**Discriminative Sounding Objects Localization via Self-supervised Audiovisual Matching**](https://arxiv.org/abs/2010.05466) (The previous title is **Learning to Discriminatively Localize Sounding Objects in a Cocktail-party Scenario**). The code is implemented on PyTorch with python3. The code for TPAMI version can be found here: https://github.com/GeWu-Lab/CSOL_TPAMI2021. ### Requirements - PyTorch 1.1 - torchvision - scikit-learn - librosa - Pillow - opencv

DVLP-CMATERJU/Skip-Connected-Multi-column-Network

['optical character recognition']

['A Skip-connected Multi-column Network for Isolated Handwritten Bangla Character and Digit recognition']

Multi-Scale-Multi-Column-CNN/Other Datasets/Level_Wise_Concatenation/MSCNN_Level_Wise.py Multi-Scale-Multi-Column-CNN/Column_Wise_Concatenation/MSCNN_Column_Wise.py Multi-Scale-Multi-Column-CNN/All_Feature_Concatenation/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Level_Wise_Concatenation/MSCNN_Level_Wise.py Multi-Scale-Multi-Column-CNN/NormalCNN/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Multi_Column/MSCNN_Multi_Column.py Multi-Scale-Multi-Column-CNN/NormalCNN/MSCNN_NormalCNN.py Multi-Scale-Multi-Column-CNN/Column_Wise_Concatenation/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Level_Wise_Concatenation/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Other Datasets/All_Feature_Concatenation/MSCNN_All_Feature_Concatenation.py Multi-Scale-Multi-Column-CNN/Other Datasets/Multi_Column/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Other Datasets/Multi_Column/MSCNN_Multi_Column.py Multi-Scale-Multi-Column-CNN/Other Datasets/Column_Wise_Concatenation/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Other Datasets/NormalCNN/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Other Datasets/NormalCNN/MSCNN_NormalCNN.py Multi-Scale-Multi-Column-CNN/Multi_Column/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/Other Datasets/Column_Wise_Concatenation/MSCNN_Column_Wise.py Multi-Scale-Multi-Column-CNN/Other Datasets/Level_Wise_Concatenation/Softmax-SVM.py Multi-Scale-Multi-Column-CNN/All_Feature_Concatenation/MSCNN_All_Feature_Concatenation.py Multi-Scale-Multi-Column-CNN/Other Datasets/All_Feature_Concatenation/Softmax-SVM.py generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval generateNum eval_cpu Skip_Model train_cpu save_checkpoint Skip_Model flatten eval append range data time loss_fun model Variable backward print zero_grad cpu train step max enumerate data model squeeze eval cpu max enumerate print save append model tolist ravel cpu numpy append enumerate

# Skip-Connected-Multi-column-Network

Daftstone/Random-Directional-Attack

['speech recognition']

['Random Directional Attack for Fooling Deep Neural Networks']

cleverhans/scripts/plot_success_fail_curve.py cleverhans/cleverhans/canary.py cleverhans/tests_tf/test_mnist_tutorial_cw.py cleverhans/cleverhans/picklable_model.py cleverhans/cleverhans_tutorials/mnist_tutorial_picklable.py cleverhans/cleverhans/utils.py cleverhans/examples/multigpu_advtrain/test_run_multigpu.py cleverhans/scripts/make_confidence_report_spsa.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/fake_cloud_client.py models/svhn_model.py cleverhans/cleverhans/utils_keras.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/worker.py cleverhans/cleverhans/evaluation.py utils.py cleverhans/examples/test_imagenet_attacks.py cleverhans/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_submission_lib.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_targeted_attacks/step_target_class/attack_step_target_class.py cleverhans/cleverhans/confidence_report.py cleverhans/cleverhans/plot/pyplot_image.py models/imagenet_model.py models/cifar10_model.py cleverhans/cleverhans/model_zoo/all_convolutional.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/image_batches_test.py cleverhans/tests_tf/test_attacks.py cleverhans/tests_tf/test_picklable_model.py cleverhans/examples/multigpu_advtrain/test_runner.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py cleverhans/examples/multigpu_advtrain/resnet_tf.py cleverhans/examples/multigpu_advtrain/model.py cleverhans/cleverhans_tutorials/mnist_tutorial_pytorch.py cleverhans/cleverhans/devtools/tests/test_format.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_attacks/fgsm/attack_fgsm.py cleverhans/scripts/print_report.py cleverhans/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py cleverhans/tests_tf/test_mnist_tutorial_jsma.py cleverhans/tests_tf/test_mnist_tutorial_keras_tf.py cleverhans/cleverhans/attacks_tf.py cleverhans/cleverhans/utils_mnist.py cleverhans/tests_pytorch/test_mnist_tutorial_pytorch.py cleverhans/cleverhans_tutorials/tutorial_models_tfe.py cleverhans/cleverhans/utils_tfe.py cleverhans/tests_tf/test_utils.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/fake_cloud_client_test.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py cleverhans/tests_tf/test_confidence_report.py cleverhans/cleverhans/model_zoo/madry_lab_challenges/cifar10_model.py cleverhans/cleverhans/plot/success_fail.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/master.py white_box.py cleverhans/cleverhans/attacks_tfe.py cleverhans/tests_tf/test_utils_tf.py cleverhans/examples/multigpu_advtrain/run_multigpu.py cleverhans/cleverhans/augmentation.py cleverhans/examples/madry_lab_challenges/mnist/madry_mnist_model.py cleverhans/cleverhans/initializers.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/adv_inception_v3/defense.py cleverhans/cleverhans/experimental/certification/tests/dual_formulation_test.py cleverhans/examples/multigpu_advtrain/utils_svhn.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/work_data_test.py cleverhans/cleverhans/serial.py cleverhans/examples/imagenet_featadvs/model.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/image_batches.py cleverhans/examples/RL-attack/model.py cleverhans/tests_tf/test_evaluation.py cleverhans/cleverhans/experimental/certification/read_weights.py cleverhans/cleverhans/experimental/certification/tests/optimization_test.py cleverhans/cleverhans/experimental/certification/certify.py cleverhans/examples/multigpu_advtrain/runner.py cleverhans/cleverhans/model_zoo/__init__.py cleverhans/cleverhans_tutorials/evaluate_pickled_model.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py cleverhans/cleverhans/utils_pytorch.py cleverhans/cleverhans_tutorials/tutorial_models.py cleverhans/examples/multigpu_advtrain/trainer.py cleverhans/docs/conf.py cleverhans/cleverhans/loss.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/__init__.py cleverhans/scripts/make_confidence_report.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/submissions_test.py cleverhans/cleverhans/__init__.py models/fmnist_model.py models/mnist_model.py cleverhans/cleverhans/train.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/base_inception_model/defense.py cleverhans/tests_tf/test_utils_keras.py cleverhans/tests_tf/test_attack_bundling.py cleverhans/scripts/make_confidence_report_bundle_examples.py cleverhans/examples/multigpu_advtrain/utils_cifar.py cleverhans/examples/multigpu_advtrain/attacks_multigpu.py cleverhans/tests_tf/test_serial.py cleverhans/cleverhans/devtools/mocks.py cleverhans/cleverhans/devtools/checks.py cleverhans/examples/madry_lab_challenges/cifar10/attack_model.py cleverhans/cleverhans/experimental/certification/tests/neural_net_params_test.py cleverhans/tests_tf/test_defenses.py cleverhans/scripts/compute_accuracy.py cleverhans/setup.py cleverhans/cleverhans_tutorials/mnist_tutorial_cw.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/validation_tool/validate_submission.py cleverhans/tests_tf/test_mnist_tutorial_tf.py cleverhans/cleverhans/model_zoo/madry_lab_challenges/__init__.py cleverhans/examples/multigpu_advtrain/make_model.py cleverhans/cleverhans_tutorials/cifar10_tutorial_tf.py cleverhans/cleverhans_tutorials/mnist_tutorial_tfe.py cleverhans/examples/multigpu_advtrain/utils.py cleverhans/cleverhans/devtools/autopep8_all.py cleverhans/examples/robust_vision_benchmark/cleverhans_attack_example/main.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/submissions.py cleverhans/cleverhans/compat.py cleverhans/cleverhans/attacks/__init__.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/validation_tool/submission_validator_lib.py cleverhans/cleverhans/plot/save_pdf.py cleverhans/examples/facenet_adversarial_faces/set_loader.py cleverhans/examples/RL-attack/enjoy-adv.py cleverhans/tests_tf/test_dataset.py cleverhans/cleverhans_tutorials/mnist_blackbox.py cleverhans/tests_tf/test_projected_gradient_descent.py cleverhans/scripts/make_confidence_report_bundled.py cleverhans/cleverhans/model.py cleverhans/cleverhans_tutorials/mnist_tutorial_jsma.py cleverhans/cleverhans/utils_tf.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_targeted_attacks/iter_target_class/attack_iter_target_class.py cleverhans/cleverhans_tutorials/__init__.py cleverhans/cleverhans/plot/image.py cleverhans/examples/imagenet_featadvs/attack_model_featadv.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/dataset_helper.py cleverhans/examples/madry_lab_challenges/mnist/attack_model.py cleverhans/scripts/show_images.py cleverhans/cleverhans/experimental/certification/optimization.py cleverhans/examples/multigpu_advtrain/test_attack_multigpu.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/defense.py cleverhans/cleverhans/experimental/certification/dual_formulation.py cleverhans/cleverhans/devtools/version.py cleverhans/cleverhans/plot/__init__.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/tests/classification_results_test.py cleverhans/examples/RL-attack/train.py cleverhans/examples/facenet_adversarial_faces/facenet_fgsm.py cleverhans/cleverhans/attack_bundling.py cleverhans/cleverhans_tutorials/mnist_tutorial_tf.py cleverhans/tests_tf/test_attacks_tf.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/classification_results.py cleverhans/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/cloud_client.py black_box.py cleverhans/cleverhans_tutorials/mnist_tutorial_keras_tf.py cleverhans/examples/nips17_adversarial_competition/dataset/download_images.py attacks.py cleverhans/examples/multigpu_advtrain/evaluator.py cleverhans/tests_tf/test_mnist_blackbox.py cleverhans/examples/robust_vision_benchmark/cleverhans_attack_example/utils.py cleverhans/cleverhans/devtools/list_files.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_attacks/noop/attack_noop.py cleverhans/cleverhans/model_zoo/madry_lab_challenges/make_cifar10_joblib.py cleverhans/examples/nips17_adversarial_competition/dev_toolkit/sample_attacks/random_noise/attack_random_noise.py cleverhans/cleverhans/experimental/certification/neural_net_params.py cleverhans/cleverhans/experimental/certification/utils.py cleverhans/cleverhans/plot/pyplot_defaults.py cleverhans/cleverhans/dataset.py datasets.py cleverhans/tests_tf/test_model.py get_adv_examples main attack data_svhn load_images data_fmnist data_cifar10 data_mnist data_imagenet imagenet_model GetGradient get_metrix_index get_gradient train mnist_model get_value imagenet_train get_adv cal_dot svhn_model get_model cifar10_model RDA main attack CarliniWagnerL2 jsma_symbolic deepfool_batch jacobian pgd_attack apply_perturbations LBFGS_attack jsma saliency_map vatm UnrolledOptimizer TensorAdam parallel_apply_transformations jacobian_graph TensorOptimizer spm TensorGradientDescent deepfool_attack UnrolledGradientDescent margin_logit_loss SPSAAdam fgm ZERO jacobian_augmentation _apply_transformation _apply_black_border fgsm UnrolledAdam ElasticNetMethod Attack FastGradientMethod BasicIterativeMethod fixed_max_confidence_recipe unfinished_attack_configs bundle_examples_with_goal bundle_attacks Misclassify single_run_max_confidence_recipe AttackConfig _WrongConfidenceFactory run_batch_with_goal _ExtraCriteriaFactory basic_max_confidence_recipe MaxConfidence save random_search_max_confidence_recipe spsa_max_confidence_recipe bundle_attacks_with_goal AttackGoal _CriteriaFactory batch_augment random_crop_and_flip random_shift run_canary softmax_cross_entropy_with_logits reduce_any reduce_max reduce_sum reduce_function reduce_prod reduce_mean reduce_min make_confidence_report make_confidence_report_bundled ConfidenceReport ConfidenceReportEntry print_stats MNIST Factory data_mnist data_cifar10 CIFAR10 maybe_download_file Dataset download_and_parse_mnist_file _AttackFactory correctness_and_confidence _check_x batch_eval batch_eval_multi_worker accuracy _CorrectFactory _CorrectAndProbFactory run_attack _check_y HeReLuNormalInitializer LossMixUp WeightedSum MixUp Loss CrossEntropy FeaturePairing WeightDecay LossCrossEntropy LossFeaturePairing wrapper_warning_logits CallableModelWrapper Model NoSuchLayerError wrapper_warning ResidualWithGroupNorm Softmax Sigmoid TanH GroupNorm ReLU Flatten Add Layer Dropout BatchNorm LeakyReLU PicklableModel GlobalAveragePool MLP ELU Tanh PerImageStandardize ResidualWithBatchNorm Linear Conv2D Print SELU PicklableVariable load NoRefModel save train avg_grads get_log_level safe_zip ordered_union to_categorical grid_visual deep_copy create_logger get_logits_over_interval deterministic_dict TemporaryLogLevel set_log_level random_targets pair_visual batch_indices other_classes shell_call AccuracyReport linear_extrapolation_plot _ArgsWrapper KerasModelWrapper cnn_model conv_2d data_mnist download_and_parse_mnist_file maybe_download_mnist_file convert_pytorch_model_to_tf _py_func_with_gradient mul model_train div clip_by_value assert_less_equal infer_devices assert_greater_equal tf_model_load clip_eta kl_with_logits model_loss get_available_gpus l2_batch_normalize model_argmax assert_equal initialize_uninitialized_global_variables model_eval batch_eval op_with_scalar_cast train silence model_eval train model_argmax CarliniWagnerL2 projected_optimization LBFGS_impl LBFGS VirtualAdversarialMethod FastFeatureAdversaries Noise Semantic _project_perturbation vatm SPSA MadryEtAl FastGradientMethod fgm optimize_linear arg_type ProjectedGradientDescent DeepFool SpatialTransformationMethod Attack MaxConfidence ElasticNetMethod BasicIterativeMethod MomentumIterativeMethod SaliencyMapMethod CleverHansTest list_files _list_files SimpleDataset random_feed_dict dev_version append_dev_version update_whitelist test_format_pep8 main DualFormulation NeuralNetParams Optimization read_weights diag initialize_dual DualFormulationTest NeuralNetParamsTest OptimizationTest ModelAllConvolutional _stride_arr _relu Softmax make_wresnet ResNet _decay Linear Flatten _batch_norm _conv _residual Conv2D Input _global_avg_pool Layer main show as_pil make_grid save get_logits_over_interval pair_visual linear_extrapolation_plot grid_visual save_pdf plot_report_from_path make_curve plot_report cifar10_tutorial main main evaluate_model ModelSubstitute train_sub prep_bbox setup_tutorial mnist_blackbox main main mnist_tutorial_cw main mnist_tutorial_jsma main mnist_tutorial main mnist_tutorial main PytorchMnistModel mnist_tutorial main mnist_tutorial main mnist_tutorial attack_selection make_basic_picklable_cnn ModelBasicCNN ModelBasicCNNTFE check_installation TestInception _top_1_accuracy TestSPSA InceptionModel load_images InceptionResnetV1Model load_testset main make_imagenet_cnn ModelImageNetCNN main main MadryMNIST MadryEtAlMultiGPU Evaluator create_adv_by_name make_basic_ngpu make_model make_basic_cnn make_madry_ngpu Conv2DnGPU unify_device_name Softmax MLP MaxPool LayerNorm Linear MLPnGPU clone_variable ReLU Conv2D LinearnGPU LayernGPU Flatten Layer ResNetTF Runner RunnerSingleGPU RunnerMultiGPU main run_trainer TestMadryEtAlMultiGPU TestRunnerMultiGPU TestRunMultiGPU TrainManager TrainerSingleGPU TrainerMultiGPU preprocess_batch read_CIFAR10 read_CIFAR100 cifar_tf_preprocess unpickle read_SVHN svhn_tf_preprocess get_image parse_args main download_image load_defense_output read_submissions_from_directory Submission AttacksOutput DatasetMetadata Attack Defense main parse_args compute_and_save_scores_and_ranking load_images main save_images InceptionModel load_images main save_images load_images main save_images load_images main load_images main load_images main inception_resnet_v2_arg_scope inception_resnet_v2 inception_resnet_v2_base block8 block35 block17 load_images save_images load_target_class main load_images save_images load_target_class main get_extract_command_template make_directory_writable SubmissionValidator load_defense_output shell_call main print_in_box main EvaluationMaster print_header save_dict_to_file make_directory_writable WorkerError AttackSubmission ExecutableSubmission sudo_remove_dirtree is_docker_still_running get_id_of_running_docker EvaluationWorker main DefenseSubmission kill_docker_container ClassificationBatches ResultMatrix read_classification_results analyze_one_classification_result CompetitionStorageClient CompetitionDatastoreClient NoTransactionBatch iterate_with_exp_backoff download_dataset DatasetMetadata enforce_epsilon_and_compute_hash ImageBatchesBase AversarialBatches DatasetBatches participant_from_submission_path CompetitionSubmissions is_unclaimed get_integer_time WorkPiecesBase DefenseWorkPieces AttackWorkPieces FakeDatasetMeta ClassificationResultsTest FakeStorageClient make_entity FakeDatastoreClientTransaction FakeBlob FakeDatastoreKey FakeDatastoreClient FakeDatastoreEntity FakeDatastoreClientBatch FakeDatastoreClientTest FakeDatastoreKeyTest FakeStorageClientTest FakeDatastoreClientTransactionTest FakeDatastoreEntityTest AdversarialBatchesTest ImageBatchesBaseTest DatasetBatchesTest ParticipantFromSubmissionPathTest SubmissionsTest WorkPiecesBaseTest AttackWorkPiecesTest DefenseWorkPiecesTest main ValidationStats SubmissionValidator get_extract_command_template make_directory_writable SubmissionValidator load_defense_output shell_call DQNModel parse_args make_env play dueling_model model parse_args maybe_load_model maybe_save_model make_env attack cleverhans_attack_wrapper RVBCleverhansModel py_func_grad main impl print_accuracies main main main main make_confidence_report_spsa current deprecated TestMNISTTutorialPytorch TestElasticNetMethod TestMadryEtAl TestSpatialTransformationMethod TestAttackClassInitArguments TestOptimizeLinear TestBasicIterativeMethod TestProjectedGradientDescent SimpleSpatialBrightPixelModel TestFastGradientMethod TestSaliencyMapMethod CommonAttackProperties TestDeepFool TestLBFGS TestCarliniWagnerL2 TestMomentumIterativeMethod TestParseParams DummyModel TestVirtualAdversarialMethod TestSPSA SimpleModel TrivialModel TestFastFeatureAdversaries TestAttackTF SimpleModel test_misclassify_request_examples test_unfinished_attack_configs test_make_confidence_report_bundled test_save_load_confidence_report test_confidence_report LightweightDataset TestDataset TestDefenses SimpleModel TestEvaluation TestMNISTBlackboxF TestMNISTTutorialCW TestMNISTTutorialJSMA TestMNISTTutorialKerasTF TestMNISTTutorialTF TestCallableModelWrapperInitArguments TestModelClass TestPerImageStandardize TestDropout test_rejects_callable test_no_logits TestSerial TestUtils TestKerasModelWrapper numpy_kl_with_logits TestUtilsTF CIFAR10_model FMNIST_model IMAGENET_model MNIST_model SVHN_model argmin float32 copy placeholder get_adv generate zeros FastGradientMethod BasicIterativeMethod MomentumIterativeMethod range len data model get_value set_random_seed sign is_train stop_gradient KerasModelWrapper clip subtitute_model placeholder generate predict sub_is_train GetGradient get_model eps get_adv_examples model_eval AccuracyReport print float32 train RDA attack check_installation zeros load_images list print shuffle print to_categorical astype load_data str print labels images shape vstack read_data_sets print labels images vstack read_data_sets join print transpose to_categorical astype ravel loadmat to_float dtype stop_gradient softmax_cross_entropy_with_logits gradients assert_greater_equal reduce_max reduce_sum cast assert_less_equal append equal CIFAR10_model MNIST_model FMNIST_model print IMAGENET_model SVHN_model shape ndarray shape ndarray ndarray cos shuffle pi sin array range range copy len pop list arange get_metrix_index reshape diag shuffle copy sign range shape cal_dot append argmax array clip predict len Sequential add Dense AveragePooling2D Convolution2D Activation range Flatten Sequential add Dense MaxPooling2D Conv2D Activation Flatten Model Input resnet MobileNetV2 data print Adam fit_generator load_weights ReduceLROnPlateau flow ImageDataGenerator ModelCheckpoint compile fit list print Adam shuffle fit_generator load_weights ReduceLROnPlateau flow_from_directory ImageDataGenerator ModelCheckpoint compile max_angle nb_dimensions warn inputs warn minimum maximum argmax int list fill_diagonal discard reshape set abs max range len update reshape other_classes run zeros sum enumerate append range gradients int product reshape debug float jacobian copy model_argmax shape saliency_map apply_perturbations floor info bool len int value fill_diagonal constant ones reshape while_loop float32 warn cast int32 floor bool update get_shape list min vstack zip range deepfool_attack sum norm clip inf debug squeeze model_argmax copy flatten shape info zeros abs array range run warn to_float one_hot reduce_max reduce_sum warn resize_images convert_to_tensor min astype rotate translate sqrt pad int32 float to_float list product reduce_max parallel_apply_transformations reduce_sum choice map_fn stack gather_nd linspace zip stop_gradient get_probs argmax equal einsum convert_to_tensor reshape map_fn _apply_black_border tile to_float str one_hot ones bundle_attacks ProjectedGradientDescent copy AttackConfig int32 Noise range append to_float str one_hot ones bundle_attacks ProjectedGradientDescent copy AttackConfig int32 Noise range append to_float str one_hot ones bundle_attacks ProjectedGradientDescent copy AttackConfig int32 Noise range append Noise bundle_attacks AttackConfig str correctness_and_confidence copy ConfidenceReport mean ConfidenceReportEntry info zeros bundle_attacks_with_goal str get_criteria start mean run_batch_with_goal save info request_examples time hasattr get_criteria params print_progress attack new_wins ConfidenceReportEntry save get_attack_config run_attack enumerate print print_stats str min mean info append str zeros_like ConfidenceReport mean new_wins ConfidenceReportEntry info save range len to_float str one_hot ones bundle_attacks copy AttackConfig int32 append SPSA range pad Graph time vstring warn op_func stop_gradient softmax_cross_entropy_with_logits_v2 hasattr get_set run_recipe set_log_level getattr factory max_val callable dataset_factory Session INFO mean sum print maximum Semantic time correctness_and_confidence get_set run_attack print set_log_level set_random_seed ConfidenceReport MaxConfidence ConfidenceReportEntry save print_stats dataset_factory factory Session INFO join urlretrieve gettempdir isfile maybe_download_file join open expand_dims to_categorical download_and_parse_mnist_file reshape _check_x _CorrectFactory batch_eval_multi_worker _check_y _check_x min batch_eval_multi_worker _CorrectAndProbFactory max _check_y _AttackFactory _check_x batch_eval_multi_worker _check_y infer_devices int str update debug extend ceil run_canary run zip append float range enumerate len update str list debug warn dict run zip append range len split warn warn dump batch_size randint warn get_next avg_grads run infer_devices str list ema_decay placeholder append ceil range update RandomState shuffle initialize_uninitialized_global_variables info float nb_epochs _ArgsWrapper int time evaluate run_canary AdamOptimizer int32 ExponentialMovingAverage global_variables_initializer callable len append add_n zip len int list range remove zeros ravel warn sum to_categorical astype choice other_classes shape int32 zeros argmax range warn warn warn warn setLevel setFormatter getLogger addHandler StreamHandler Formatter OrderedDict sorted keys append len join list debug group match range compile len copy Sequential model Activation add join warn warn MNIST warn get_default_graph getrandbits out_features softmax_cross_entropy_with_logits inputs op warn reduce_mean variables_initializer len global_variables run fprop argmax equal _ArgsWrapper update run minimum get_shape list maximum reduce_sum square sqrt div clip_by_value abs range len RandomState minimize warn model_loss AdamOptimizer _ArgsWrapper get_available_gpus warn list_local_devices cast_clip is_scalar dtype cast gradient assign Saver save apply_gradients filename eager train_dir batch_indices zip get_params join Variable int str batch_size Variable debug min get_probs range assign generate ceil zeros float eager len Variable get_probs eager to_float dtype stop_gradient softmax_cross_entropy_with_logits optimize_linear gradients assert_greater_equal reduce_max reduce_sum cast assert_less_equal clip_by_value append equal append tuple dtype as_np_dtype to_float get_shape list mul reduce_max equal maximum reduce_sum square sign sqrt stop_gradient abs range len dtype init_state nest while_loop flatten shape project_perturbation cast assert_less_equal random_uniform no_op join isdir _list_files abspath pardir endswith listdir append isdir astype md5 update sorted list_files dev_version extend join relpath append shell_call pardir adv_class input_minval num_classes model_json input_maxval read_weights range sizes true_class NeuralNetParams get_full_psd_matrix set_differentiable_objective init_dual_file checkpoint print reshape DualFormulation initialize_dual epsilon list get_tensor load_checkpoint transpose reshape append keys get_variable_to_shape_map str reshape num_hidden_layers astype float32 item append range get_variable debug get_shape append trainable_variables l2_loss ResNet checkpoint_dir latest_checkpoint exit set_log_level DEBUG Session mkstemp close shell_call save fromarray dtype min issubdtype max floating int concatenate shape sqrt ceil zeros float split show squeeze pause add_subplot axis set_window_title imshow figure ion enumerate ioff show add_subplot axis set_window_title imshow figure range get_logits l2_normalize reshape float lin_space placeholder expand_dims generate FastGradientMethod show use plot xlabel ylabel set_window_title clf savefig figure linspace legend xlim argmax range gcf PdfPages close savefig load plot_report plot concatenate make_curve xlabel print ylabel get_color array str sorted warn append float len do_eval set_random_seed CrossEntropy stop_gradient DEBUG Session map placeholder set_log_level attack prefetch generate get_logits RandomState CIFAR10 FastGradientMethod get_set batch print AccuracyReport float32 dict train ModelAllConvolutional cifar10_tutorial MNIST do_eval get_logits get_set float32 placeholder set_log_level dict set_random_seed generate FastGradientMethod Session INFO evaluate_model set_random_seed str get_logits print model_eval CrossEntropy ModelBasicCNN train ModelSubstitute str int get_logits print hstack jacobian_augmentation CrossEntropy argmax range jacobian_graph MNIST str Session get_logits RandomState train_sub prep_bbox print model_eval float32 placeholder set_log_level generate DEBUG FastGradientMethod argmax get_set mnist_blackbox CarliniWagnerL2 grid_visual set_random_seed CrossEntropy Saver save DEBUG exists Session str placeholder set_log_level tf_model_load shape sum range get_logits RandomState format close mean ModelBasicCNN get_set MNIST generate_np print AccuracyReport model_eval reshape float32 zeros train array mnist_tutorial_cw grid_visual set_random_seed CrossEntropy DEBUG argmax Session run str placeholder set_log_level sum range pair_visual get_logits RandomState format close mean other_classes ModelBasicCNN float get_set zeros MNIST int generate_np print AccuracyReport model_eval reshape float32 global_variables_initializer train SaliencyMapMethod mnist_tutorial_jsma model set_image_dim_ordering set_random_seed CrossEntropy Saver stop_gradient KerasModelWrapper Session restore set_learning_phase set_session placeholder attack generate format RandomState get_checkpoint_state mkdir FastGradientMethod get_set MNIST evaluate print AccuracyReport model_eval float32 model_2 cnn_model train mnist_tutorial do_eval DEBUG make_basic_picklable_cnn set_log_level get_factory get_logits dict CallableModelWrapper zero_grad DataLoader numpy tf_model_fn cuda run convert_pytorch_model_to_tf Adam append range nll_loss item is_available float torch_model backward parameters PytorchMnistModel step ModelBasicCNN list keys clip ModelBasicCNNTFE attack_class attack_selection MLP warn join int index range asarray read_pairs min choice load_data append get_paths max range len make_imagenet_cnn set_random_seed random_uniform generate FastFeatureAdversaries abs ys to_categorical CIFAR10Data dataset_dir xs float32 placeholder MadryMNIST Saver data_mnist FastGradientMethod BasicIterativeMethod get_probs MLP make_basic_cnn layers MLPnGPU MLPnGPU basicConfig model_train eval TrainerSingleGPU TrainerMultiGPU info list namedtuple HParams keys run_trainer load close open join unpickle concatenate reshape transpose mean append range len join concatenate reshape transpose mean append unpickle random_crop per_image_standardization random_saturation random_flip_left_right resize_image_with_crop_or_pad random_brightness random_contrast join concatenate transpose flatten mean append loadmat range len print resize_image_with_crop_or_pad random_crop add_argument ArgumentParser print str join read convert urlopen save resize crop exists join format partial write index set add ThreadPool enumerate imap_unordered close parse_args flush len set_defaults join Attack Defense append listdir items list join targeted_attack_names image_by_base_filename len dataset_image_count write_ranking write_score_matrix get_target_class zeros sum keys attack_names get_true_label dataset_metadata save_all_classification load_defense_output intermediate_results_dir name DatasetMetadata AttacksOutput compute_and_save_scores_and_ranking mkdir output_dir save_target_classes clip_and_copy_attack_outputs run basename Glob append enumerate max_epsilon set_verbosity INFO load_images save_images input_dir warn iter_alpha num_iter load_target_class endswith iteritems error shell_call print len seed use_gpu SubmissionValidator mkdtemp temp_dir submission_filename call submission_type validate_submission info print_in_box print len prepare_attacks compute_results limited_dataset EvaluationMaster show_status cleanup_datastore round_name cleanup_defenses blacklisted_submissions warning prepare_defenses cleanup_attacks_with_zero_images cleanup_failed_attacks check_output shell_call get_id_of_running_docker run_work EvaluationWorker shell_call decode int BytesIO seek reader size get_blob warning download_to_file StringIO iteritems read_classification_results hasattr iter join convert astype hexdigest BICUBIC warning save resize array clip data join iteritems basename bucket_name copyfile call mkdir exists endswith isdigit basename startswith time FakeDatastoreKey isinstance boolean_flag SimpleMonitor make wrap_dqn str print close capture_frame render reset VideoRecorder float step len join time format put relatively_safe_pickle_dump save_state log get join pickle_load format log load_state exists FastGradientMethod str get_default_graph randint get_set set_log_level set_random_seed impl dataset_factory factory Session INFO Semantic time print accuracy ProjectedGradientDescent print_accuracies make_confidence_report make_confidence_report_bundled train_start get_set bundle_examples_with_goal report_path warn which_set MaxConfidence factory max_val dataset_factory test_end test_start train_end get_set float32 set_log_level set_random_seed NB_CLASSES dataset_factory factory spsa_max_confidence_recipe Session INFO make_confidence_report_spsa hasattr print warn mean correctness completed mean print warn append array AttackConfig unfinished_attack_configs request_examples list Misclassify AttackConfig array ConfidenceReportEntry ConfidenceReport SimpleDataset MLP make_confidence_report_bundled get_factory Session Linear load copy ConfidenceReport ConfidenceReportEntry save zeros assert_raises ones ProjectedGradientDescent NoLogitsModel generate Session assert_raises Session exp sum numpy_softmax log

# Random Directional Attack for Fooling Deep Neural Networks This project is for the paper "Random Directional Attack for Fooling Deep Neural Networks". Our implementation is based on [cleverhans](https://github.com/tensorflow/cleverhans/tree/v.3.0.1) . The code was developed on Python 3.6 ## 1. Install dependencies. Our experiment runs on GPU,, install this list: ```bash pip install -r requirements_gpu.txt ```

DagnyT/hardnet

['image retrieval', 'patch matching']

["Working hard to know your neighbor's margins: Local descriptor learning loss"]

code/Losses.py code/download_all_datasets.py examples/extract_hardnet_desc_from_hpatches_file.py code/Loggers.py benchmarks/hpatches_extract_HardNet.py code/dataloaders/HPatchesDatasetCreator.py code/check_gor_triplet.py code/HardNet.py code/HardNetClassicalHardNegMiningSiftInit.py code/HardNetMultipleDatasets.py code/W1BS.py code/HardNetHPatchesSplits.py code/EvalMetrics.py code/dataloaders/TotalDataLoader.py examples/caffe/extract_hardnetCaffe_desc_from_hpatches_file.py code/HardNetClassicalHardNegMining.py code/Utils.py code/dataset.py examples/extract_DenseHardNet.py examples/caffe/convert_weights_to_caffe.py code/check_gor_HardNet.py L2Norm L1Norm HardNet hpatches_sequence cd usage cd usage np2torch HPatchesDM read_patch_file read_image_dir TripletPhotoTour find_files TotalDatasetsLoader ErrorRateAt95Recall create_optimizer create_loaders test adjust_learning_rate CorrelationPenaltyLoss weights_init TripletPhotoTour main train HardNet BuildKNNGraphByFAISS_GPU get_hard_negatives create_optimizer PhototourTrainingData get_descriptors_for_dataset showImagesHorizontally create_loaders TripletPhotoTourHardNegatives test pre_init_with_sift remove_descriptors_with_same_index adjust_learning_rate TripletPhotoTour weights_init main train CorrelationPenaltyLoss TNet create_optimizer create_loaders test adjust_learning_rate CorrelationPenaltyLoss weights_init TripletPhotoTour main TotalDatasetsLoader train HardNet create_optimizer create_loaders test adjust_learning_rate CorrelationPenaltyLoss weights_init TripletPhotoTour main TotalDatasetsLoader train HardNet FileLogger Logger loss_random_sampling global_orthogonal_regularization loss_HardNet distance_matrix_vector loss_L2Net distance_vectors_pairwise L1Norm str2bool L2Norm w1bs_extract_descs_and_save mean_image std_image HPatches TotalDatasetsLoader LocalNorm2d load_grayscale_var DenseHardNet L2Norm L1Norm HardNet L2Norm L1Norm copy_weights HardNet L2Norm extract_tfeats preprocess_patch print exit endswith join listdir append from_numpy rollaxis uint8 size convert astype mean append range len append print read_patch_file find_files cat sum argmax data constant isinstance orthogonal Conv2d remove training_set Compose copy DataLoader TripletPhotoTour model zero_grad loss_HardNet model_dir set_description adjust_learning_rate save dataset cuda loss_L2Net loss_random_sampling gor format decor stat item enumerate enable_logging backward Variable tqdm step len format enable_logging model print reshape size len tqdm sqrt eval set_description ErrorRateAt95Recall log_value append dataset sum cuda enumerate batch_size param_groups n_triplets lr float epochs parameters Adam SGD draw_and_save_plots w1bs_extract_descs_and_save features cuda load_state_dict draw_and_save_plots_with_loggers range format create_optimizer replace create_loaders test get_list_of_patch_images match_descriptors_and_save_results start_epoch lr resume vars load enable_logging print isfile train epochs add_subplot axis imshow figure range len Linear numpy get_hard_negatives pre_init_with_sift DataLoader save TripletPhotoTourHardNegatives Compose log_string eval get_descriptors_for_dataset GpuIndexFlatConfig search GpuIndexFlatL2 add shape StandardGpuResources data DataLoader resize cuda FloatTensor append expand_dims astype eval SIFT type enumerate PhototourTrainingData Variable SIFTNet float32 tqdm extend numpy data transform PhototourTrainingData model print extend tqdm eval DataLoader append numpy cuda enumerate len append all range len BuildKNNGraphByFAISS_GPU items list print labels remove_descriptors_with_same_index hardnegatives create_indices TotalDatasetsLoader unsqueeze sqrt sum exp print clamp min exit mean distance_vectors_pairwise exp print type_as exit mean distance_matrix_vector ge sum cuda diag exp view print clamp squeeze type_as exit min gather t mean distance_matrix_vector float cuda diag mul clamp size mean pow sum model resize cuda from_numpy shape savetxt dirname append imread range replace concatenate astype eval enumerate int time Variable reshape print float32 tqdm zeros makedirs mean std view Variable size convert astype float32 mean from_numpy array str isinstance print Conv2d BatchNorm2d numpy astype float32 int reshape ceil zeros float range len

# HardNet model implementation HardNet model implementation in PyTorch for NIPS 2017 paper ["Working hard to know your neighbor's margins: Local descriptor learning loss"](https://arxiv.org/abs/1705.10872) [poster](http://cmp.felk.cvut.cz/~mishkdmy/posters/hardnet2017.pdf), [slides](http://cmp.felk.cvut.cz/~mishkdmy/slides/HardNet2017.pdf) ## An example how to compile HardNet to Torchscript to be used in C++ code [Notebook](notebook/convert_HardNet_to_JIT.ipynb) ## Update April 06 2018 We have added small shift and rot augmentation, which improves results up to 1mAP point on HPatches. It is in HardNet.py, turn on by --augmentation=True. All the weight will be updated soon. Version, which is trained on Brown + HPatches + PS datasets is in progress, stay tuned :) ## Re: popular question about BoW retrieval engine Unfortunately, it is proprietary and we cannot release it. But you can try the following open source repos, both Matlab-based: - [ASMK](https://github.com/gtolias/asmk)

Dahee96/Seq2seq-

['speech recognition', 'noisy speech recognition', 'distant speech recognition']

['The PyTorch-Kaldi Speech Recognition Toolkit']

kaldi_decoding_scripts/utils/nnet/gen_hamm_mat.py kaldi_decoding_scripts/utils/reverse_arpa.py kaldi_decoding_scripts/utils/nnet/gen_splice.py kaldi_decoding_scripts/utils/nnet/gen_dct_mat.py kaldi_decoding_scripts/utils/filt.py kaldi_decoding_scripts/utils/nnet/make_nnet_proto.py kaldi_decoding_scripts/utils/nnet/make_cnn_proto.py tune_hyperparameters.py kaldi_decoding_scripts/utils/nnet/make_lstm_proto.py run_exp.py kaldi_decoding_scripts/utils/nnet/make_blstm_proto.py plot_acc_and_loss.py save_raw_fea.py data_io.py utils.py neural_networks.py core.py kaldi_decoding_scripts/utils/nnet/make_cnn2d_proto.py extract_data_from_shared_list convert_numpy_to_torch run_nn run_nn_refac01 read_next_chunk_into_shared_list_with_subprocess load_counts UnknownMatrixHeader _read_vec_flt_binary open_or_fd _read_mat_ascii read_vec_int_ark context_window_old read_vec_flt_scp UnknownVectorHeader read_cntime_ark load_chunk read_vec_flt_ark write_mat read_cntime UnsupportedDataType write_vec_int BadInputFormat read_post_ark SubprocessFailed write_vec_flt read_vec_int read_mat load_dataset BadSampleSize read_vec_flt read_post_rxspec read_post_scp read_ali_ark read_lab_fea_refac01 _read_vec_flt_riff _read_mat_binary read_key read_cnet_ark _read_compressed_mat read_segments_as_bool_vec read_lab_fea context_window popen read_mat_scp read_post read_mat_ark CNN RNN PositionwiseFeedForward liGRU SincNet LayerNorm TransformerLayer LSTM_cudnn GRU_cudnn MultiSequential channel_averaging _pre_hook get_attn_pad_mask get_non_pad_mask GRU SincConv_fast minimalGRU act_fun TLayerNorm MLP PASE Conv2dSubsampling GELU MultiHeadedAttention flip Transformer1 repeat SincConv SRU LSTM RNN_cudnn PositionalEncoding logMelFb _max_nr_of_parallel_forwarding_processes _is_first_validation _run_forwarding_in_subprocesses nth_replace_string run_command compute_cw_max create_curves run_shell read_args_command_line check_cfg parse_model_field create_lists get_chunks_after_which_to_validate optimizer_init create_block_diagram run_shell_display dump_epoch_results is_sequential export_loss_acc_to_txt split_chunks write_cfg_chunk expand_section_proto get_val_cfg_file_path expand_str_ep create_block_connection get_val_lst_file_path compute_avg_performance check_consistency_with_proto parse_fea_field change_lr_cfg check_field expand_section get_all_archs get_val_info_file_path forward_model terminal_node_detection model_init create_configs parse_lab_field dict_fea_lab_arch is_sequential_dict do_validation_after_chunk _get_val_file_name_base shift check_cfg_fields cfg_item2sec list_fea_lab_arch forward_model_refac01 compute_n_chunks progress fix_filt_step Glorot start Thread join float cuda view _optimization_step load_counts strtobool convert_numpy_to_torch write_mat _prepare_input log _write_info_file list _get_dim_from_data_set len map _read_chunk_specific_config sum range detach extract_data_from_shared_list close _get_batch_size_from_config _get_batch_config read_next_chunk_into_shared_list_with_subprocess forward_model join time _save_model is_sequential_dict _initialize_random_seed shift _update_progress_bar _open_forward_output_files_and_get_file_handles _load_model_and_optimizer numpy split load_counts strtobool open_or_fd zero_grad DataParallel numpy save write_mat round cuda max log seed str list optimizer_init len exit map load_state_dict sum range detach state_dict Thread replace ConfigParser close start manual_seed item float keys model_init forward_model load int read join time backward print is_sequential_dict contiguous shift write randint step progress split _read_features_and_labels_with_kaldi _match_feature_and_label_sequence_lengths _chunk_features_and_labels _concatenate_features_and_labels _input_is_wav_file flatten empty range concatenate empty range roll min mean context_window load_dataset std column_stack _reorder_data_set _append_to_shared_list _read_features_and_labels _read_from_config _read_chunk_specific_config update int read list dict_fea_lab_arch ConfigParser is_sequential_dict write exit shuffle load_chunk compute_cw_max keys append column_stack rsplit int seek search popen split open start Popen open decode strip read_vec_int open_or_fd read_key decode remove read open_or_fd close frombuffer array split pack char write open_or_fd encode range len read_vec_flt open_or_fd split read_vec_flt open_or_fd read_key decode remove open_or_fd close array split read frombuffer unpack decode read frombuffer pack char write open_or_fd encode tobytes read_mat open_or_fd split read_mat open_or_fd read_key decode _read_mat_ascii _read_mat_binary open_or_fd decode read reshape startswith frombuffer decode vstack append array split dtype read reshape zeros frombuffer array pack char write open_or_fd encode tobytes print exit startswith open_or_fd read_post split open_or_fd read_post read_key decode read tolist open_or_fd close append frombuffer range read_cntime open_or_fd read_key decode read tolist open_or_fd close frombuffer loadtxt repeat astype float size range new_zeros lt get_non_pad_mask expand pop size view contiguous strtobool get_chunks_after_which_to_validate _get_nr_of_valid_per_epoch_from_config decode readline print append Popen decode write Popen flush communicate wait Popen int str findall group write exit nth_replace_string split append range compile len read ConfigParser mean append float sum int list write exit map float split append sections read list add_section ConfigParser remove_section set sections append keys range values len ConfigParser read list set list write exit any sections keys read ConfigParser exit check_cfg_fields expand_section open rstrip strtobool values open run_shell str list sorted parse_model_field len exit create_block_diagram append sum range replace check_consistency_with_proto parse_fea_field sections join items parse_lab_field int read write split findall makedirs write sections exit append range len list _partition_chunks append get_chunks_after_which_to_validate _get_nr_of_valid_per_epoch_from_config format _get_val_lst_file_name _get_val_info_file_name _get_val_cfg_file_name strtobool max open str check_cfg list exit log10 ceil append range write_cfg_chunk expand_str_ep get_val_cfg_file_path format get_val_lst_file_path replace close get_all_archs float get_val_info_file_path keys int items do_validation_after_chunk write split compute_n_chunks len __add__ max open seed str sorted list map log10 reverse writelines ceil append split_chunks range format get_val_lst_file_path parse_fea_field close shuffle _get_validation_data_for_chunks _shuffle_forward_data int do_validation_after_chunk cfg_item2sec split len add_section str list sorted remove_section append range replace check_consistency_with_proto ConfigParser glob remove_option sections keys int read join items cfg_item2sec findall len sorted write exit sub append split write exit sub append split glob int sorted format read list replace ConfigParser len write exit findall float range append open list str list index append range len run_shell str read list remove replace create_block_connection ConfigParser findall append list replace strtobool len map cfg_item2sec findall range append split list replace strtobool len map cfg_item2sec findall range append split strtobool list keys strtobool int list max append keys list out_dim strtobool nn_class print set eval import_module getattr train cuda CrossEntropyLoss list strtobool map SGD Adam RMSprop parameters float keys split list _get_network_output _get_labels_from_input mean _add_input_features_to_outs_dict shape _compute_layer_values float cat len list view float mean shape long bool keys cat len str list int print write close keys log10 ceil max open int write float round flush str asarray ndarray readlines makedirs savetxt split append float range len arange axis str use exit ylabel title savefig legend append export_loss_acc_to_txt range plot readlines clear print loadtxt xlabel write amax len find str read list ConfigParser set keys int write extend exit append float split range with_glorot

<<<<<<< HEAD # The PyTorch-Kaldi Speech Recognition Toolkit <img src="pytorch-kaldi_logo.png" width="220" img align="left"> PyTorch-Kaldi is an open-source repository for developing state-of-the-art DNN/HMM speech recognition systems. The DNN part is managed by PyTorch, while feature extraction, label computation, and decoding are performed with the Kaldi toolkit. This repository contains the last version of the PyTorch-Kaldi toolkit (PyTorch-Kaldi-v1.0). To take a look into the previous version (PyTorch-Kaldi-v0.1), [click here](https://bitbucket.org/mravanelli/pytorch-kaldi-v0.0/src/master/). If you use this code or part of it, please cite the following paper: *M. Ravanelli, T. Parcollet, Y. Bengio, "The PyTorch-Kaldi Speech Recognition Toolkit", [arXiv](https://arxiv.org/abs/1811.07453)* ``` @inproceedings{pytorch-kaldi, title = {The PyTorch-Kaldi Speech Recognition Toolkit},

Daikenan/ASRCF

['visual tracking']

['Visual Tracking via Adaptive Spatially-Regularized Correlation Filters']

external_libs/matconvnet/utils/proto/caffe_fastrcnn_pb2.py external_libs/matconvnet/doc/matdocparser.py external_libs/matconvnet/doc/matdoc.py external_libs/matconvnet/utils/proto/caffe_6e3916_pb2.py external_libs/matconvnet/utils/proto/caffe_pb2.py external_libs/matconvnet/utils/proto/caffe_b590f1d_pb2.py external_libs/matconvnet/utils/proto/caffe_0115_pb2.py external_libs/matconvnet/utils/layers.py external_libs/matconvnet/utils/proto/caffe_old_pb2.py external_libs/matconvnet/utils/proto/vgg_caffe_pb2.py external_libs/matconvnet/utils/import-caffe.py extract render_L render_V render_P render_DIVL render_DH render_BL render_L_from_indent render_SL render_S render Context render_DI Frame render_B findNextFunction getFunctionDoc readText MatlabFunction render_DL clean Lexer P PL Parser BH EOF DI L DL SL BL DH DIVL Terminal S DIV NonTerminal B V Symbol blobproto_to_array versiontuple dict_to_struct_array keyboard tolist escape bilinear_interpolate getopts find rowcell CaffeInnerProduct ConversionError CaffeScale CaffeBatchNorm CaffeLayer CaffeCrop CaffeConcat CaffeConv CaffePooling CaffeData reorder CaffeElementWise getFilterOutputSize CaffeROIPooling CaffeSoftMaxLoss CaffeReLU getFilterTransform CaffeModel CaffeTransform CaffeDeconvolution row dictToMatlabStruct rowarray CaffeBlob transposeTransform CaffeDropout CaffeLRN CaffeEltWise composeTransforms CaffeSoftMax HingeLossParameter BlobProto BlobProtoVector NetStateRule LayerParameter PowerParameter FillerParameter ArgMaxParameter V0LayerParameter InnerProductParameter ConvolutionParameter SolverState EltwiseParameter SliceParameter WindowDataParameter DummyDataParameter HDF5OutputParameter TanHParameter TransformationParameter SoftmaxParameter ConcatParameter DataParameter SolverParameter MVNParameter ContrastiveLossParameter NetState NetParameter PoolingParameter DropoutParameter Datum SigmoidParameter AccuracyParameter MemoryDataParameter LRNParameter ReLUParameter ImageDataParameter InfogainLossParameter HDF5DataParameter ThresholdParameter ReductionParameter HingeLossParameter BlobProto BlobProtoVector NetStateRule LayerParameter PowerParameter FillerParameter ArgMaxParameter V0LayerParameter InnerProductParameter ConvolutionParameter SolverState EltwiseParameter LossParameter SliceParameter WindowDataParameter DummyDataParameter HDF5OutputParameter TanHParameter TransformationParameter SoftmaxParameter ConcatParameter DataParameter SPPParameter ParamSpec EmbedParameter SolverParameter MVNParameter ContrastiveLossParameter NetState NetParameter PoolingParameter DropoutParameter Datum SigmoidParameter BlobShape ExpParameter AccuracyParameter LogParameter ThresholdParameter TileParameter MemoryDataParameter LRNParameter ReLUParameter ImageDataParameter ReshapeParameter InfogainLossParameter V1LayerParameter HDF5DataParameter PReLUParameter FlattenParameter PythonParameter ReductionParameter HingeLossParameter BlobProto BlobProtoVector NetStateRule LayerParameter PowerParameter FillerParameter ArgMaxParameter V0LayerParameter InnerProductParameter ConvolutionParameter SolverState EltwiseParameter LossParameter SliceParameter BatchNormParameter WindowDataParameter DummyDataParameter HDF5OutputParameter TanHParameter TransformationParameter SoftmaxParameter ConcatParameter DataParameter SPPParameter ParamSpec EmbedParameter SolverParameter MVNParameter ContrastiveLossParameter NetState NetParameter BiasParameter PoolingParameter DropoutParameter Datum SigmoidParameter BlobShape ExpParameter AccuracyParameter LogParameter ThresholdParameter TileParameter MemoryDataParameter LRNParameter ReLUParameter ImageDataParameter ELUParameter ReshapeParameter InfogainLossParameter ScaleParameter V1LayerParameter HDF5DataParameter PReLUParameter FlattenParameter PythonParameter ROIPoolingParameter HingeLossParameter BlobProto BlobProtoVector NetStateRule LayerParameter PowerParameter FillerParameter ArgMaxParameter V0LayerParameter InnerProductParameter ConvolutionParameter SolverState EltwiseParameter LossParameter SliceParameter WindowDataParameter DummyDataParameter HDF5OutputParameter TanHParameter TransformationParameter SoftmaxParameter ConcatParameter DataParameter ParamSpec SolverParameter MVNParameter ContrastiveLossParameter NetState NetParameter PoolingParameter DropoutParameter Datum SigmoidParameter BlobShape ExpParameter AccuracyParameter ThresholdParameter MemoryDataParameter LRNParameter ReLUParameter ImageDataParameter InfogainLossParameter V1LayerParameter HDF5DataParameter PReLUParameter PythonParameter NetParameter LayerConnection BlobProto BlobProtoVector LayerParameter FillerParameter Datum SolverParameter SolverState BlobProto BlobProtoVector PowerParameter LayerParameter FillerParameter V0LayerParameter InnerProductParameter ConvolutionParameter SolverState WindowDataParameter HDF5OutputParameter ConcatParameter DataParameter SolverParameter NetParameter PoolingParameter DropoutParameter Datum MemoryDataParameter LRNParameter ImageDataParameter InfogainLossParameter HDF5DataParameter NetParameter EvalHistoryIter LayerConnection BlobProto BlobProtoVector LayerParameter FillerParameter Datum EvalHistory SolverParameter SolverState search clean match strip group append getFunctionDoc findNextFunction print print print children render_SL print pop render_DH print Frame push render_DIVL children render_DI print children render_L print pop children render_L_from_indent isa Frame render_B push indent pop children Frame push render_DIVL children render_BL render_V render_S render_DL isa render_P print Context render_DIVL dim tolist hasattr list empty keys RepeatedScalarFieldContainer isinstance update print f_locals interact copy reshape asarray astype clip hasattr list ndarray isinstance append empty keys CaffeTransform CaffeTransform

# ASRCF - Visual Tracking via Adaptive Spatially-Regularized Correlation Filters(**CVPR2019 Oral**). <div align="center"> <img src="https://github.com/Daikenan/ASRCF/blob/master/faceocc1.gif" width="500px" /> </div> ## abstract 　　In this work, we propose a novel adaptive spatially-regularized correlation filters (ASRCF) model to simultaneously optimize the filter coefficients and the spatial regularization weight. First, this adaptive spatial regularization scheme could learn an effective spatial weight for a specific object and its appearance variations, and therefore result in more reliable filter coefficients during the tracking process. Second, our ASRCF model can be effectively optimized based on the alternating direction method of multipliers, where each subproblem has the closed-from solution. Third, our tracker applies two kinds of CF models to estimate the location and scale respectively. The location CF model exploits ensembles of shallow and deep features to determine the optimal position accurately. The scale CF model works on multi-scale shallow features to estimate the optimal scale efficiently. Extensive experiments on five recent benchmarks show that our tracker performs favorably against many state-of-the-art algorithms, with real-time performance of 28fps. ## Paper link - [Google Drive](https://drive.google.com/file/d/1zsUnEmXTLwXqTKytpv3dWTqEreK90_bI/view?usp=sharing) ## Citation

DanailKoychev/neural-style

['style transfer']

['A Neural Algorithm of Artistic Style']

utils.py vgg16_nofc.py style_transfer.py loss_functions.py frobenious_norm gram_matrix style_transfer_loss content_loss style_loss parse_arguments load_image save_image trim_colors Vgg16 as_list reshape matmul content_loss style_loss subtract conv4_2 as_list frobenious_norm pack frobenious_norm conv3_3 conv2_2 subtract conv1_2 conv4_3 gram_matrix add_argument ArgumentParser minimum shape maximum copy save trim_colors int imread min resize

Neural Style =================== A simple implementaton of the style transfer algorithm described here: https://arxiv.org/abs/1508.06576 --- <img src="https://raw.githubusercontent.com/DanailKoychev/neural-style/master/sample-images/jf1.png" height="300px"> --- <p float="left"> <img src="https://raw.githubusercontent.com/DanailKoychev/neural-style/master/sample-images/original-images/natalie_dormer.jpeg" height="210px"> <img src="https://raw.githubusercontent.com/DanailKoychev/neural-style/master/sample-images/original-images/fmi.jpg" height="210px"> <img src="https://raw.githubusercontent.com/DanailKoychev/neural-style/master/sample-images/original-images/chicago.jpg" height="210px">

Danial-Alh/fast-bleu

['text generation']

['Jointly Measuring Diversity and Quality in Text Generation Models']

old_metrics/bleu-old.py old_metrics/self_bleu.py old_metrics/bleu.py fast_bleu/__python_wrapper__.py fast_bleu/__init__.py fast_bleu/__test__.py setup.py test_cases.py test_cases/test_case2.py old_metrics/utils.py InstallCommand CleanCommand BuildExtWithoutPlatformSuffix nltk_bleu cpp_bleu nltk_self_bleu cpp_self_bleu nltk_org_bleu compare get_execution_time SelfBLEU _encode_listoflist_str BLEU _load_cdll test closest_ref_length corpus_bleu modified_precision Bleu brevity_penalty closest_ref_length corpus_bleu modified_precision Bleu brevity_penalty SelfBleu get_ngrams Ngram tokenize Threader user_options Bleu BLEU get_score SelfBleu get_score SelfBLEU time array func str format print fromarrays get_execution_time float sum diff dirname print get_score BLEU SelfBLEU exp Counter modified_precision fsum method0 smoothing_function enumerate brevity_penalty len sum max values min isinstance Ngram run

# fast-bleu Package This is a fast multithreaded C++ implementation of NLTK BLEU with Python wrapper; computing BLEU and SelfBLEU scores for a fixed reference set. It can return (Self)BLEU for different (max) n-grams simultaneously and efficiently (e.g. BLEU-2, BLEU-3, etc.). ## Installation The installation requires `c++11`. The `requirements.txt` file is the required python packages to run the `test_cases.py` file. ### Linux and WSL Installing [PyPI latest stable release](https://pypi.org/project/fast-bleu/): ``` bash pip install --user fast-bleu

DanielTakeshi/gym-cloth

['imitation learning']

['Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisor']

render/ext/nanogui/ext/pybind11/tests/test_smart_ptr.py gym_cloth/envs/__init__.py render/ext/nanogui/ext/pybind11/tools/clang/enumerations.py render/ext/nanogui/python/example1.py render/ext/nanogui/docs/exhale.py render/ext/nanogui/ext/pybind11/tests/test_issues.py render/ext/nanogui/ext/pybind11/tests/test_numpy_array.py render/ext/nanogui/ext/pybind11/tests/test_alias_initialization.py analysis/convert_hdf5_visual_dynamics.py render/ext/nanogui/ext/pybind11/tests/test_stl_binders.py gym_cloth/physics/util.py render/ext/nanogui/ext/pybind11/tests/test_methods_and_attributes.py render/ext/nanogui/python/example3.py gym_cloth/blender/test.py examples/demo_bed.py examples/demo_render.py render/ext/nanogui/ext/pybind11/pybind11/__init__.py render/ext/nanogui/ext/pybind11/tests/test_inheritance.py render/ext/nanogui/ext/pybind11/tools/clang/cindex.py render/ext/nanogui/ext/pybind11/tests/test_keep_alive.py examples/env_init.py gym_cloth/blender/get_image_rep_279.py render/ext/nanogui/ext/pybind11/tests/test_callbacks.py render/ext/nanogui/ext/pybind11/tests/test_docstring_options.py render/ext/nanogui/ext/eigen/scripts/relicense.py analysis/combine_demo_data.py render/ext/nanogui/ext/pybind11/tests/test_kwargs_and_defaults.py render/ext/nanogui/ext/pybind11/docs/benchmark.py render/ext/nanogui/ext/pybind11/pybind11/_version.py render/ext/nanogui/ext/pybind11/tests/test_class_args.py render/ext/nanogui/docs/conf.py gym_cloth/__init__.py render/ext/nanogui/docs/mkdoc_rst.py render/ext/nanogui/ext/pybind11/tests/test_numpy_dtypes.py render/ext/nanogui/ext/pybind11/tests/test_opaque_types.py examples/demo_spaces.py render/ext/nanogui/ext/pybind11/tests/conftest.py examples/analytic.py render/ext/nanogui/ext/pybind11/tests/test_virtual_functions.py render/ext/nanogui/ext/pybind11/tests/test_sequences_and_iterators.py render/ext/nanogui/ext/pybind11/setup.py render/ext/nanogui/ext/pybind11/tools/mkdoc.py gym_cloth/envs/cloth_env.py render/ext/nanogui/ext/pybind11/tests/test_enum.py analysis/check_demo_data.py render/ext/nanogui/ext/pybind11/tests/test_numpy_vectorize.py render/ext/nanogui/python/example2.py render/ext/nanogui/ext/eigen/debug/gdb/__init__.py render/ext/nanogui/ext/pybind11/tests/test_constants_and_functions.py render/ext/nanogui/python/example4.py render/ext/nanogui/ext/pybind11/tools/libsize.py render/ext/nanogui/ext/eigen/debug/gdb/printers.py render/ext/nanogui/ext/pybind11/tests/test_modules.py render/ext/nanogui/ext/pybind11/tests/test_copy_move_policies.py render/ext/nanogui/ext/pybind11/tests/test_operator_overloading.py render/ext/nanogui/ext/pybind11/tools/clang/__init__.py setup.py render/ext/nanogui/ext/pybind11/tests/test_pickling.py render/ext/nanogui/ext/pybind11/tests/test_chrono.py render/ext/nanogui/ext/pybind11/tests/test_eigen.py render/ext/nanogui/ext/pybind11/docs/conf.py render/ext/nanogui/ext/pybind11/tests/test_eval_call.py render/ext/nanogui/ext/pybind11/tests/test_multiple_inheritance.py render/ext/nanogui/ext/pybind11/tests/test_cmake_build/test.py render/ext/nanogui/ext/pybind11/tests/test_exceptions.py render/ext/nanogui/ext/pybind11/tests/test_python_types.py render/ext/nanogui/ext/pybind11/tests/test_eval.py render/ext/nanogui/ext/pybind11/tests/test_buffers.py analyze get_numpy OracleCornerRevealPolicy HighestPointPolicy HarrisCornerPolicy OracleCornerPolicy Policy WrinklesPolicy run RandomPolicy test_init _create_scene_and_offscreen_render _save_trimesh _save_render_images _preprocess_depth run _corners_nodelta analytic_corners analytic _corners_delta run load_mesh set_floor_pose set_dom_rand set_bed_color set_camera_optical_center set_camera_pose render_image get_occlusion_vec set_cloth_color set_active BVHTreeAndVerticesInWorldFromObj set_bed_pose set_cloth_texture main compute_depth set_lighting set_camera_focal_length ClothEnv write_to_file load_from_file generateDoxygenXML setup qualifyKind kindAsBreatheDirective ExhaleNode exclaimError generate specificationsForKind ExhaleRoot extract ExtractionThread d sanitize_name process_comment EigenQuaternionPrinter lookup_function register_eigen_printers build_eigen_dictionary EigenMatrixPrinter update generate_dummy_code_pybind11 generate_dummy_code_boost get_include test_isinstance test_inheritance test_automatic_upcasting suppress _strip_and_dedent Output doc _sanitize_general pytest_namespace _test_import_pybind11 gc_collect _make_explanation Capture msg _sanitize_docstring _split_and_sort _sanitize_message pytest_assertrepr_compare SanitizedString Unordered capture test_alias_delay_initialization2 test_alias_delay_initialization1 test_from_python test_to_python test_callbacks test_keyword_args_and_generalized_unpacking test_function_signatures test_cpp_function_roundtrip test_lambda_closure_cleanup test_chrono_duration_roundtrip test_chrono_system_clock test_chrono_steady_clock_roundtrip test_chrono_duration_subtraction_equivalence test_floating_point_duration test_chrono_system_clock_roundtrip test_chrono_steady_clock test_class_args test_constants test_exception_specifiers test_bytes test_function_overloading test_lacking_copy_ctor test_lacking_move_ctor test_docstring_options test_dense_signature test_dense test_special_matrix_objects test_nonunit_stride_to_python assert_equal_ref assert_sparse_equal_ref test_eigen_ref_to_python test_sparse_signature test_fixed test_nonunit_stride_from_python test_sparse test_scoped_enum test_implicit_conversion test_binary_operators test_unscoped_enum test_evals test_error_already_set test_python_call_in_catch test_custom test_dupe_assignment test_reference_wrapper test_shared_ptr_gc test_dispatch_issue test_override_ref test_regressions test_complex_cast test_operators_notimplemented test_iterator_rvpolicy test_enable_shared_from_this_with_reference_rvp test_inheritance_override_def_static test_move_fallback test_str_issue test_no_id test_non_destructed_holders test_iterator_passthrough test_nested test_keep_alive_return_value test_keep_alive_argument test_return_none test_function_signatures test_arg_and_kwargs test_named_arguments test_property_rvalue_policy_static test_properties test_static_properties test_property_return_value_policies test_dynamic_attributes test_cyclic_gc test_methods_and_attributes test_property_rvalue_policy test_importing test_reference_internal test_nested_modules test_multiple_inheritance_cpp test_multiple_inheritance_mix1 test_multiple_inheritance_virtbase test_multiple_inheritance_mix2 test_at_fail test_constructors test_dim_check_fail test_numpy_view test_isinstance test_make_c_f_array test_cast_numpy_int64_to_uint64 test_at test_data test_wrap test_index_offset test_mutate_data test_mutate_readonly test_array_attributes arr test_bounds_check test_register_dtype test_enum_array test_signature test_array_constructors test_dtype test_string_array test_scalar_conversion packed_dtype test_format_descriptors assert_equal simple_dtype test_recarray test_vectorize test_docs test_type_selection test_pointers test_string_list test_operator_overloading test_roundtrip_with_dict test_roundtrip test_str_api test_repr test_move_out_container test_exp_optional test_method_docs test_class_docs test_instance test_module test_static test_implicit_casting test_dict_api test_print test_constructors test_accessors test_optional test_sequence test_map_iterator allclose isclose test_generalized_iterators test_shared_ptr_and_references test_smart_ptr test_unique_nodelete test_shared_ptr_from_this_and_references test_smart_ptr_refcounting test_map_string_double test_vector_int test_vector_custom test_vector_bool test_noncopyable_unordered_map test_noncopyable_vector test_noncopyable_map test_noncopyable_deque test_map_string_double_const test_override test_move_support test_inheriting_repeat extract ExtractionThread d sanitize_name process_comment Config SourceRange register_enumerations RefQualifierKind CursorKind TokenGroup register_functions TokenKind FixIt CompileCommands register_function CachedProperty ClangObject CompileCommand CompilationDatabase FileInclusion Diagnostic LibclangError CompletionString CompletionChunk Index StorageClass TranslationUnitLoadError BaseEnumeration AccessSpecifier CodeCompletionResults TypeKind TemplateArgumentKind TranslationUnitSaveError Token CCRStructure _CXUnsavedFile File _CXString CodeCompletionResult TranslationUnit Type Cursor SourceLocation CompilationDatabaseError TestApp cb make_accessors TestApp MyGLCanvas TestApp join format imwrite std print min zfill mean array resize append zeros sum max range enumerate len format print len append array enumerate seed cfg_file defaultdict set_env_cfg format max_episodes print ClothEnv mean stop_render render reset terminate append get_action step std range join time format print action_space ClothEnv load_state observation_space stop_render realpath render reset shape dirname terminate save_state step depth_scaled_to_255 depth_to_3ch format imwrite zfill render _preprocess_depth len format remove_node zfill from_trimesh Node _save_render_images Trimesh export append array range add_node len OffscreenRenderer add SpotLight Scene array PerspectiveCamera join _create_scene_and_offscreen_render _save_trimesh get_random_action save_state get_random_action _get_xy_vals pts pi arctan2 _get_xy_vals pi sqrt pts analytic_corners obj replace append new array minimum remove new maximum links node_tree append float array render save_render replace FromPolygons matrix_world areas Vector ray_cast BVHTreeAndVerticesInWorldFromObj mode_set normalized append select_all enumerate remove new nodes links node_tree load_mesh int set_dom_rand set_bed_color argv set_floor_pose set_camera_pose set_camera_optical_center get_occlusion_vec render_image set_cloth_color delete set_bed_pose compute_depth set_camera_focal_length shade_smooth dump close open open communicate bytes close Popen generateDoxygenXML dedent generate parse ExhaleRoot customSpecificationFunction abspath generateFullAPI breathe_parse format write items join sub replace items rstrip replace endswith strip min lstrip split splitlines sub startswith TextWrapper fill float len d sanitize_name process_comment append get_children spelling append strip_typedefs search tag target type randint range randint range Pet Hamster Rabbit Dog strip sub replace _sanitize_general __doc__ str sub _sanitize_general hasattr collect skipif get rows Matrix astype cols float32 range get gc_collect Matrix array test_callback4 test_callback5 payload_cstats test_cleanup abs today test_chrono1 abs today test_chrono2 test_chrono3 today today test_chrono4 test_chrono5 timedelta test_chrono6 test_chrono7 class_args_noop C assert_array_equal assert_equal_ref todense assert_equal_ref fixed_r fixed_passthrough_r fixed_c fixed_passthrough_c assert_equal_ref dense_passthrough_c dense_passthrough_r dense_r dense_c reshape enumerate range array enumerate chol array range sparse_c sparse_passthrough_c assert_sparse_equal_ref sparse_r sparse_passthrough_r ETwo Two ESecondMode test_function EFirstMode Write Read join dirname PyClass2 get ElementA ElementList gc_collect add range enumerate gc_collect as_base NestB NestC NestA get_moveissue1 get_moveissue2 OverrideTest A_ref get_child SharedParent get holder_cstats child SpecialHolderObj make make2 dict get add8 add1 add7 add5 ExampleMandA add4 add6 add9 add3 add10 add2 TestProperties TestPropRVP getattr TestPropRVP rvalue static_rvalue get cls DynamicClass get DynamicClass A B MIType MITypePy MITypePy MITypePy array view assert_references transpose random wrap diagonal array function_taking_uint64 uint64 list default_constructors converting_constructors assert_array_equal values array dtype dtype create_rec_nested assert_equal func create_rec_partial create_rec_partial_nested arange reshape assert_array_equal range test_array_ctors create_string_array dtype dtype create_enum_array enumerate StringList push_back pop_back ClassWithSTLVecProperty enumerate return_unique_ptr get Vector Vector2 setExtra1 Pickleable setExtra2 dumps loads loads PickleableWithDict HIGHEST_PROTOCOL dumps get_array get get_valarray new_instance test_str_format test_accessor_api test_accessor_assignment TestObject raises raises move_list MoveOutContainer get_implicit_casting get Sequence reversed items list StringMap get print zip enumerate cstats_ref get MyObject4 get holder_ref ref holder_copy copy SharedPtrRef get holder_ref ref holder_copy copy bad_wp SharedFromThisRef append insert VectorInt append VectorVectorEl VectorEl El append VectorBool range UnorderedMapStringDouble MapStringDouble MapStringDoubleConst UnorderedMapStringDoubleConst str range enumerate get_vnc range get_dnc items list range get_mnc get_umnc list range items get ExtendedExampleVirt2 ExtendedExampleVirt ExampleVirt BT CR DT CCR CCT AT DR AR CT get NCVirtExt2 NCVirtExt getattr register register TokenKinds print

# Gym Cloth Quick logistics overview: this is *one* of the code bases used in our paper "Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisor" with [arXiv here][3] and [project website here][4]. The arXiv version will have the most up-to-date version of the paper. <hr> This creates a gym environment based on our cloth simulator. The path directory is structured following [standard gym conventions][1], and we also include our `.pyx` files here for Cython compilation. Platforms tested: - Mac OS X (renderer working) - Ubuntu 16.04 (renderer not working, unfortunately) - Ubuntu 18.04 (renderer working)

DanieleAlessandro/KENN2

['multi label classification']

['Neural Networks Enhancement with Logical Knowledge']

src/KENN2/layers/relational/Join.py src/KENN2/layers/residual/ClauseEnhancer.py src/KENN2/parsers.py src/KENN2/layers/Kenn.py src/KENN2/layers/residual/KnowledgeEnhancer.py src/KENN2/layers/RelationalKENN.py src/KENN2/layers/relational/GroupBy.py src/KENN2/layers/RangeConstraint.py relational_parser unary_parser unary_parser_ke Kenn RangeConstraint RelationalKENN GroupBy Join ClauseEnhancer KnowledgeEnhancer split split append

# KENN: Knowledge Enhanced Neural Networks KENN2 (Knowledge Enhanced Neural Networks 2.0) is a library for Python 3 built on top of TensorFlow 2 that allows you to modify neural network models by providing logical knowledge in the form of a set of universally quantified FOL clauses. It does so by adding a new final layer, called **Knowledge Enhancer (KE)**, to the existing neural network. The KE changes the original predictions of the standard neural network enforcing the satisfaction of the knowledge. Additionally, it contains **clause weights**, learnable parameters which represent the strength of each clause. **NB:** version 1.0 of KENN was released for Python 2.7 and TensorFlow 1.x and it is available at [KENN v1.0](https://github.com/DanieleAlessandro/KENN). Notice that this version is not backward compatible. Additionally, this implementation of KENN can work with relational domains, meaning that one can use also binary predicates to express logical rules which involve the relationship between two objects. This is an implementation of the model presented in our paper: [Knowledge Enhanced Neural Networks](https://link.springer.com/chapter/10.1007/978-3-030-29908-8_43). If you use this software for academic research, please, cite our work using the following BibTeX: ``` @InProceedings{10.1007/978-3-030-29908-8_43, author="Daniele, Alessandro and Serafini, Luciano",

DanieleGammelli/CensoredGP

['gaussian processes']

['Estimating Latent Demand of Shared Mobility through Censored Gaussian Processes']

GPy/likelihoods/loggaussian.py GPy/kern/src/psi_comp/linear_psi_comp.py GPy/core/parameterization/transformations.py GPy/kern/src/sde_matern.py GPy/plotting/matplot_dep/variational_plots.py GPy/core/parameterization/priors.py GPy/inference/latent_function_inference/var_dtc.py GPy/kern/src/standard_periodic.py GPy/kern/src/psi_comp/ssrbf_psi_comp.py GPy/old_tests/gplvm_tests.py GPy/plotting/abstract_plotting_library.py GPy/plotting/matplot_dep/util.py GPy/inference/latent_function_inference/grid_posterior.py GPy/plotting/matplot_dep/__init__.py GPy/likelihoods/weibull.py GPy/util/cluster_with_offset.py GPy/util/parallel.py GPy/plotting/matplot_dep/svig_plots.py benchmarks/regression/run.py GPy/models/gp_heteroscedastic_regression.py GPy/kern/src/todo/ODE_1.py GPy/core/sparse_gp.py GPy/kern/src/static.py GPy/mappings/mlpext.py GPy/testing/pickle_tests.py GPy/kern/src/kern.py GPy/models/warped_gp.py GPy/plotting/plotly_dep/plot_definitions.py GPy/models/one_vs_all_classification.py GPy/kern/src/grid_kerns.py GPy/models/gp_offset_regression.py GPy/testing/link_function_tests.py GPy/inference/latent_function_inference/var_dtc_parallel.py GPy/likelihoods/gamma.py GPy/testing/tp_tests.py GPy/plotting/gpy_plot/latent_plots.py GPy/kern/src/rbf.py GPy/testing/kernel_tests.py GPy/models/input_warped_gp.py GPy/examples/dimensionality_reduction.py GPy/models/sparse_gplvm.py GPy/plotting/matplot_dep/priors_plots.py GPy/inference/latent_function_inference/exact_gaussian_inference.py GPy/core/svgp.py GPy/kern/src/sde_standard_periodic.py GPy/models/one_vs_all_sparse_classification.py GPy/core/sparse_gp_mpi.py GPy/kern/src/splitKern.py GPy/plotting/matplot_dep/maps.py GPy/util/initialization.py GPy/inference/latent_function_inference/vardtc_md.py GPy/plotting/gpy_plot/kernel_plots.py GPy/models/sparse_gp_minibatch.py GPy/util/config.py GPy/kern/src/ODE_UYC.py setup.py GPy/plotting/gpy_plot/plot_util.py GPy/testing/variational_tests.py GPy/kern/src/trunclinear.py GPy/inference/__init__.py GPy/testing/grid_tests.py GPy/models/ss_mrd.py GPy/plotting/matplot_dep/plot_definitions.py GPy/testing/state_space_main_tests.py GPy/kern/src/integral.py GPy/kern/src/stationary.py GPy/core/symbolic.py GPy/mappings/kernel.py GPy/mappings/piecewise_linear.py GPy/models/gplvm.py GPy/testing/pep_tests.py GPy/old_tests/cgd_tests.py GPy/models/__init__.py GPy/inference/latent_function_inference/exact_studentt_inference.py GPy/models/gp_regression.py GPy/kern/src/psi_comp/rbf_psi_gpucomp.py GPy/testing/minibatch_tests.py GPy/testing/prior_tests.py GPy/util/squashers.py GPy/util/mocap.py benchmarks/regression/outputs.py doc/source/conf.py GPy/kern/src/todo/hetero.py GPy/models/state_space.py GPy/models/mrd.py GPy/core/mapping.py GPy/inference/mcmc/samplers.py GPy/likelihoods/poisson.py GPy/util/linalg.py GPy/likelihoods/multioutput_likelihood.py GPy/inference/latent_function_inference/inferenceX.py GPy/models/bayesian_gplvm.py GPy/likelihoods/loglogistic.py GPy/testing/gpy_kernels_state_space_tests.py GPy/core/parameterization/variational.py GPy/testing/util_tests.py GPy/kern/src/todo/finite_dimensional.py travis_tests.py GPy/old_tests/mapping_tests.py GPy/examples/non_gaussian.py GPy/likelihoods/exponential.py GPy/testing/mapping_tests.py GPy/inference/latent_function_inference/fitc.py GPy/kern/src/multidimensional_integral_limits.py benchmarks/regression/methods.py GPy/core/parameterization/priorizable.py GPy/kern/src/todo/symmetric.py GPy/inference/optimization/__init__.py GPy/models/state_space_setup.py GPy/testing/gp_tests.py GPy/models/gp_kronecker_gaussian_regression.py GPy/util/subarray_and_sorting.py GPy/kern/src/poly.py benchmarks/regression/tasks.py GPy/old_tests/psi_stat_gradient_tests.py GPy/likelihoods/censored_gaussian.py GPy/plotting/matplot_dep/controllers/imshow_controller.py GPy/testing/linalg_test.py GPy/kern/src/periodic.py GPy/util/choleskies.py GPy/testing/model_tests.py GPy/kern/src/brownian.py GPy/testing/plotting_tests.py GPy/models/sparse_gp_regression.py GPy/inference/latent_function_inference/dtc.py GPy/examples/classification.py GPy/models/gradient_checker.py GPy/models/state_space_model.py GPy/mappings/compound.py GPy/old_tests/psi_stat_expectation_tests.py GPy/util/debug.py GPy/kern/src/psi_comp/__init__.py GPy/models/gp_classification.py GPy/kern/src/spline.py GPy/models/tp_regression.py GPy/old_tests/gp_transformation_tests.py GPy/kern/src/linear.py GPy/plotting/Tango.py GPy/util/__init__.py GPy/testing/fitc.py GPy/testing/likelihood_tests.py GPy/__version__.py GPy/plotting/matplot_dep/defaults.py GPy/kern/src/eq_ode1.py GPy/plotting/matplot_dep/base_plots.py GPy/plotting/gpy_plot/__init__.py GPy/likelihoods/mixed_noise.py GPy/kern/src/psi_comp/rbf_psi_comp.py GPy/kern/src/mlp.py GPy/kern/src/ODE_st.py GPy/models/gp_censored_regression.py GPy/models/sparse_gp_regression_md.py GPy/plotting/__init__.py GPy/kern/src/add.py GPy/kern/src/integral_limits.py GPy/util/multioutput.py GPy/models/bcgplvm.py GPy/old_tests/bcgplvm_tests.py GPy/mappings/mlp.py GPy/inference/latent_function_inference/posterior.py GPy/kern/src/todo/gibbs.py GPy/kern/src/__init__.py GPy/testing/ep_likelihood_tests.py GPy/models/gp_coregionalized_regression.py GPy/models/sparse_gp_coregionalized_regression.py GPy/mappings/__init__.py GPy/models/gp_multiout_regression_md.py GPy/models/gp_grid_regression.py GPy/plotting/matplot_dep/visualize.py GPy/testing/meanfunc_tests.py GPy/util/classification.py GPy/util/diag.py GPy/util/input_warping_functions.py GPy/util/linalg_gpu.py GPy/util/misc.py GPy/mappings/identity.py GPy/kern/__init__.py GPy/examples/__init__.py GPy/models/ibp_lfm.py GPy/inference/latent_function_inference/var_gauss.py GPy/inference/latent_function_inference/laplace.py GPy/inference/latent_function_inference/vardtc_svi_multiout.py GPy/testing/cython_tests.py GPy/likelihoods/__init__.py GPy/util/quad_integrate.py GPy/mappings/linear.py GPy/inference/latent_function_inference/vardtc_svi_multiout_miss.py GPy/examples/regression.py GPy/models/ss_gplvm.py GPy/kern/src/todo/eq_ode1.py GPy/inference/latent_function_inference/gaussian_grid_inference.py GPy/testing/__init__.py GPy/kern/src/todo/fixed.py GPy/inference/latent_function_inference/pep.py GPy/plotting/gpy_plot/gp_plots.py GPy/core/gp.py GPy/plotting/plotly_dep/defaults.py GPy/core/model.py GPy/kern/src/psi_comp/sslinear_psi_comp.py GPy/util/normalizer.py GPy/testing/mpi_tests.py GPy/models/dpgplvm.py GPy/util/decorators.py GPy/inference/optimization/stochastics.py GPy/inference/latent_function_inference/expectation_propagation.py GPy/inference/mcmc/hmc.py GPy/plotting/matplot_dep/img_plots.py GPy/models/bayesian_gplvm_minibatch.py GPy/testing/misc_tests.py GPy/inference/latent_function_inference/svgp.py GPy/likelihoods/likelihood.py GPy/testing/svgp_tests.py GPy/kern/src/coregionalize.py GPy/util/gpu_init.py GPy/core/gp_grid.py GPy/likelihoods/binomial.py GPy/models/gp_multiout_regression.py GPy/plotting/gpy_plot/data_plots.py GPy/core/__init__.py GPy/plotting/matplot_dep/mapping_plots.py GPy/util/datasets.py GPy/plotting/matplot_dep/ssgplvm.py GPy/kern/src/symmetric.py GPy/util/netpbmfile.py GPy/util/warping_functions.py GPy/testing/serialization_tests.py GPy/examples/state_space.py GPy/kern/src/sde_static.py GPy/plotting/matplot_dep/controllers/axis_event_controller.py GPy/models/gp_var_gauss.py GPy/plotting/gpy_plot/inference_plots.py GPy/kern/src/prod.py GPy/core/parameterization/__init__.py GPy/models/sparse_gp_classification.py GPy/core/parameterization/parameterized.py GPy/kern/src/eq_ode2.py GPy/likelihoods/link_functions.py GPy/kern/src/sde_brownian.py GPy/kern/src/symbolic.py GPy/mappings/constant.py GPy/testing/rv_transformation_tests.py GPy/kern/src/todo/spline.py GPy/kern/src/psi_comp/ssrbf_psi_gpucomp.py GPy/testing/examples_tests.py GPy/kern/src/todo/rbf_inv.py GPy/models/state_space_main.py GPy/testing/inference_tests.py GPy/util/ln_diff_erfs.py benchmarks/regression/evaluation.py GPy/kern/src/kernel_slice_operations.py GPy/likelihoods/gaussian.py GPy/kern/src/independent_outputs.py GPy/util/pca.py GPy/kern/src/todo/poly.py GPy/kern/src/sde_stationary.py GPy/mappings/additive.py GPy/plotting/matplot_dep/controllers/__init__.py GPy/old_tests/sparse_gplvm_tests.py GPy/util/block_matrices.py GPy/util/functions.py GPy/testing/quadrature_tests.py GPy/util/univariate_Gaussian.py GPy/kern/src/sde_linear.py GPy/likelihoods/student_t.py GPy/inference/mcmc/__init__.py GPy/kern/src/psi_comp/gaussherm.py GPy/likelihoods/bernoulli.py GPy/core/parameterization/param.py GPy/kern/src/basis_funcs.py GPy/kern/src/ODE_UY.py GPy/inference/latent_function_inference/__init__.py GPy/kern/src/multioutput_kern.py GPy/kern/src/ODE_t.py GPy/__init__.py ismac read read_to_rst RMSE Evaluation RegressionMethod GP_RBF SparseGP_RBF SVIGP_RBF H5Output ScreenOutput Output CSVOutput RegressionTask Housing WineQuality Mock load tests GP GpGrid Bijective_mapping Mapping Model SparseGP SparseGP_MPI SVGP setInDict getFromDict Symbolic_core randomize Param Parameterized Priorizable DGPLVM Gamma LogGaussian DGPLVM_T DGPLVM_Lamda StudentT HalfT Gaussian gamma_from_EV Prior InverseGamma DGPLVM_KFDA Exponential MultivariateGaussian Uniform VariationalPrior VariationalPosterior NormalPosterior SpikeAndSlabPosterior SpikeAndSlabPrior NormalPrior toy_heaviside sparse_toy_linear_1d_classification_uncertain_input sparse_toy_linear_1d_classification oil crescent_data toy_linear_1d_classification toy_linear_1d_classification_laplace bcgplvm_linear_stick stick gplvm_oil_100 robot_wireless brendan_faces bgplvm_simulation bcgplvm_stick stick_bgplvm cmu_mocap ssgplvm_simulation_linear bgplvm_oil ssgplvm_simulation bgplvm_simulation_missing_data _generate_high_dimensional_output mrd_simulation mrd_simulation_missing_data sparse_gplvm_oil ssgplvm_oil bgplvm_simulation_missing_data_stochastics swiss_roll _simulate_matern gplvm_simulation stick_play bgplvm_test_model olivetti_faces _simulate_sincos student_t_approx boston_example _contour_data robot_wireless silhouette uncertain_inputs_sparse_regression coregionalization_sparse toy_ARD_sparse olympic_marathon_men warped_gp_cubic_sine coregionalization_toy sparse_GP_regression_1D olympic_100m_men toy_ARD multiple_optima toy_poisson_rbf_1d_laplace toy_rbf_1d sparse_GP_regression_2D parametric_mean_function simple_mean_function toy_rbf_1d_50 epomeo_gpx state_space_example vDTC DTC ExactGaussianInference ExactStudentTInference EPCensored gaussianApproximation posteriorParamsDTC cavityParams posteriorParamsBase posteriorParams EP EPBase EPDTC marginalMoments FITC GaussianGridInference GridPosterior InferenceX infer_newX warning_on_one_line LaplaceBlock Laplace PEP Posterior PosteriorExact PosteriorEP StudentTPosterior SVGP VarDTC_MD VarDTC_SVI_Multiout PosteriorMultioutput VarDTC_SVI_Multiout_Miss _compute_log_marginal_likelihood _compute_dL_dpsi VarDTC _compute_dL_dR VarDTC_minibatch update_gradients update_gradients_sparsegp VarGauss InferenceMethodList LatentFunctionInference HMC_shortcut HMC Metropolis_Hastings StochasticStorage SparseGPMissing SparseGPStochastics ODE_t Add LogisticBasisFuncKernel DomainKernel PolynomialBasisFuncKernel LinearSlopeBasisFuncKernel BasisFuncKernel ChangePointBasisFuncKernel Brownian Coregionalize EQ_ODE1 lnDifErf EQ_ODE2 GridRBF GridKern IndependentOutputs Hierarchical index_to_slices Integral Integral_Limits CombinationKernel Kern _slice_gradients_XX_diag _slice_update_gradients_expectations put_clean _slice_gradients_Z_expectations _slice_update_gradients_diag _slice_gradients_qX_expectations KernCallsViaSlicerMeta _slice_K _slice_psi _Slice_wrap _slice_gradients_X_diag _slice_gradients_X _slice_Kdiag _slice_update_gradients_full _slice_gradients_XX LinearFull Linear MLP Multidimensional_Integral_Limits ZeroKern MultioutputKern ODE_st ODE_UY ODE_UYC PeriodicMatern32 PeriodicExponential Periodic PeriodicMatern52 Poly Prod numpy_invalid_op_as_exception dkron RBF sde_Brownian sde_Linear sde_Matern52 sde_Matern32 seriescoeff sde_StdPeriodic sde_White sde_Bias sde_Exponential sde_RatQuad sde_RBF Spline SplitKern DEtime SplitKern_cross StdPeriodic WhiteHeteroscedastic Fixed Static Precomputed Bias White ExpQuad Cosine RatQuad Matern52 Exponential Stationary OU Matern32 Symbolic Symmetric TruncLinear TruncLinear_inf PSICOMP_GH psiDerivativecomputations psicomputations _psi2computations __psi2computations psicomputations _psi2compDer __psi1computations _psi1compDer psiDerivativecomputations PSICOMP_RBF_GPU psiDerivativecomputations psicomputations _psi2computations _psicomputations psiDerivativecomputations PSICOMP_SSRBF_GPU PSICOMP_RBF PSICOMP_Linear PSICOMP Eq_ode1 FiniteDimensional Fixed Gibbs Hetero ODE_1 POLY RBFInv Spline theta Symmetric Bernoulli Binomial CensoredGaussian Exponential Gamma Gaussian HeteroscedasticGaussian Likelihood GPTransformation Reciprocal Heaviside Cloglog Log Probit Identity Log_ex_1 LogGaussian LogLogistic MixedNoise MultioutputLikelihood Poisson StudentT Weibull Additive Compound Constant Identity Kernel Linear MLP MLPext PiecewiseLinear GPVariationalGaussianApproximation BayesianGPLVM BayesianGPLVMMiniBatch BCGPLVM DPBayesianGPLVM GPLVM GPCensoredRegression GPClassification GPCoregionalizedRegression GPRegressionGrid GPHeteroscedasticRegression GPKroneckerGaussianRegression GPMultioutRegression GPMultioutRegressionMD GPOffsetRegression GPRegression get_shape GradientChecker SkewChecker HessianChecker flatten_if_needed at_least_one_element IBPPosterior VarDTC_minibatch_IBPLFM IBPLFM update_gradients IBPPrior InputWarpedGP MRD OneVsAllClassification OneVsAllSparseClassification SparseGPLVM SparseGPClassification SparseGPClassificationUncertainInput SparseGPCoregionalizedRegression SparseGPMiniBatch SparseGPRegression SparseGPRegressionMD IBPPosterior SLVMPosterior SSGPLVM SLVMPrior IBPPrior IBPPrior_SSMRD SpikeAndSlabPrior_SSMRD SSMRD StateSpace Struct ContDescrStateSpace balance_ss_model DescreteStateSpaceMeta matrix_exponent Std_Dynamic_Callables_Python Measurement_Callables_Python Q_handling_Python Std_Measurement_Callables_Python DescreteStateSpace balance_matrix R_handling_Python Dynamic_Callables_Python AddMethodToClass StateSpace TPRegression WarpedGP BCGPLVMTests callback Test GPLVMTests TestTransformations MappingGradChecker MappingTests ard Test DPsiStatTest PsiStatModel sparse_GPLVMTests AbstractPlottingLibrary hex2rgb nextMedium currentLight nextLight nextDark reset currentMedium currentDark show plotting_library change_plotting_library inject_plotting plot_inducing _plot_data_error _plot_inducing _plot_errorbars_trainset plot_data_error plot_errorbars_trainset _plot_data plot_data _plot_samples _plot plot plot_density plot_samples _plot_confidence _plot_mean _plot_density plot_confidence plot_mean plot_f plot_sgd_traces plot_optimizer plot_covariance plot_ARD plot_latent_inducing _plot_latent_scatter _plot_latent plot_steepest_gradient_map _plot_steepest_gradient_map _wait_for_updates plot_magnification plot_latent_scatter _new_canvas _plot_magnification plot_latent get_fixed_dims find_best_layout_for_subplots get_which_data_ycols scatter_label_generator x_frame2D in_ipynb helper_predict_with_model update_not_existing_kwargs get_free_dims subsample_X get_which_data_rows helper_for_plot_data x_frame1D get_x_y_var gpplot x_frame1D meanplot gradient_fill align_subplot_array x_frame2D removeRightTicks fewerXticks align_subplots ax_default gperrors removeUpperTicks plot_2D_images _calculateFigureSize plot_mapping plot_string_match bbox_match apply_bbox plot plot_bbox string_match new_shape_string MatplotlibPlots univariate_plot plot SSGPLVM_plot plot_traces plot legend_ontop align_subplot_array fixed_inputs removeRightTicks fewerXticks align_subplots removeUpperTicks plot_SpikeSlab plot stick_show lvm_dimselect data_play data_show matplotlib_show mocap_data_show image_show vpython_show vector_show skeleton_show lvm lvm_subplots mocap_data_show_vpython BufferedAxisChangedController AxisEventController AxisChangedController ImAnnotateController ImshowController PlotlyPlotsBase PlotlyPlotsOnline PlotlyPlotsOffline test_choleskies_backprop test_stationary CythonTestChols TestObservationModels ExamplesTests model_checkgrads test_models model_instance flatten_nested FITCtest StateSpaceKernelsTests Test GridModelTest HMCSamplerTest VarDtcTest MCMCSamplerTest InferenceGPEP InferenceXTestCase KernelGradientTestsContinuous Kern_check_dK_dX Kern_check_dK_dtheta KernelTestsNonContinuous Coregionalize_cython_test Kern_check_dKdiag_dtheta Kern_check_d2K_dXdX Kern_check_d2Kdiag_dXdX KernelTestsProductWithZeroValues Kern_check_model Kernel_Psi_statistics_GradientTests Kern_check_dKdiag_dX KernelTestsMiscellaneous check_kernel_gradient_functions dparam_partial TestNoiseModels dparam_checkgrad LaplaceTests LinalgTests LinkFunctionTests MappingGradChecker MappingTests MFtests SparseGPMinibatchTest BGPLVMTest MiscTests _create_missing_data_model MiscTests GradientTests MPITests PEPgradienttest toy_model Test ListDictTestCase test_bayesian_gplvm flatten_axis test_figure test_classification test_sparse_classification test_gplvm _image_comparison compare_axis_dicts _image_directories ConfigTest test_plot test_twod _a test_kernel test_sparse test_threed PriorTests QuadTests f RVTransformationTestCase TestModel Test generate_linear_data generate_linear_plus_sin generate_sine_data generate_random_y_data StateSpaceKernelsTests generate_brownian_data generate_x_points SVGP_nonconvex SVGP_classification SVGP_Poisson_with_meanfunction Test TestUnivariateGaussian TestDebug TestStandardize Test KLGrad deepTest block_dot unblock get_blocks get_blocks_3d get_block_shapes get_block_shapes_3d _flat_to_triang_pure indexes_to_fix_for_low_rank _flat_to_triang_cython multiple_dpotri safe_root _backprop_gradient_pure _triang_to_flat_cython triang_to_cov _triang_to_flat_pure conf_matrix cluster get_log_likelihood get_log_likelihood_offset olympic_200m_women cmu_mocap_35_walk_jog robot_wireless olympic_sprints global_average_temperature brendan_faces swiss_roll_1000 boston_housing data_details_return crescent_data olympic_200m_men silhouette cifar10_patches osu_run1 cmu_mocap_49_balance cmu_mocap oil_100 drosophila_knirps drosophila_protein singlecell fruitfly_tomancak spellman_yeast ripley_synth download_url google_trends data_available football_data xw_pen download_rogers_girolami_data mauna_loa toy_linear_1d_classification olympic_400m_women singlecell_rna_seq_deng olympic_marathon_men boxjenkins_airline cmu_urls_files decampos_digits isomap_faces olivetti_glasses olympic_100m_men sample_class creep_data download_data swiss_roll singlecell_rna_seq_islam prompt_user della_gatta_TRP63_gene_expression simulation_BGPLVM spellman_yeast_cdc15 hapmap3 olympic_100m_women olympic_400m_men toy_rbf_1d sod1_mouse oil reporthook pumadyn lee_yeast_ChIP olivetti_faces swiss_roll_generated toy_rbf_1d_50 authorize_download epomeo_gpx checkFullRank checkFinite silence_errors offdiag_view view subtract divide add times _diag_ufunc logisticln normcdf logistic clip_exp normcdfln differfln closeGPU initialize_latent IdentifyWarping KumarWarping InputWarpingFunction InputWarpingTest dtrtri ijk_ljk_to_ilk ijk_jlk_to_il multiple_pdinv pddet dpotri pca symmetrify tdot_blas trace_dot ij_jlk_to_ilk force_F_ordered_symmetric _symmetrify_cython DSYR_blas ppca DSYR_numpy _symmetrify_numpy _mdot_r backsub_both_sides pdinv force_F_ordered tdot_numpy jitchol mdot dtrtrs DSYR tdot dpotrs ln_diff_erfs safe_quad chain_2 kmm_init param_to_array blockify_dhess_dtheta linear_grid blockify_third safe_three_times blockify_hessian safe_cube safe_square chain_1 safe_exp chain_3 opt_wrapper rotation_matrix acclaim_skeleton tree parse_text read_connections vertex skeleton load_text_data Private ICM build_likelihood get_slices LCM build_XY NetpbmFile imread imsave Standardize _Norm get_id_within_node divide_data optimize_parallel PCA quadvgk quadgk_int getSubs sigmoid softmax single_softmax common_subarrays logPdfNormal _erfRationalHelper derivLogCdfNormal cdfNormal std_norm_pdf _erfRationalHelperR3 inv_std_norm_cdf logCdfNormal WarpingFunction IdentityFunction TanhFunction LogFunction read format convert test getmembers optimization src normal ones copy rand_gen update_model warn SparseGPClassification optimize print conf_matrix optimize subplots GPClassification plot print plot_f RBF Bernoulli optimize subplots plot print plot_f GP toy_linear_1d_classification Laplace SparseGPClassification optimize subplots plot print toy_linear_1d_classification plot_f optimize subplots plot print shape uniform toy_linear_1d_classification SparseGPClassificationUncertainInput plot_f RBF Bernoulli optimize subplots plot print EP GP toy_linear_1d_classification range plot_f SparseGPClassification optimize GPClassification plot print FITCClassification parameters_changed RBF T optimize plot BayesianGPLVM rand title nan K VarDTCMissingData RBF optimize GPLVM Bias plot_latent argmax oil_100 seed RBF optimize Bias oil mean SparseGPLVM plot_ARD plot_latent argmax RBF ones_like exp optimize set_title randn BayesianGPLVM hstack add_subplot Bias White scatter figure swiss_roll_generated embedding_ clip fit seed RBF optimize subplots lvm_dimselect BayesianGPLVM close oil uniform eval vector_show plot_latent input argmax seed RBF optimize subplots lvm_dimselect close SSGPLVM oil uniform eval vector_show plot_latent input argmax add_subplot clf K Matern32 seed set_title _generate_high_dimensional_output imshow legend range format plot tight_layout zip enumerate T draw figure len add_subplot sS clf seed set_title _generate_high_dimensional_output imshow legend format plot tight_layout s2 zip enumerate draw figure s3 vectorize s1 len dot hstack randn optimize plot BayesianGPLVM print shape uniform _simulate_matern plot_ARD Linear optimize plot print GPLVM plot_ARD _simulate_matern Linear optimize plot print SSGPLVM shape uniform _simulate_matern plot_ARD Linear optimize plot print astype copy plot_ARD _simulate_matern nan BayesianGPLVMMiniBatch Linear optimize plot print astype copy plot_ARD _simulate_matern nan BayesianGPLVMMiniBatch Linear optimize MRD plot print plot_scales White _simulate_sincos Linear optimize MRD plot print astype plot_scales White _simulate_matern nan append Linear optimize input BayesianGPLVM copy mean image_show eval plot_latent lvm optimize input BayesianGPLVM copy mean image_show eval plot_latent lvm data_play stick_show osu_run1 copy optimize stick_show input close GPLVM copy eval clf osu_run1 plot_latent lvm optimize stick_show input copy BCGPLVM clf eval osu_run1 plot_latent lvm Linear Kernel RBF optimize stick_show copy BCGPLVM clf osu_run1 plot_latent lvm optimize plot_latent BayesianGPLVM RBF optimize subplots stick_show lvm_dimselect BayesianGPLVM draw copy eval osu_run1 plot_latent input sca optimize input close GPLVM copy mean eval skeleton_show plot_latent lvm std randn dot sample_X empty range randn f_hat constrain_fixed Y max subplot ylim title sin plot copy White GP RBF randomize optimize suptitle print constrain_bounded StudentT figure GPRegression X Laplace boston_housing add_subplot grid constrain_fixed set_xlabel gaussian shape title scatter setp student_t predict KFold format log_predictive_density Laplace copy mean boxplot enumerate RBF T optimize print set_axisbelow constrain_bounded white bias set_ylabel rmse GPRegression constrain_positive figure zeros std len GPRegression optimize plot optimize plot randn rand get_slices GPCoregionalizedRegression sin optimize plot randn rand ylim get_slices sin SparseGPCoregionalizedRegression SparseGPRegression RBF list optimize hstack constrain_bounded range Coregionalize constrain_fixed zip append zeros sum array len _contour_data linspace seed exp arrow ylabel uniform log10 gca append range set_xlim mean lengthscale get_ylim get_xlim della_gatta_TRP63_gene_expression empty contour RBF optimize xlabel variance GPRegression set_ylim var GPRegression kernel_call append log_likelihood optimize plot GPRegression optimize plot GPRegression optimize plot GPRegression RBF optimize exp multivariate_normal plot Poisson GP K zeros Laplace RBF exp optimize randn sort reshape rand cos hstack dot RBF_inv plot_ARD GPRegression sin log Linear SparseGPRegression RBF exp optimize randn ones sort rand cos hstack dot shape RBF_inv plot_ARD sin log Linear str plot print axis title GPRegression legend sum print GPRegression optimize SparseGPRegression RBF optimize plot randn uniform sin checkgrad seed RBF SparseGPRegression optimize plot randn print uniform nan sin binomial checkgrad SparseGPRegression RBF optimize subplots set_title plot randn ones print draw uniform sin RBF optimize plot randn reshape cos Gaussian GP sin Mapping RBF optimize plot randn reshape cos Gaussian GP sin Mapping Linear show normal RBF optimize_restarts plot print random pi plot_warping constrain_fixed GPRegression sin TanhFunction array WarpedGP optimize randn print StateSpace sde_Matern32 GPRegression sin Matern32 InferenceX optimize T backsub_both_sides flatten dot eye eye T dtrtrs diag pi ravel flatten trace sum log Y_local kern float64 Z Y KL_divergence update_gradients_full update_gradients_KL update_gradients_expectations set_X_gradients Allreduce copy inference_minibatch empty inference_likelihood isinstance gradients_qX_expectations likelihood array X Y_local kern inference_likelihood float64 update_gradients_diag Allreduce copy likelihood inference_minibatch Z Y update_gradients update_gradients_full empty X exp where erf shape erfcx zeros log asarray hstack func zeros kron range eye iv list exp arange any bb meshgrid zeros sum range factorial T variance tdot mean dot sum T variance square mean dot _psi2computations T reshape square dot shape empty swapaxes inner sum _psi1computations empty _psi2computations exp sum square einsum exp reshape sum square _psi2compDer _psi1compDer sum _psi1computations sum square einsum T reshape square dot swapaxes sum _psi2computations square binary_prob inner binary_prob sum param_to_array variance inline square mean binary_prob float empty log psicomputations param_to_array inline zeros float log ndarray isinstance isinstance update_gradients_diag Zp expm array T perm_matr diag gebal get_lapack_funcs copy dot eye range T dot shape balance_matrix zeros range append draw copy ARD dot ones_like clip randn append pop append pop append pop append pop PlotlyPlotsOffline MatplotlibPlots PlotlyPlotsOnline inject_plotting plot_inducing plot_latent_inducing plot plot_density plot_steepest_gradient_map plot_data_error plot_covariance plot_samples plot_magnification plot_ARD plot_latent_scatter plot_latent plot_confidence plot_mean plot_errorbars_trainset plot_optimizer plot_data plot_f _plot_data new_canvas data_1d get_which_data_ycols data_2d scatter get_free_dims update_not_existing_kwargs append get_which_data_rows get_x_y_var _plot_data_error new_canvas append get_which_data_ycols yerrorbar flatten sqrt get_free_dims update_not_existing_kwargs xerrorbar get_which_data_rows get_x_y_var _plot_inducing new_canvas inducing_1d inducing_2d plot_axis_lines scatter inducing_3d update_not_existing_kwargs get_free_dims get_most_significant_input_dimensions values new_canvas _plot_errorbars_trainset append get_which_data_ycols copy yerrorbar vstack get_free_dims update_not_existing_kwargs helper_predict_with_model get_which_data_rows get_x_y_var get_which_data_ycols _plot_mean new_canvas helper_predict_with_model helper_for_plot_data meanplot_1d meanplot_2d meanplot_3d dict update_not_existing_kwargs get_which_data_ycols _plot_confidence new_canvas helper_predict_with_model helper_for_plot_data confidence_interval update_not_existing_kwargs append fill_between range _plot_samples get_which_data_ycols new_canvas helper_predict_with_model helper_for_plot_data samples_3d samples_1d update_not_existing_kwargs get_which_data_ycols _plot_density linspace new_canvas helper_predict_with_model helper_for_plot_data fill_gradient density update_not_existing_kwargs append range update _plot_samples _plot get_which_data_ycols _plot_data_error _plot_inducing new_canvas helper_predict_with_model _plot_data helper_for_plot_data update _plot_samples _plot_confidence _plot_density _plot_mean dict print new_canvas list plot add_to_canvas dict new_canvas figure fopt_trace append keys range len atleast_2d arange format nextMedium _effective_input_dim print name input_sensitivity traverse reset ard new_canvas update_not_existing_kwargs barplot append range meanplot_1d meanplot_2d format isinstance ones tolist meanplot_3d dict new_canvas K update_not_existing_kwargs helper_for_plot_data input eval deactivate new_canvas get_most_significant_input_dimensions nextMedium latent_scatter scatter_label_generator reset scatter update_not_existing_kwargs append subsample_X _plot_latent_scatter ones num_data _new_canvas get_x_y_var _plot_inducing inducing_2d _new_canvas update_not_existing_kwargs magnification update_not_existing_kwargs _plot_latent_scatter ones _wait_for_updates num_data add_to_canvas new_canvas _plot_magnification helper_for_plot_data latent update_not_existing_kwargs _plot_latent_scatter _plot_latent ones _wait_for_updates num_data add_to_canvas dict new_canvas helper_for_plot_data output_dim list gradient annotation update_not_existing_kwargs range update ones _plot_steepest_gradient_map _wait_for_updates num_data add_to_canvas dict new_canvas helper_for_plot_data config int astype Gaussian transf posterior_samples Identity predict_quantiles range predict len get_fixed_dims x_frame2D get_free_dims zeros x_frame1D str list format isinstance size cycle append zeros next int format print hstack scatter_label_generator choice shape append update issparse ndarray view Y_normalized values asanyarray arange input_dim atleast_1d list map isinstance add_subplot figure ax_default meanplot hstack flatten fill ax_default append reduce pairwise mask_or ones append ax_default ones_like _process_unit_info asarray convert_yunits add_collection float contiguous_regions pop extend PolyCollection convert_xunits masked_invalid autoscale_view zeros bool len remove errorbar flatten append ax_default get_yticklines set_visible enumerate get_xticklines set_visible enumerate set_xticks subplot min ylim removeUpperTicks removeRightTicks xlim xticks max range yticks set_yticks min set_xlim flatten shape set_xticks removeRightTicks max enumerate set_ylim removeUpperTicks T int jet reshape ones min add_subplot _calculateFigureSize axis sqrt empty shape imshow ceil float abs max range arange add_subplot _Xscale x_frame1D range predict setdiff1d plot slice set_xlim x_frame2D input_dim empty contour T _Xoffset figure array X set_ylim len list parts Polygon min add_subplot set_xlim add_collection vstack figure points append PatchCollection max range set_ylim len append shapeRecords search enumerate append shapeRecords bbox enumerate bbox_match plot string_match plot line POINT record print field string_match Writer points save append len set_ylim set_xlim bbox rvs plot pdf hist linspace xlim rvs T reshape ylim xlim contour data zeros_like _Xoffset copy _Xscale scatter Y Z X_batch subplot T asarray _ll_trace plot zip _get_param_names figure legend _param_trace array append_axes make_axes_locatable set_axis_off legend get_legend_handles_labels isinstance X copy append range arange clf iter legend set_xticklabels tight_layout sqrt isinstance draw extend cycle fill_between arange add_subplot clf max bar iter legend range plot set_xticklabels set_xlim tight_layout sqrt min draw extend cycle figure fill_between set_ylim modify float sleep randomize append hasattr extend getmembers format print __file__ iter_modules dirname load_module flatten_nested output_dim Kern_check_dK_dX randn print name fix is_positive_semi_definite Kern_check_d2K_dXdX Kern_check_d2Kdiag_dXdX Kern_check_dKdiag_dX input_dim randint checkgrad list format partial_df GradientChecker randomize print reshape partial_f min grep_param_names zip dparam_partial constraint range __name__ checkgrad enumerate astype copy _simulate_matern nan BayesianGPLVMMiniBatch GPRegression sin join mkdirs join format flatten_axis get_fignums draw close savefig figure zip savez_compressed getmembers format ndarray _flatten isinstance asarray range len assert_array_almost_equal _a seed update rcParamsDefault seed update rcParamsDefault seed update _image_comparison rcParamsDefault seed update normal SparseGPRegression plot_inducing rcParamsDefault plot_data_error cos _image_comparison shape uniform plot_mean sin plot_data seed update normal SparseGPRegression plot_inducing rcParamsDefault plot_samples cos close _image_comparison shape uniform plot_mean sin plot_data seed update normal SparseGPRegression subplots rcParamsDefault plot_data_error cos _image_comparison shape uniform sin plot_data seed update normal subplots GPClassification plot rcParamsDefault cos _image_comparison shape uniform sin plot_errorbars_trainset seed update normal SparseGPClassification plot rcParamsDefault cos _image_comparison shape uniform sin seed join update load plot_scatter rcParamsDefault initialize_parameter plot_steepest_gradient_map _image_comparison plot_magnification plot_latent update_model seed join update load plot_inducing plot_scatter rcParamsDefault initialize_parameter plot_steepest_gradient_map _image_comparison plot_magnification plot_latent update_model sort asarray rand linspace randn sin_function sqrt generate_x_points len randn linear_function sqrt generate_x_points len randn sqrt zeros generate_x_points range generate_linear_data generate_sine_data empty range randn list empty sum range enumerate len empty sum enumerate len empty sum get_block_shapes enumerate vectorize sqrt int shape range zeros safe_root shape safe_root shape empty range range copy append hstack arange range ones print size float sum optimize vstack GPRegression zeros log_likelihood optimize ones GPOffsetRegression Gaussian set_prior vstack zeros log_likelihood enumerate T nanargmax get_log_likelihood_offset print hstack write extend delete isnan shape repeat unravel_index vstack append zeros range flush len write flush print lower set zip_longest join print rfind urlopen makedirs print str join join download_url zip_longest update str int join makedirs append range len join parse close download_data open binomial exp where download_data join genfromtxt download_data join T loadmat download_data join loadmat double writer deepcopy join int reader writerow loadtxt close reversed download_data any open download_data join read_csv download_data join read_csv join T asarray download_data read_csv download_data join read_csv join T download_data linspace meshgrid download_data join read_csv join hstack download_data vstack read_csv join read columns asarray replace quote print sort len to_csv loads sub append DataFrame read_csv makedirs join reshape close download_data open seed permutation oil join permutation print loadtxt extractall sort close download_data open genfromtxt join list sort download_data any nonzero unique zip zeros range len join var download_data sqrt mean loadmat load join reshape download_data shape array download_data join genfromtxt download_data join print loadtxt download_data join print loadtxt download_data join loadtxt extract join download_data namelist ZipFile load_text_data argsort reduce DataFrame exists values list map from_csv getsize sum to_pickle astype download_data population join remove set_index print loadtxt reshape write_status dict read_pickle array join columns index download_data read_csv join concatenate concat tolist astype groups apply download_data DataFrame read_csv to_series join T from_tuples flush format print name write index apply download_data read_csv compile len download_data join transpose loadmat download_data join transpose loadmat join loadmat array seed RBF multivariate_normal sort reshape White uniform K zeros seed toy_rbf_1d permutation sort seed normal load join T seed permutation reshape download_data extract join str asarray download_data flatten namelist append ZipFile range download_data loadtxt join join print extractall close download_data open download_rogers_girolami_data download_rogers_girolami_data download_rogers_girolami_data download_rogers_girolami_data download_rogers_girolami_data download_rogers_girolami_data download_data join genfromtxt zeros dataset vstack enumerate seed normal dot sqrt vstack append round array range join list print loadtxt extractall close extend copy download_data range open join rollaxis concatenate reshape extractall astype download_data zeros range open cmu_mocap cmu_mocap cmu_urls_files join load_channels acclaim_skeleton download_data eye tile append zeros range len str print isfinite logical_not id any str print id real eigvals as_strided squeeze func view detach normal var T min PCA add tdot uniform asfortranarray project print dpotrf ascontiguousarray mean any cholesky eye diag asfortranarray force_F_ordered force_F_ordered symmetrify diag sum jitchol log mdot dtrtri log jitchol dpotri sum diag symmetrify force_F_ordered print svd mean std T randn mean shape masked_invalid range zeros asfortranarray dsyrk symmetrify dsyr symmetrify _symmetrify_numpy _symmetrify_cython triu_indices_from dtrtrs zeros zeros swapaxes zeros_like view reshape sign shape real tile zeros clip clip clip clip clip all clip clip optimize ones linspace argmax T dot append sum array range warn radians cos dot eye sin array range len join T concatenate read_connections range uint loadtxt close NaN open range append len readline strip len open append zeros array range split cumsum len hstack vstack len MixedNoise len prod Coregionalize warn ICM len pop list ICM fix range NetpbmFile asarray NetpbmFile write COMM_WORLD rank Get_processor_name allgather int empty range join int optimize name now strftime save ceil float range bcast ones flatten vstack arange feval size delete flatten dot getSubs swapaxes zeros sum array quadvgk exp exp defaultdict count clip sqrt log pi sign logPdfNormal exp

# Censored Gaussian Processes --------------------------------------------------------------- This repository is the official implementation of the CGP, from *Estimating latent demand of shared mobility through censored Gaussian Processes*. The full paper is available here: [link1](https://arxiv.org/abs/2001.07402), [link2](https://www.sciencedirect.com/science/article/pii/S0968090X20306859?via%3Dihub) | | | |:-------------------------:|:-------------------------:| |<img width="1604" alt="Data" src="images/synthetic_1.png"> | <img width="1604" alt="NCGP" src="images/synthetic_2.png">| |<img width="1604" alt="NCGP-A" src="images/synthetic_3.png"> | <img width="1604" alt="CGP" src="images/synthetic_4.png">| The implementation is based on [GPy](https://github.com/SheffieldML/GPy) ## Summary

DanieleGammelli/multi-output-gp-censored-regression

['stochastic optimization']

['Generalized Multi-Output Gaussian Process Censored Regression']

models.py distributions.py likelihoods.py PyroCensoredNegBinomial PyroCensoredPoison PyroNegBinomial PyroCensoredNormal Poisson CensoredNegBinomial CensoredHomoscedGaussian CensoredPoisson Gaussian CensoredHeteroscedGaussian NegBinomial HeteroscedVariationalGP VariationalGP real nonnegative_integer nonnegative_integer nonnegative_integer

# Generalized Multi-Output Gaussian Process Censored Regression --------------------------------------------------------------- This repository is the official implementation of the HMOCGP, from *Generalized Multi-Output Gaussian Process Censored Regression*. The full paper is available [here](https://arxiv.org/abs/2009.04822) | | | |:-------------------------:|:-------------------------:| |<img width="1604" alt="NCGP" src="images/ncgp.PNG"> | <img width="1604" alt="HMOCGP_1" src="images/hmocgp1.PNG">| |<img width="1604" alt="CGP" src="images/cgp.PNG"> | <img width="1604" alt="HMOCGP_2" src="images/hmocgp2.PNG">| ## Summary This repository contains:

Daniil-Selikhanovych/Shampoo_optimizer

['stochastic optimization']

['Shampoo: Preconditioned Stochastic Tensor Optimization']

matrix_square_root_power.py shampoo_optimizer.py matrix_square_root matrix_inverse_pth_root ShampooOptimizer GetParam while_loop reduce_sum sqrt cast int32 eye norm while_loop pow cast int32 eye callable

# Shampoo_optimizer Our implementation of Shampoo optimizer based on https://arxiv.org/pdf/1802.09568.pdf. It consists of different notebooks, which we used on our own computers or Google Colab. Use scripts matrix_square_root_power.py and shampoo_optimizer.py both for using our code. Method `apply_gradients` does one iteration of Shampoo optimization process.

DanyWind/fastai_bs_finder

['dota 2']

['An Empirical Model of Large-Batch Training']

bs_finder.py bs_find BSFinder mom3 lin_comb get_flatten_grad parameters list cat lin_comb train_dl int BSFinder batch_size fit ceil train_ds len

DanyWind/fastai_bs_finder

DaoD/ConstraintGraph4NSO

['sentence ordering']

['Neural Sentence Ordering Based on Constraint Graphs']

FirstPhase/prepare_data.py SecondPhase/Dataset.py SecondPhase/Evaluate.py SecondPhase/run.py FirstPhase/model.py SecondPhase/GINPointer.py SecondPhase/prepare_data.py SecondPhase/NeuralNetwork.py SecondPhase/Model.py set_seed evaluate evaluate_test PairProcessor MyTestDataset main train load_and_cache_examples main DataHandler Dataset evaluate_pmr evaluate_first_last_accuracy evaluate_tau evaluate_accuracy evaluate_lcs GIN4Ordering GraphIsomorphismNetworkLayer BertClassification GraphConvolutionNetwork TransformerBlock BertClassification2 GraphIsomorphismNetwork GraphConvolutionNetworkLayer NeuralNetwork train_model test_model set_seed seed manual_seed_all manual_seed model get_linear_schedule_with_warmup tuple clip_grad_norm_ zero_grad DataLoader DataParallel max_grad_norm output_dir save max str set_seed format mean save_pretrained num_train_epochs info trange per_gpu_train_batch_size enumerate int join n_gpu evaluate_during_training evaluate backward AdamW print makedirs tqdm parameters step len tuple DataLoader save tensor argmax max open writer str eval_batch_size max_seq_length data_dir per_gpu_eval_batch_size len get_labels PairProcessor TensorDataset convert_examples_to_features compute_metrics append range update format close eval window_size MyTestDataset info join n_gpu writerow tqdm get_test_examples numpy makedirs tuple DataLoader argmax max eval_batch_size per_gpu_eval_batch_size compute_metrics append update format eval info load_and_cache_examples join n_gpu makedirs tqdm numpy len load join str format max_seq_length data_dir get_labels PairProcessor window_size TensorDataset convert_examples_to_features save info tensor exists from_pretrained do_eval ArgumentParser device do_train output_dir save basicConfig set_seed device_count parse_args to update save_pretrained info join evaluate add_argument evaluate_test dict train load_and_cache_examples do_test makedirs str get_convert_write_sind write_test get_convert_write_roc get_convert_write data_dir write_test_sind DataHandler out_dir task_name write_test_roc sum range len append range len range lcs len append range kendall_tau len range len str load sum print parameters is_finetuning GIN4Ordering fit load load_model evaluate save_path GIN4Ordering

DaoD/ConstraintGraph4NSO

Darg-Iztech/gender-prediction-from-tweets

['gender prediction']

['Gender Prediction from Tweets: Improving Neural Representations with Hand-Crafted Features']

train.py visualizer.py modelDeleter.py eval.py parameters_es.py test.py parameters.py preprocess.py parameters_en.py main.py parameters_ar.py model.py test network flags flags flags flags readCaptions prepCharBatchData prepWordBatchData_tweet readData prepWordBatchData readGloveEmbeddings partite_dataset_vectors user2target char2id partite_dataset readCharEmbeddings prepVectorBatchData prepTestData word2id readVectors partite_dataset_tweet Saver list model_tuple randn dict LineSentence array append zeros keys range values len ord model_tuple randn dict LineSentence append zeros range len lang join list findall TweetTokenizer text len shuffle tokenize getroot open keys append split lang join list str TweetTokenizer len shuffle tokenize array open keys append split join close index open append float listdir split training_set_size int list shuffle zip len list batch_size user2target shuffle zip len user2target append max range len append append lower range batch_size append lower list batch_size user2target shuffle zip append max range len list tweet_per_user batch_size user2target tolist shuffle zip append max range len list tweet_per_user batch_size user2target tolist extend shuffle sequence_length zip append range len int str validation_set_size print len training_set_size training_set_size int list str validation_set_size print shuffle zip len

# RNN and Captioning for Gender Classification Gender Classification From Tweets and Posted Images ## Requirements - Python 2.7 (it may work on python3 as well, not guaranteed) - (Preferable) CUDA 9.0 and Nvidia GPU compatible with CUDA - Word embeddings, we prefer GLoVe (https://nlp.stanford.edu/projects/glove/) - As a corpus to test, we used PAN2018 Author Profiling dataset (https://pan.webis.de/clef18/pan18-web/author-profiling.html) #### _Packages_: * Tensorflow 1.5 or higher * Gensim

Darthholi/DocumentConcepts

['table detection']

['Table understanding in structured documents']

concepts_fixed.py generators.py concepts.py run_experiments.py concepts_rendered.py attention.py distributions.py utils.py boxgeometry.py concepts_test.py GatherFromIndices SinCosPositionalEmbedding AttentionTransformer BoundingBox get_items_reading_layout range_distance_ordered analyze_bboxes_positions range_overlap_size filter_seen_boxes range_overlap_percent_min produce_ordered_and_overlapping_chunks search_lines_in_bboxes range_distance_side project_fun_bbox range_overlap_percent ConceptsPageGen InputBoxRuleScorable absolute_to_relative ConceptRulesDistribution ConceptRuleBaseScorable ConceptRuleDefinition clip_move_bbox_bounds all_possibly_same ConceptItem positions_sizes_to_ltrb Concept DistributionOfDistributions relative_to_absolute small_testing_setting_fixdim_class random_testing_setting random_testing_setting_distances constant_testing_setting_more_inboxes constant_testing_setting run_keras_fixed_experiment_binary_bigger constant_testing_setting_2pg constant_testing_setting_multiclass run_keras_fixed_experiment_categorical deep_eps_compare fixed_experiment_binary constant_wrong_testing_setting_2pg small_testing_setting realistic_setting run_keras_articlemodel run_keras_all2all_model run_keras_rendered_experiment_binary run_keras_rendered_experiment_categorical test_concepts_page_gen test_constant_models_categorical test_dist_of_dists_nonfix test_rendered_concepts_tf_gen test_concepts_tf_gen test_constant_models_binary test_struct_indexer test_scoring_samples test_jensen_shannon test_dist_of_dists test_rendered_models single_number_param_s conditioned_continuous is_np_array StochasticScorable FixdimDistribution _squeeze_last gaussian_smoothed_discrete MultiDistribution tuple_param_s is_single_float single_input_as_bool_param_s jensen_snannon_divergence_monte_carlo is_iterable smoothed_js_distances Determined StochasticScorableWrapper clipnorm is_single_integer is_single_number jensen_snannon_distance_monte_carlo DistributionAsRadial np_samesize PagesPacker FixedNeighboursAllPacker FixedNeighboursPacker RenderedConceptsPacker paint_2ddistribution paint_distribution fixed_known_borders_all_boxes_noshuffle model_sees_all_to_all realistic_experiment_articlemodel_local deepdictify baseline_rendered lrtb_center fixed_known_borders sample_concepts_example fixed_known_borders_all_boxes_shuffle fixed_known_borders_bigger_all_boxes_noshuffle articlemodel fixed_known_borders_bigger draw_texted_bbox noop draw_lrtg_bbox tempmap make_product_matrix EvaluateFCallback equal_ifarray GlobalMaxPooling1DFrom4D evaluate expand_batch iterate_predictions_batch array_all_classification_metrics StructIndexer collect_eval_data tf_dataset_as_iterator np_as_tmp_map project_fun_bbox all sort analyze_bboxes_positions key_f append max min sorted min key_f max enumerate pop sort min add append union enumerate append produce_ordered_and_overlapping_chunks search_lines_in_bboxes list list copy Counter InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition ConceptsPageGen list is_iterable isinstance is_np_array zip keys get_final_tf_data_dataset EarlyStopping fit_generator FixedNeighboursAllPacker Model tf_dataset_as_iterator summary FixedNeighboursPacker Input compile get_final_tf_data_dataset EarlyStopping fit_generator FixedNeighboursAllPacker Model tf_dataset_as_iterator summary FixedNeighboursPacker Input compile int get_final_tf_data_dataset EarlyStopping fit_generator Model tf_dataset_as_iterator summary FixedNeighboursPacker Input compile ConceptsPageGen InputBoxRuleScorable ConceptRuleDefinition StochasticScorableWrapper append randint range get_final_tf_data_dataset EarlyStopping fit_generator Model tf_dataset_as_iterator summary RenderedConceptsPacker use_neighbours Input compile get_final_tf_data_dataset EarlyStopping fit_generator Model tf_dataset_as_iterator summary RenderedConceptsPacker use_neighbours Input compile get_final_tf_data_dataset int format on_epoch_end print Input EarlyStopping range fit_generator Model summary tf_dataset_as_iterator RenderedConceptsPacker use_neighbours next zeros compile get_final_tf_data_dataset on_epoch_end print EarlyStopping fit_generator Model tf_dataset_as_iterator summary RenderedConceptsPacker use_neighbours Input compile pack_from unpack_from from_example small_testing_setting_fixdim_class draw_objects small_testing_setting constant_testing_setting get_final_tf_data_dataset constant_testing_setting dataflow_packer reset_state make_one_shot_iterator get_data get_next FixedNeighboursPacker tf_data_dataset_batcher_from_generator fixed_experiment_binary run_keras_fixed_experiment_categorical run_keras_rendered_experiment_categorical get_final_tf_data_dataset constant_testing_setting dataflow_packer reset_state make_one_shot_iterator get_data get_next RenderedConceptsPacker tf_data_dataset_batcher_from_generator draw_samples score_samples concept_scorable generated_to_objects uniform DistributionOfDistributions draw_samples score_samples generated_to_objects uniform DistributionOfDistributions draw_samples score_samples rvs log2 mean pdf gaussian_smoothed_discrete isinstance isinstance is_single_number isinstance sum is_iterable isinstance is_single_number all enumerate len list isinstance copy dict Rectangle annotate lrtb_center add_patch draw_lrtg_bbox in_concepts format plot print add_subplot lrtb_center ylim savefig figure realistic_setting xlim range draw_texted_bbox bbox rvs plot print rv_discrete add_subplot conditioned_continuous pdf ylim hist savefig figure linspace xlim MultiDistribution plot add_subplot savefig figure draw_samples DistributionAsRadial print fixed_experiment_binary realistic_setting print realistic_setting run_keras_fixed_experiment_binary_bigger print fixed_experiment_binary realistic_setting print realistic_setting run_keras_fixed_experiment_binary_bigger print realistic_setting run_keras_fixed_experiment_binary_bigger realistic_setting run_keras_rendered_experiment_binary run_keras_articlemodel print realistic_setting print run_keras_all2all_model realistic_setting run_keras_articlemodel print realistic_setting tempmap ndarray isinstance sum str format int append print classification_report confusion_matrix flatten shape round precision_recall_fscore_support zeros float argmax range roc_auc_score concat Assert tile expand_dims equal get_next make_one_shot_iterator run PrettyPrinter print index array_all_classification_metrics pprint collect_eval_data sum tempmap sum reshape iterate_predictions_batch nonzero zip array_represent next range append unpack_from range from_example len expand_batch predict_on_batch zip

#### Simulating invoice generation process This repository displays a process we went through to create a simulated business documents for fast experimentation. All experiments could be run as `python run_experiments.py command_name`, where command name is the name of the experiment. Simplest experiments are provided as tests in `concepts_test.py`. The code contains codes for model from our article https://arxiv.org/abs/1904.12577 together with generators code and some easier models (using tf.data.dataset and tensorpack dataflows) possibly enabling fast experimentation. ###### Installation Needs cuda and tensorflow installed (we recommend either a https://lambdalabs.com/lambda-stack-deep-learning-software or google collab), but possibly can run also on tensorflow cpu system. other requirements are written in `requirements.txt` ready for pip install.

Darthholi/similarity-models

['one shot learning', 'table detection']

['Table understanding in structured documents', 'Learning from similarity and information extraction from structured documents']

utils/sqlite_experiment_generator.py utils/__init__.py ds_local_sqlite.py utils/keras_utils_crop_render.py utils/manipulations_utils.py utils/textutils.py utils/k_utils.py experiments_reuse.py utils/dataflows.py experiments_ft.py utils/boxgeometry.py utils/attention.py experiments_ftreuse.py experiments_copy.py experiments_ftreuse_linear.py utils/evals.py utils/generic_utils.py doc_type sqlite_anonymize_texts sqlite_super_anonymize_texts sqlite_split_val_test add_embeddings_to_pages anonymize_memorize store_pic_data sqlite_anonymize_texts_as_dict clisap doc_info convert_array pic_data_for_page ask_memory_anonymize load_pdf streaming_dataset_from_file train_val_split dataset_to_sqlite anonymize ModelBase adapt_array range_by_char evaluating_copy CopyModel run_experiment_copy run_keras_rendered_experiment_cls_extract_types_only doc_inputs_siam_part doc_inputs_all_for_linear run_keras_rendered_experiment_cls_extract_types run_keras_rendered_experiment_cls_extract_types xpred_1 xcount_1 xacc run_keras_rendered_experiment_binary xacc_1 SinCosPositionalEmbedding AttentionTransformer BoundingBox area analyze_bboxes_positions filter_seen_boxes search_lines_in_bboxes sanitize_bboxes range_distance_side get_items_reading_layout get_doc_annotations overlaps_ids range_overlap_size produce_ordered_and_overlapping_chunks cut_percentage range_overlap_percent_min range_overlap project_fun_bbox sanitize_bbox overlaps range_distance_ordered produce_annotations_for_page range_overlap_percent RandomPhaseSequence BatchDataTflike RandomOrderSequence tf_dataset_as_iterator BatchDataPadder positive_samples_with_weights_mt evaluating_ftypes_targets evaluate_ftypes_legacy evaluating_ftypes_targets_reuse repair_annotations EvaluateFCallbackToLogs precision distances_and_classified_with_weights_v2 f1 binary_crossentropy_with_weights_mt eval_match_annotations accuracy_binary_multitarget array_all_classification_metrics num_predicted_ones_binary_multitarget collect_eval_data binary_crossentropy_with_weights_matrix matrix_weighted_target_loss evaluating_f_reuse GWME_sum recall_multitarget report_ftypes real_sampled_mean positive_samples_with_weights_predicted_mt positive_samples_with_weights lastaxissum GWME_f1 binary_accuracy_positive_with_weights_mt are_wordboxes_in_cls_extract_type binary_accuracy_with_weights_mt recall evaluating_ftypes_targets_separated distances_and_classified_with_weights_and_mean EvaluateFCallback T_prec binary_accuracy_with_weights num_truth_ones_binary_multitarget binary_accuracy_positive_with_weights distances_and_classified_with_weights T_rec T_f1 matrix_weighted_target_bce_tempsum positive_samples_with_weights_predicted tempmap PythonLiteralOption make_product_matrix EvaluateFCallback equal_ifarray GlobalMaxPooling1DFrom4D evaluate PythonLiteralOptionOrString expand_batch iterate_predictions_batch array_all_classification_metrics StructIndexer collect_eval_data tf_dataset_as_iterator np_as_tmp_map render_bboxes_pyfunc_2d render_nd_bboxes_tf_spreading keras_crop_and_resize_batches_of_arrays tf_crop_and_resize_batches_of_arrays load_our_model tile_for_product_matrix make_product_matrix tile_for_product_matrix_shape tile_to_match_shape GatherFromIndices tile_to_match load_weights_as_possible log_class_weights_from_counts_binary OpenTriMemmapGet is_np_array analyze_bboxes_positions care_weights_save_file range_distance_side ss_to_center bb_center produce_fov_ids dynamic_trimap is_single_float log_class_weights_from_counts fov_dilate class_weights_from_counts_binary is_iterable multiclass_temporal_class_weights np_pad_to_size hash_generator project_fun_bbox dynamic_memmap get_weights_save_file np_as_tmp_map tempmap OpenMemmap is_single_integer np_samesize model_timestamped_name equal_ifarray class_weights_from_counts range_distance_ordered hash_numpy is_single_number OpenTriMemmapSet run_dflow_epochs try_precompute_nearest DFSourceDataset try_generator run_dflow_final_epochs IndexObject count_wordboxes_total FtypesTrgtDocsTextsSqliteNearest convert_array try_find_nearest try_dflow_FtypesTrgtDocsTextsSqlite FtypesTrgtDocsTextsSqliteNearestSeparated dfobj_cache_logic try_dflow FtypesTrgtDocsTextsSqliteSeparated DocsTextsSqliteWeightedFtypesDebug FtypesTrgtDocsTextsSqlite count_wordboxes_per_cls_extract_type all_pairs measure_classcounts put_item DocsTextsSqliteWeightedFtypes put_item_batched DocsTextsSqliteBase IndexObjectNearest try_tf_dataset adapt_array DocsTextsSqliteNearest all_pairs_batched features_from_sentence text_onehot_chars features_from_text_len base_text_features remove_accents features_from_text text_tokens produce_drawings BytesIO seek save BytesIO seek load fetchall commit cursor close connect build tqdm load_pdf execute commit cursor close connect tqdm execute streaming_dataset_from_file commit remove cursor list close connect train_val_split tqdm page_count zip execute produce_annotations_for_page streaming_dataset_from_file range enumerate fetchall int commit cursor print close connect execute len fetchall join list cursor commit print close connect set tqdm execute split fetchall commit cursor print close connect execute islower isalpha isdigit isupper anonymize split fetchall join list cursor commit print close connect set tqdm anonymize_memorize execute repair_annotations eval_match_annotations str list defaultdict predict_on_batch append sum union next range GWME_sum format GWME_f1 set zip keys produce_drawings enumerate print tqdm dict T_f1 CopyModel get_index obj_gen pass_cls_extract_types get_final_dataflow_dataset Input Input keras_crop_and_resize_batches_of_arrays get_report_f ModelCheckpoint EvaluateFCallback build_model print stem EarlyStopping dfclass OrderedDict EvaluateFCallbackToLogs load_weights summary care_weights_save_file get_index get_final_dataflow_dataset get_report_f ModelCheckpoint EvaluateFCallback build_model print stem EarlyStopping OrderedDict EvaluateFCallbackToLogs load_weights summary care_weights_save_file get_index obj_gen get_final_dataflow_dataset get_report_f ModelCheckpoint EvaluateFCallback build_model print stem EarlyStopping dfclass OrderedDict EvaluateFCallbackToLogs load_weights summary care_weights_save_file get_index get_final_dataflow_dataset project_fun_bbox all sort analyze_bboxes_positions key_f append max min sorted min key_f max enumerate pop sort min add append union enumerate append produce_ordered_and_overlapping_chunks search_lines_in_bboxes append overlaps bbox enumerate array append sanitize_bbox get_items_reading_layout sanitize_bbox append sum full range bbox enumerate area min max get_next get_session make_one_shot_iterator run mean fn ndim w_fn greater switch sum binary_crossentropy min maximum sum max binary_crossentropy binary_crossentropy min maximum cast sum max equal sum cast equal T_rec T_prec items list defaultdict max values list defaultdict sorted predict_on_batch sum union next range format set zip are_wordboxes_in_cls_extract_type produce_drawings keys enumerate print tqdm dict T_f1 zeros sum list defaultdict format print predict_on_batch tqdm T_f1 zip next range enumerate append len set len set defaultdict enumerate repair_annotations eval_match_annotations str list defaultdict predict_on_batch append sum next range GWME_sum format GWME_f1 zip keys produce_drawings enumerate print tqdm dict T_f1 sum list defaultdict format print predict_on_batch union tqdm T_f1 zip next produce_drawings range enumerate array_all_classification_metrics sum print collect_eval_data tempmap sum list reshape predict_on_batch tqdm zip array_represent next range append enumerate sum str format int append print classification_report confusion_matrix flatten shape round precision_recall_fscore_support zeros float argmax range roc_auc_score evaluate_ftypes_legacy evaluating_ftypes_targets_reuse tempmap ndarray isinstance concat Assert tile expand_dims equal PrettyPrinter print index array_all_classification_metrics pprint collect_eval_data sum iterate_predictions_batch nonzero unpack_from range from_example len expand_batch predict_on_batch zip greater size equal Assert expand_dims tile expand_dims tile tile_for_product_matrix load_weights now expanduser format print dirname get_weights_save_file makedirs update encode update uint8 view ones len range enumerate is_single_number all enumerate len max tuple min full range enumerate len range float sum log len float sum range len len float sum log enumerate sum enumerate len bb_center intersect insert sort bb_center min len fov_dilate full max Index enumerate print precompute_embcache load_embcache format print dataflow_packer reset_state min tqdm get_data mean append median DocsTextsSqliteNearest max len range sum append range sum DocsTextsSqliteNearest precompute_embcache time format start_t subplots plot print reset_state tqdm get_data get_doc_pair title savefig last_print_t get_indices append n DocsTextsSqliteNearest len format start_t print dataflow_packer reset_state tqdm get_data last_print_t DocsTextsSqliteNearest n format start_t print dataflow_packer reset_state tqdm get_data FtypesTrgtDocsTextsSqliteNearest last_print_t n format asizeof start_t print dataflow_packer reset_state tqdm get_data last_print_t DocsTextsSqliteNearest range n format asizeof start_t print tqdm last_print_t n DocsTextsSqliteNearest range get_final_dataflow_dataset get_final_tf_data_dataset format start_t batch_size print tqdm last_print_t tf_dataset_as_iterator DocsTextsSqliteNearest n fetchall time cursor print close connect execute fetchall cursor print close connect fetchall defaultdict cursor format print min close connect tqdm mean loads execute median sum max append len encode normalize zeros lower enumerate text_histogram min extend range len remove_accents lower sub split ensure_pad sum isinstance float base_text_features replace len set_aspect join draw_rectangles format subplots axis close savefig zip append enumerate

# Similarity models Similarity and data extraction Source codes to accompany the following publications: - https://ieeexplore.ieee.org/document/8892877 (older manuscript: https://arxiv.org/abs/1904.12577) - https://rdcu.be/cmoAk (or manuscript here: https://arxiv.org/abs/2011.07964) Dataset is hosted under kaggle datasets: - https://www.kaggle.com/martholi/anonymized-invoices - Note that there can never be any pictures or text content, as the data must be anonymized! All codes and datasets are published under the LICENSE attached (GNU AFFERO GENERAL PUBLIC LICENSE). Any derivative work or service should be published under the same license (for any other licensing options, feel free to reach out).

DataMining-ClusteringAnalysis/CRAD-Clustering

['time series', 'time series clustering']

['CRAD: Clustering with Robust Autocuts and Depth']

CreateDistanceMatrix.py ExtensionToDBSCAN.py CRAD.py clustering_ _hist_find_new cal_adjM_cutOff dfs dbscan_newM _expand_cluster _region_query_cutOff arange tolist histogram append float array range len _hist_find_new remove extend append array range range len range dfs _hist_find_new remove DataFrame array append len range _region_query_cutOff range _expand_cluster

## CRAD-Clustering Python implementation of CRAD clustering algorithm (CRAD.py) and extended DBSCAN algorithm using CRAD framework (ExtensionToDBSCAN.py). ## Setup `python setup.py install` ## Documentation For CRAD-Clustering: Call the function `cal_adjM_cutOff(xxDist, StepSize, Nbin)` to calculate adjancey matrix where Inputs: xxDist - A distance matrix using robust mahalanobis distance.

Daulbaev/IRDM

['density estimation']

['Interpolation Technique to Speed Up Gradients Propagation in Neural ODEs']

experiments/density/vae_lib/utils/plotting.py experiments/density/vae_lib/models/flows.py interpolated_torchdiffeq/_impl/rk_common.py experiments/density/vae_lib/utils/load_data.py interpolated_torchdiffeq/_impl/odeint_interpolated.py interpolated_torchdiffeq/_impl/spline.py experiments/density/vae_lib/utils/visual_evaluation.py interpolated_torchdiffeq/_impl/odeint.py experiments/density/diagnostics/plot_compare_multiscale.py experiments/density/datasets/power.py experiments/density/train_misc.py experiments/density/diagnostics/plot_bottleneck_losses.py experiments/density/diagnostics/viz_toy.py experiments/density/vae_lib/utils/distributions.py experiments/classification/anode/__init__.py experiments/classification/scripts/save_best_checkpt.py experiments/density/datasets/miniboone.py experiments/classification/anode/scheme.py experiments/classification/models/sqnxt.py experiments/density/datasets/gas.py interpolated_torchdiffeq/_impl/adjoint.py experiments/classification/train_cifar.py experiments/density/vae_lib/optimization/loss.py interpolated_torchdiffeq/_impl/fixed_adams.py interpolated_torchdiffeq/_impl/tsit5.py experiments/classification/models/resnet.py interpolated_torchdiffeq/__init__.py interpolated_torchdiffeq/_impl/misc_old.py experiments/density/train_toy_new.py interpolated_torchdiffeq/_impl/bosh3.py experiments/density/vae_lib/models/layers.py experiments/classification/anode/utils.py interpolated_torchdiffeq/_impl/dopri5_old.py interpolated_torchdiffeq/_impl/solvers.py experiments/density/diagnostics/viz_high_fidelity_toy.py experiments/classification/anode/odesolver.py interpolated_torchdiffeq/_impl/dopri5.py experiments/density/diagnostics/plot_nfe_vs_dim_vae.py experiments/density/train_tabular.py experiments/density/datasets/hepmass.py experiments/density/diagnostics/approx_error_1d.py experiments/density/vae_lib/utils/log_likelihood.py interpolated_torchdiffeq/_impl/misc.py experiments/density/datasets/__init__.py experiments/density/vae_lib/models/VAE.py experiments/density/vae_lib/optimization/training.py experiments/classification/anode/adjoint.py experiments/density/diagnostics/viz_multiscale.py experiments/density/datasets/bsds300.py interpolated_torchdiffeq/_impl/adams.py experiments/density/train_vae_flow.py experiments/density/vae_lib/models/CNFVAE.py experiments/classification/models/preresnet.py experiments/density/diagnostics/plot_sn_losses.py experiments/density/diagnostics/plot_flows.py interpolated_torchdiffeq/_impl/cheb.py setup.py interpolated_torchdiffeq/_impl/adaptive_heun.py interpolated_torchdiffeq/_impl/fixed_grid.py interpolated_torchdiffeq/_impl/interp.py interpolated_torchdiffeq/_impl/__init__.py experiments/density/diagnostics/scrap_log.py experiments/density/diagnostics/viz_cnf.py experiments/classification/anode/time_stepper.py conv_init test ODEBlock get_activation get_param_normalization Identity train get_normalization tol_scheduler flatten_params_grad flatten_params odesolver_adjoint Checkpointing_Adjoint odesolver RK2 Euler RK4 Time_Stepper get_logger makedirs PreResNet6 PreBasicBlock PreResNet PreResNet10 PreResNet34 PreBasicBlock2 PreResNet4 PreResNet18 lr_schedule ResNet ResNet18 ResNet34 ResNet6 BasicBlock2 ResNet4 ResNet10 BasicBlock lr_schedule SqNxt_23_1x BasicBlock2 BasicBlock SqueezeNext lr_schedule get_configs get_normalizations get_best_acc add_spectral_norm standard_normal_logprob build_model_tabular count_parameters append_regularization_to_log count_nfe spectral_norm_power_iteration get_regularization set_cnf_options count_total_time override_divergence_fn create_regularization_fns batch_iter set_forward_t get_f_t set_nbe get_forward_t get_z_t restore_model get_nbe get_backward_t set_backward_t load_data compute_loss get_dt set_dt update_lr get_transforms compute_loss run BSDS300 GAS get_correlation_numbers load_data load_data_and_clean load_data_and_clean_and_split load_data_no_discrete_normalised_as_array HEPMASS load_data_no_discrete load_data_no_discrete_normalised load_data load_data_normalised load_data MINIBOONE POWER load_data_split_with_noise load_data load_data_normalised normal_log_density evaluate model_sample model_density data_sample compute_loss train data_density get_losses get_values construct_discrete_model get_transforms get_values plot_pairplot _line_to_dict log_to_csv add_noise add_spectral_norm build_model get_dataset save_im standard_normal_logprob save_density_traj trajectory_to_video get_ckpt_model_and_data makedirs add_noise create_model save_imgs_figure get_dataset FactorOut standard_normal_logprob trajectory_to_video save_trajectory get_ckpt_model_and_data makedirs AmortizedBiasODEnet AmortizedLowRankODEnet AmortizedBiasCNFVAE AmortizedCNFVAE construct_amortized_odefunc get_hidden_dims CNFVAE LyperODEnet LypernetCNFVAE AmortizedLowRankCNFVAE concat_layer_num_params HyperODEnet HypernetCNFVAE Sylvester TriangularSylvester Planar IAF GatedConvTranspose2d MaskedConv2d MaskedLinear Identity GatedConv2d TriangularSylvesterVAE HouseholderSylvesterVAE IAFVAE VAE PlanarVAE OrthogonalSylvesterVAE multinomial_loss_function binary_loss_array calculate_loss_array nll_loss calculate_loss multinomial_loss_array binary_loss_function cross_entropy train evaluate log_bernoulli log_normal_standard log_normal_diag log_normal_normalized load_omniglot load_caltech101silhouettes load_static_mnist load_dataset load_freyfaces calculate_likelihood plot_training_curve plot_reconstructions plot_images _VCABMState compute_implicit_phi VariableCoefficientAdamsBashforth g_and_explicit_phi _abs_square AdaptiveHeunSolver _ta_append _interp_fit_adaptive_heun OdeintAdjointMethod odeint_adjoint Bosh3Solver _interp_fit_bosh3 cheb1_interp_torch r8vec_cheby1space r8vec_cheby2space compute_barycentric_weights _abs_square _interp_fit_dopri5 Dopri5Solver _ta_append _abs_square _interp_fit_dopri5 Dopri5Solver _ta_append AdamsBashforth AdamsBashforthMoulton RK4 Euler Midpoint _interp_evaluate _interp_fit _scaled_dot_product _select_initial_step _check_inputs _flatten _dot_product _assert_increasing _is_iterable _compute_error_ratio _handle_unused_kwargs _is_finite _decreasing _norm _possibly_nonzero _convert_to_tensor _has_converged _flatten_convert_none_to_zeros _optimal_step_size _scaled_dot_product _select_initial_step _check_inputs _flatten _dot_product _assert_increasing _is_iterable _compute_error_ratio _handle_unused_kwargs _is_finite _decreasing _norm _possibly_nonzero _convert_to_tensor _has_converged _flatten_convert_none_to_zeros _optimal_step_size odeint OdeintChebyshevMethod odeint_chebyshev_func odeint_linear_func OdeintLinearMethod rk4_alt_step_func rk4_step_func _RungeKuttaState _runge_kutta_step FixedGridODESolver AdaptiveStepsizeODESolver make_spline compute_linear_transformation compute_smooth_torch prepare_data univariate_evaluate_torch compute_smooth prepare_data_torch make_spline_torch make_spline_using_transformation _interp_coeff_tsit5 Tsit5Solver _abs_square _interp_eval_tsit5 _optimal_step_size items list bias xavier_uniform_ weight __name__ constant_ join time format criterion backward zero_grad SGD wandb_name log parameters save append step max net enumerate list time criterion wandb_name log eval max net enumerate parameters flatten_params apply RK2 hasattr integrate base_func print squeeze dt_next z_t RK4 Euler append Dopri5SolverOld getLogger addHandler StreamHandler info DEBUG setLevel INFO FileHandler append strip format glob append int float startswith log pi apply apply AccNumEvals apply Accumulator apply find_cnf apply format enumerate item append iteritems eval tuple CNF isinstance tuple modules len batch_norm tuple map set_cnf_options zip append SequentialFlow split arange randperm cuda is_cuda split param_groups lr to sum model load load_state_dict deepcopy deepcopy deepcopy deepcopy inf_train_gen data batch_size AmortizedBiasCNFVAE IAFVAE kernel_size dims PlanarVAE plot_training_curve flow save AmortizedLowRankCNFVAE model_signature dataset HypernetCNFVAE cuda retrain_encoder OrthogonalSylvesterVAE log str list std VAE rank LypernetCNFVAE load_state_dict load_dataset append num_blocks range snap_dir format inf replace HouseholderSylvesterVAE Adamax hstack mean info model_path load join TriangularSylvesterVAE items num_householder num_ortho_vecs evaluate time wandb_name named_parameters made_h_size CNFVAE parameters isnan out_dir train epochs array makedirs read_pickle drop corr sum get_correlation_numbers mean any load_data std drop int as_matrix read_csv load_data drop mean std load_data_no_discrete int T Counter load_data_no_discrete_normalised append load int mean vstack load_data std int RandomState rand hstack shuffle delete load_data zeros load_data_split_with_noise sqrt concatenate randint randn normal_log_density standard_normal_logprob to zeros_like model standard_normal_logprob model randn data_sample model_density count_parameters count_nfe set_cnf_options compute_loss Adam to range update val avg count_total_time info item float niters RunningAverageMeter load join plot xlabel xscale ylabel tight_layout set_cnf_options savefig load_state_dict save figure to findall float append findall float append glow CouplingLayer depth append BruteForceLayer range strip sub split pairplot savefig read_csv set uniform_ MNIST CIFAR10 CelebA DataLoader SVHN recursive_apply_sn add_spectral_norm batch_norm autoencode tuple map MovingBatchNorm2d get_dataset spectral_norm load_state_dict append SequentialFlow split join view clone shape clamp_ dirname save div_ save_image makedirs eval linspace meshgrid to cat join format communicate Popen split load add_spectral_norm inf_train_gen format data checkpt print count_parameters spectral_norm set_cnf_options load_state_dict to create_regularization_fns LSUN add_spectral_norm ODENVP tuple map spectral_norm set_cnf_options split join list format save save_image eval linspace meshgrid to cat AmortizedBiasODEnet AmortizedLowRankODEnet ODEfunc get_hidden_dims LyperODEnet HyperODEnet sum log_normal_standard size reconstruction_function BCELoss log_normal_diag log_normal_standard view size sum long log_normal_diag cross_entropy size sum log_normal_standard log_normal_diag log_normal_standard view size sum long log_normal_diag cross_entropy view size NLLLoss2d dim multinomial_loss_function binary_loss_function input_size item prod log binary_loss_array multinomial_loss_array bernoulli view model min calculate_loss dynamic_binarization zeros cuda len divergence_fn input_size eval prod override_divergence_fn cuda log reciprocal clamp log astype shuffle from_numpy DataLoader TensorDataset zeros float seed reshape shuffle from_numpy DataLoader TensorDataset zeros float freyseed seed join reshape_data astype shuffle from_numpy DataLoader dynamic_binarization TensorDataset binomial zeros float loadmat join reshape_data astype shuffle from_numpy DataLoader TensorDataset float loadmat load_freyfaces load_omniglot load_caltech101silhouettes load_static_mnist model input_size unsqueeze log view logsumexp append prod range asarray format size info time calculate_loss_array reshape contiguous array len subplots arange max set_xlabel title savefig range update plot close ScalarMappable Normalize minimum set_size_inches min maximum subplots_adjust to_rgba set_ylabel brg set_ylim snap_dir str plot_images float makedirs update subplot set_aspect set_xticklabels set_yticklabels reshape axis close GridSpec imshow savefig figure swapaxes enumerate tuple deque to range append tuple min deque range append len tuple type_as append _flatten apply parameters is_tensor TupleFunc tuple type_as cos pi zeros float range cos pi zeros float range sum einsum pi sin zeros float range tuple type_as tuple dtype device _convert_to_tensor to range append len tuple tensor to type isnan iter is_tensor format __name__ warn tuple float fun to max tuple to _convert_to_tensor min max is_tensor _decreasing _check_inputs integrate hasattr base_func tuple dt_next func append _flatten apply parameters is_tensor TupleFunc _flatten apply parameters is_tensor TupleFunc dtype zip tuple alpha device append _convert_to_tensor beta tuple func tuple func list asarray reshape shape array view reshape shape tensor double ones_like T diags reshape size hstack ndim sqrt compute_smooth vstack spdiags spsolve array zeros diff vstack device view ones solve shape nelement to diags dim range cat compute_smooth_torch sqrt enumerate int diagflat t zeros reshape t nonzero item tensor range len zeros make_spline_torch range reshape einsum float tuple _interp_coeff_tsit5 type_as

# About Code for reproducing the experiments in the paper: > Daulbaev, T., Katrutsa, A., Markeeva, L., Gusak, J., Cichocki, A., & Oseledets, I. (2020). Interpolation Technique to Speed Up Gradients Propagation in Neural ODEs. Advances in Neural Information Processing Systems, 33. > [[arxiv]](https://arxiv.org/abs/2003.05271) [[bibtex]](https://scholar.googleusercontent.com/scholar.bib?q=info:pRV1HO0t_YYJ:scholar.google.com/&output=citation&scisdr=CgWvWYqaEPD_3T0aF_Y:AAGBfm0AAAAAX8wfD_Y4Ya2WUusJ2ZIm1BUcz2gdrE5S&scisig=AAGBfm0AAAAAX8wfD9hsDNUadyTkwE-UuiXNGTeD19jt&scisf=4&ct=citation&cd=-1&hl=ru) This code is based on the following repositories: * https://github.com/rtqichen/torchdiffeq * https://github.com/rtqichen/ffjord * https://github.com/amirgholami/anode * https://github.com/juliagusak/neural-ode-norm ### Installation

Davidham3/ASTGCN

['traffic prediction']

['Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting']

train.py test/test_data_preperation.py lib/data_preparation.py test/test_utils.py lib/utils.py test/test_metrics.py test/test_model.py model/mstgcn.py model/astgcn.py model/model_config.py lib/metrics.py MyInit normalization read_and_generate_dataset mean_absolute_error mean_squared_error masked_mape_np cheb_polynomial evaluate compute_val_loss get_sample_indices get_adjacency_matrix search_data predict scaled_Laplacian get_backbones MSTGCN_block cheb_conv MSTGCN_submodule MSTGCN test_normalize test_mae test_mape test_mse test_predict2 test_ASTGCN_submodule test_predict1 test_search_data5 test_search_data1 test_search_data2 test_get_adjacency_matrix2 test_scaled_Laplacian test_get_adjacency_matrix1 test_cheb_polynomial2 test_get_sample_indices2 test_search_data4 test_get_sample_indices1 test_search_data3 test_get_sample_indices3 test_cheb_polynomial1 Xavier Uniform mean normalize std int format print get_sample_indices shape normalization append range len append range search_data concatenate zeros sum diag real append range add_scalar print tolist extend loss_function sum net enumerate len concatenate print len append asnumpy enumerate print reshape mean_absolute_error masked_mape_np mean_squared_error predict add_scalar int read ConfigParser get_adjacency_matrix scaled_Laplacian normalization uniform uniform masked_mape_np mean_squared_error uniform uniform mean_absolute_error initialize ASTGCN_submodule net random_uniform initialize get_backbones ASTGCN random_uniform cpu net initialize get_backbones ASTGCN random_uniform cpu net search_data search_data search_data search_data search_data get_sample_indices uniform get_sample_indices uniform get_sample_indices uniform get_adjacency_matrix get_adjacency_matrix get_adjacency_matrix get_adjacency_matrix cheb_polynomial scaled_Laplacian get_adjacency_matrix cheb_polynomial scaled_Laplacian

# ASTGCN Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting (ASTGCN)  # References [Shengnan Guo, Youfang Lin, Ning Feng, Chao Song, Huaiyu Wan(*). Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting. The 33rd AAAI Conference on Artificial Intelligence (AAAI'19) 2019.](https://github.com/Davidham3/ASTGCN/blob/master/papers/2019%20AAAI_Attention%20Based%20Spatial-Temporal%20Graph%20Convolutional%20Networks%20for%20Traffic%20Flow%20Forecasting.pdf)

Davidham3/STSGCN

['traffic prediction']

['Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting']

models/stsgcn.py test/test_stsgcn.py load_params.py utils.py main.py training construct_model generate_data mask_np construct_adj masked_mse_np masked_mae_np generate_seq generate_from_train_val_test get_adjacency_matrix masked_mape_np generate_from_data sthgcn_layer_sharing output_layer huber_loss weighted_loss stsgcn sthgcn_layer_individual position_embedding stsgcl stsgcm gcn_operation test_stsgcn test_stsgcl test_position_embedding test_stsgcm test_weighted_loss test_huber_loss test_output_layer test_model_construction test_gcn_operation update time format print forward_backward extend dict masked_mae_np reset update_metric zip append label range asnumpy enumerate var format FullyConnected print Variable construct_adj BlockGrad shape Constant get_adjacency_matrix stsgcn Activation zeros zeros range len mean transformer generate_seq std mean transformer generate_seq std load generate_from_data generate_from_train_val_test keys isnan mask_np abs mask_np var format broadcast_add dot FullyConnected split append range gcn_operation len slice reshape transpose expand_dims swapaxes append position_embedding range stsgcm slice reshape concat stsgcm swapaxes append position_embedding range reshape Activation FullyConnected swapaxes MakeLoss abs square where arange broadcast_mul huber_loss expand_dims flip var format concat output_layer huber_loss append stsgcl range enumerate position_embedding gcn_operation stsgcm stsgcl output_layer huber_loss weighted_loss Constant stsgcn join listdir construct_model

# STSGCN AAAI 2020. Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting url: paper/AAAI2020-STSGCN.pdf # Usage Docker is recommended. 1. install docker 2. install nvidia-docker 3. build image using `cd docker && docker build -t stsgcn/mxnet_1.41_cu100 .` 4. download the data [STSGCN_data.tar.gz](https://pan.baidu.com/s/1ZPIiOM__r1TRlmY4YGlolw) with code: `p72z` 5. uncompress data file using `tar -zxvf data.tar.gz`

Davidnh8/artAI

['style transfer']

['A Neural Algorithm of Artistic Style']

Artist.py save_vgg19_weights_notop.py customvgg19_notop gramMatrix styleLossSingleLayer Artist contentLoss sum square dot transpose shape int gramMatrix Model Input load_weights

# AI Art This repo implements the art style transfer algorithm from "A Neural Algorithm of Artistic Style" (https://arxiv.org/abs/1508.06576) by Leon A with few modifications. Artist.py requires two base images, one for style and one for content. Then it extracts style and content respectively and merge the two to produce a hybrid image. Note: 1. At its core, it uses pre-trained VGG19 network. However, all local max pooling were replaced by local average pooling to achieve better image quality. 2. Uses fmin_l_bfgs_b to minimize the loss function. Gradient descent works as well, but it is slower and in my opinion, it achieves lower quality results. Below are some examples. Some images are my personal painting or a photograph taken by me. ## Example 1 Picasso (The Family) | | Van gogh (Vincent van Gogh) | | Van gogh with Picasso style ----------- | -- |------------ | -- | ------------

Davidzhangyuanhan/CelebA-Spoof

['face anti spoofing']

['CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations', 'CelebA-Spoof Challenge 2020 on Face Anti-Spoofing: Methods and Results']

intra_dataset_code/tsn_predict.py intra_dataset_code/models.py intra_dataset_code/main.py intra_dataset_code/client.py intra_dataset_code/detector.py get_image verify_output get_tpr_from_threshold read_image get_thresholdtable_from_fpr CelebASpoofDetector run_test conv3x3 BasicBlock AENet Bottleneck TSNPredictor pretrain shape imread cvtColor COLOR_BGR2RGB list format array info append keys read_image enumerate len int sort zip append float max len sort zip append float len format get_tpr_from_threshold info append get_thresholdtable_from_fpr time format detector_class verify_output info float predict enumerate Parameter data list items replace isinstance copy_ state_dict

# **CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations**  **CelebA-Spoof: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations** [Yuanhan Zhang](https://github.com/Davidzhangyuanhan/CelebA-Spoof), [Zhenfei Yin](https://github.com/yinzhenfei), [Yidong Li](http://faculty.bjtu.edu.cn/8408/), [Guojun Yin](https://gjyin91.github.io/), [Junjie Yan](https://yan-junjie.github.io/), [Jing Shao](https://amandajshao.github.io/) and [Ziwei Liu](https://liuziwei7.github.io/) In ECCV 2020. [[paper](https://arxiv.org/abs/2007.12342)] | [[video](https://www.youtube.com/watch?v=A7XjSg5srvI&t=4s)] > Abstract: **CelebA-Spoof** is a large-scale face anti-spoofing dataset that has **625,537** images from **10,177** subjects, which includes **43** rich attributes on face, illumination,environment and spoof types. Live image selected from the CelebA dataset. We collect and annotate spoof images of CelebA-Spoof. Among 43 rich attributes, 40 attributes belong to Live images including all facial components and accessories such as skin, nose, eyes, eyebrows, lip, hair, hat, eyeglass. 3 attributes belong to spoof images including spoof types, environments and illumination conditions.CelebA-Spoof can be used to **train and evaluate algorithms of face anti-spoofing**.  ## Updates [02/2021] The [technical report](https://arxiv.org/abs/2102.12642) of CelebA-Spoof Challenge 2020 is released on arXiv.

DePengW/PSENet

['optical character recognition', 'scene text detection', 'curved text detection']

['Shape Robust Text Detection with Progressive Scale Expansion Network', 'Shape Robust Text Detection with Progressive Scale Expansion Network']

util/statistic.py util/tf.py pypse.py util/event.py train_ic15.py metrics.py util/feature.py util/proc.py test_ctw1500.py util/rand.py util/t.py util/neighbour.py util/caffe_.py models/__init__.py util/dtype.py dataset/icdar2015_loader.py util/ml.py util/str_.py util/test.py util/log.py test_ic15.py pse/.ycm_extra_conf.py pse/__init__.py util/misc.py dataset/icdar2015_loader_Wdp.py eval/ic15/rrc_evaluation_funcs_v2.py pse/__main__.py util/__init__.py dataset/test_tools.py util/logger.py dataset/ctw1500_test_loader.py util/mask.py util/url.py eval/ic15/rrc_evaluation_funcs_v1.py util/cmd.py dataset/ctw1500_loader.py eval/ctw1500/eval_ctw1500.py eval/ctw1500/file_util.py util/thread_.py models/fpn_resnet.py util/np.py util/dec.py util/io_.py util/mod.py train_ctw1500.py dataset/ctw1500_loader_Wdp.py dataset/icdar2015_loader_Wdp_demo.py dataset/icdar2015_test_loader.py eval/ic15/rrc_evaluation_funcs.py eval/ic15/script.py dataset/__init__.py util/img.py util/plt.py eval/ic15/file_util.py runningScore pse debug test write_result_as_txt extend_3c polygon_from_points debug test write_result_as_txt extend_3c polygon_from_points cal_kernel_score dice_loss ohem_batch adjust_learning_rate save_checkpoint ohem_single main cal_text_score train cal_kernel_score dice_loss ohem_batch adjust_learning_rate save_checkpoint ohem_single main cal_text_score train get_img random_crop shrink get_bboxes random_rotate random_scale dist random_horizontal_flip scale CTW1500Loader perimeter get_img rotate_position random_horizontal_flip_points mirror_position random_crop crop_position shrink random_rotate_points CTW1500LoaderWdp get_bboxes random_scale random_rotate dist random_horizontal_flip scale random_crop_points perimeter circle get_img CTW1500TestLoader scale IC15Loader get_img random_crop shrink get_bboxes random_rotate random_scale dist random_horizontal_flip scale perimeter IC15LoaderWdp get_img rotate_position random_horizontal_flip_points mirror_position random_crop crop_position shrink random_rotate_points get_bboxes random_rotate random_scale dist random_horizontal_flip scale random_crop_points perimeter circle IC15LoaderWdp get_img rotate_position mirror_position random_crop crop_position shrink random_rotate_wdp get_bboxes random_rotate random_scale dist random_horizontal_flip_wdp random_crop_wdp random_horizontal_flip scale perimeter circle get_img IC15TestLoader scale get_union get_gt get_pred get_intersection write_file_not_cover write_file read_file read_dir write_file_not_cover write_file read_file read_dir validate_point_inside_bounds load_zip_file_keys validate_clockwise_points validate_lines_in_file decode_utf8 print_help main_validation get_tl_line_values load_zip_file get_tl_line_values_from_file_contents validate_tl_line main_evaluation validate_point_inside_bounds load_zip_file_keys validate_clockwise_points validate_lines_in_file decode_utf8 print_help main_validation get_tl_line_values load_zip_file get_tl_line_values_from_file_contents validate_tl_line main_evaluation validate_point_inside_bounds load_zip_file_keys validate_clockwise_points validate_lines_in_file decode_utf8 print_help main_validation get_tl_line_values load_zip_file get_tl_line_values_from_file_contents validate_tl_line main_evaluation evaluation_imports evaluate_method default_evaluation_params validate_data ResNet resnet50 Bottleneck resnet152 conv3x3 resnet34 resnet18 BasicBlock resnet101 GetCompilationInfoForFile IsHeaderFile MakeRelativePathsInFlagsAbsolute FlagsForFile DirectoryOfThisScript pse get_data get_params draw_log cmd print_calling_in_short_for_tf timeit print_calling print_test print_calling_in_short is_tuple int is_number is_str cast is_list double wait_key hog get_contour_min_area_box blur imwrite get_rect_iou black get_value put_text bgr2rgb get_roi render_points bgr2gray get_contour_region_iou resize convex_hull draw_contours get_contour_rect_box get_shape set_value is_in_contour is_valid_jpg move_win get_contour_region_in_rect fill_bbox imshow apply_mask random_color_3 imread bilateral_blur find_contours points_to_contours maximize_win rect_area rgb2bgr contour_to_points ds_size points_to_contour eq_color find_two_level_contours get_wh average_blur rgb2gray get_contour_region_in_min_area_rect rotate_point_by_90 white filter2D is_white translate get_contour_area rectangle min_area_rect rect_perimeter get_rect_points rotate_about_center gaussian_blur circle get_dir search is_dir dump_mat read_h5_attrs exists get_filename cd join_path load_mat get_file_size get_absolute_path cat dir_mat dump_json create_h5 dump pwd copy is_path mkdir ls open_h5 load remove write_lines read_h5 make_parent_dir find_files read_lines get_date_str init_logger plot_overlap savefig Logger LoggerMonitor find_black_components find_white_components init_params AverageMeter mkdir_p get_mean_and_std kmeans try_import_by_name add_ancester_dir_to_path import_by_name is_main get_mod_by_name add_to_path load_mod_from_path n2 _in_image count_neighbours get_neighbours n1 n1_count n8 n2_count n4 norm2_squared smooth flatten empty_list norm2 sum_all angle_with_x has_infty sin arcsin norm1 eu_dist iterable is_2D shuffle cos_dist chi_squared_dist clone is_empty has_nan has_nan_or_infty show plot_solver_data line show_images get_random_line_style to_ROI draw imshow rectangle hist maximize_figure set_subtitle save_image get_pid kill wait_for_pool get_pool cpu_count set_proc_name ps_aux_grep shuffle normal randint sample D E join index_of find_all is_none_or_empty ends_with remove_all remove_invisible to_lowercase starts_with is_str int_array_to_str contains to_uppercase split replace_all add_noise crop_into get_latest_ckpt get_init_fn gpu_config Print is_gpu_available min_area_rect focal_loss_layer_initializer get_variable_names_in_checkpoint sum_gradients get_all_ckpts get_update_op get_variables_to_train focal_loss get_iter get_available_gpus wait_for_checkpoint get_current_thread ThreadPool create_and_start ProcessPool is_alive get_current_thread_name download argv cit get_count exit sit connectedComponents get transpose copy put shape Queue zeros range len reshape concatenate imwrite concatenate print makedirs append range len join_path makedirs write_lines append range enumerate len int T empty model pse sign DataLoader resize resnet152 cuda max RETR_TREE list OrderedDict shape load_state_dict resnet101 append range format min_kernel_area resnet50 synchronize findContours astype debug copy write_result_as_txt mean eval img_paths resume scale flush enumerate load items time uint8 drawContours binary_th CHAIN_APPROX_SIMPLE print Variable reshape float32 sigmoid parameters CTW1500TestLoader isfile zeros tuple minAreaRect cmd boxPoints IC15TestLoader int sort min astype sum concatenate ohem_single append float numpy range sigmoid sum view mean update astype int32 get_scores numpy update astype int32 get_scores numpy model zero_grad runningScore cuda cal_kernel_score step append cal_text_score sum range update format size astype ohem_batch item float flush enumerate time criterion backward Variable print AverageMeter numpy len param_groups lr join save SGD DataLoader adjust_learning_rate resnet34 Logger save_checkpoint resnet152 cuda hasattr CTW1500LoaderWdp load_state_dict resnet101 resnet18 append module range pretrain resnet50 close lr resume checkpoint optimizer flush join load print parameters n_epoch train set_names makedirs IC15LoaderWdp imread int asarray remove_all append read_lines split range copy len warpAffine random getRotationMatrix2D range len max resize max min choice resize array min where randint max range len range PyclipperOffset int JT_ROUND area min append AddPath array perimeter ET_CLOSEDPOLYGON Execute len random range array PFT_EVENODD Execute PT_SUBJECT Pyclipper PT_CLIP random CT_INTERSECTION AddPaths append AddPath array range len radians cos sin PFT_EVENODD Execute PT_SUBJECT Pyclipper min PT_CLIP CT_INTERSECTION where array AddPaths append randint AddPath max range len format imwrite range len len random range array PFT_EVENODD Execute PT_SUBJECT Pyclipper PT_CLIP random CT_INTERSECTION AddPaths append AddPath array range len PFT_EVENODD Execute PT_SUBJECT Pyclipper min PT_CLIP CT_INTERSECTION where array AddPaths append randint AddPath max range len print append split append int asarray split area append sort walk replace read close open join makedirs close write open join makedirs close write open write exit group match namelist append ZipFile group match namelist append ZipFile decode BOM_UTF8 replace startswith encode validate_tl_line decode_utf8 replace split get_tl_line_values validate_point_inside_bounds int replace group match replace argsort append get_tl_line_values split update default_evaluation_params_fn validate_data_fn writestr list items write dumps close dict print_help evaluate_method_fn ZipFile makedirs update default_evaluation_params_fn validate_data_fn print exit dict validate_clockwise_points float load_zip_file validate_lines_in_file compute_ap area list decode_utf8 append polygon_from_points range import_module get_intersection_over_union load_zip_file empty get_pred get_tl_line_values_from_file_contents float items namedtuple int8 rectangle_to_polygon Rectangle get_intersection zeros len load_url ResNet load_state_dict load_url ResNet load_state_dict list ResNet load_url load_state_dict keys state_dict list ResNet load_url load_state_dict keys state_dict list ResNet load_url load_state_dict keys state_dict append join startswith IsHeaderFile compiler_flags_ exists compiler_flags_ GetCompilationInfoForFile compiler_working_dir_ MakeRelativePathsInFlagsAbsolute DirectoryOfThisScript cpse array net Solver isinstance append net Solver isinstance show int get_random_line_style plot print readlines smooth len contains eval get_absolute_path save_image plt legend append float open isinstance debug waitKey ord bgr2rgb get_absolute_path wait_key namedWindow isinstance destroyAllWindows move_win rgb2bgr WINDOW_NORMAL imread maximize_win get_absolute_path rgb2bgr make_parent_dir moveWindow setWindowProperty WND_PROP_FULLSCREEN enumerate get_shape get_shape min max drawContours boundingRect get_contour_rect_box minAreaRect BoxPoints int0 get_shape get_contour_rect_box warpAffine int BoxPoints transpose hstack getRotationMatrix2D dot get_roi minAreaRect points_to_contour black draw_contours shape to_contours draw_contours assert_equal asarray range GaussianBlur bilateralFilter putText int32 FONT_HERSHEY_SIMPLEX get_shape int tuple warpAffine get_wh float32 cos deg2rad getRotationMatrix2D dot sin abs array _get_area transpose _get_inter zeros range len findContours asarray copy findContours copy pointPolygonTest convexHull randint list asarray range zip minAreaRect empty points_to_contour get_absolute_path makedirs get_dir mkdir get_absolute_path get_dir mkdir get_absolute_path get_absolute_path get_absolute_path is_dir get_absolute_path expanduser startswith chdir get_absolute_path append listdir get_absolute_path ends_with get_absolute_path open get_absolute_path make_parent_dir get_absolute_path get_absolute_path get_absolute_path savemat make_parent_dir get_absolute_path getsize get_absolute_path get_absolute_path make_parent_dir get_absolute_path get_absolute_path get_absolute_path join_path extend ls is_dir get_absolute_path find_files append get_absolute_path make_parent_dir now setFormatter basicConfig print join_path addHandler make_parent_dir StreamHandler get_date_str Formatter setLevel asarray arange plot numbers enumerate len pop black set_root insert copy get_neighbours N4 shape get_new_root get_root append set_visited range is_visited print DataLoader div_ zeros range len normal constant isinstance kaiming_normal Conv2d bias modules BatchNorm2d weight Linear makedirs asarray warn flatten append enumerate insert join_path add_to_path get_dir __import__ import_by_name get_absolute_path get_filename append _in_image append _in_image append _in_image append _in_image norm2 zip shape asarray extend len pi asarray asarray reshape flatten sqrt shape range shape asarray has_infty has_nan enumerate len show asarray join_path flatten linspace figure save_image load show val_accuracies list plot training_losses val_losses training_accuracies figure legend range len Rectangle add_patch full_screen_toggle get_current_fig_manager add_line Line2D linspace show_images show set_title set_subtitle axis colorbar bgr2rgb imshow maximize_figure subplot2grid append save_image enumerate get_absolute_path savefig imsave make_parent_dir set_xlim set_ylim suptitle maximize_figure randint len Pool join close setproctitle print get_pid cmd append int cmd split flatten flatten append pop list tuple to_lowercase is_str enumerate list tuple to_lowercase is_str enumerate to_lowercase findall replace replace_all binomial list_local_devices get_checkpoint_state is_dir get_absolute_path model_checkpoint_path get_checkpoint_state all_model_checkpoint_paths int get_latest_ckpt is_none_or_empty latest_checkpoint print get_model_variables startswith info append extend get_collection TRAINABLE_VARIABLES get_latest_ckpt NewCheckpointReader get_variable_to_shape_map dtype set_shape py_func ConfigProto ones_like zeros_like sigmoid_cross_entropy_with_logits float32 where reduce_sum sigmoid pow cast stop_gradient name reduce_mean add_n histogram zip append scalar UPDATE_OPS get_collection start Thread setName print urlretrieve stat show_images get_count imwrite get_count imwrite asarray

# Shape Robust Text Detection with Progressive Scale Expansion Network ## Requirements * Python 2.7 * PyTorch v0.4.1+ * pyclipper * Polygon2 * OpenCV 3.4 (for c++ version pse) * opencv-python 3.4 ## Introduction Progressive Scale Expansion Network (PSENet) is a text detector which is able to well detect the arbitrary-shape text in natural scene.

DeanChan/NAE4PS

['person search', 'person re identification', 'human detection']

['Norm-Aware Embedding for Efficient Person Search']

configs/res50_faster_rcnn.py lib/datasets/cuhk_sysu.py configs/__init__.py lib/datasets/ps_dataset.py lib/utils/serialization.py lib/utils/distributed.py lib/datasets/__init__.py lib/datasets/prw.py lib/loss/__init__.py lib/utils/trainer.py scripts/train_NAE.py lib/utils/misc.py lib/loss/oim.py lib/model/faster_rcnn_pixel_wise_norm_aware.py lib/model/resnet_backbone.py lib/utils/group_by_aspect_ratio.py lib/utils/evaluator.py lib/utils/transforms.py lib/utils/debug_tools.py lib/utils/logger.py scripts/test_NAE.py lib/model/faster_rcnn_norm_aware.py args_faster_rcnn_norm_aware args_faster_rcnn args_faster_rcnn_oim CUHK_SYSU PRW PersonSearchDataset PrefetchDataLoader get_data_loader get_dataset collate_fn PrefetchGenerator OIM oimsmr oim OIMLossSMR OIMLoss OIMSMR norm_aware_rcnn_loss NormAwareEmbeddingProj get_norm_aware_model CoordRegressor FasterRCNN_NormAware NormAwareRoiHeads spatial_norm_aware_rcnn_loss PixelWiseNormAwareEmbeddingProj load_NAE_weights FasterRCNN_NormAware_PW get_pixel_wise_norm_aware_model PixelWiseNormAwareRoiHeads resnet_backbone BackboneWithFasterRCNN RCNNConvHead get_rcnn_fg_bg_ratio is_dist_avail_and_initialized setup_for_distributed tensor_gather init_distributed_mode get_world_size reduce_dict dist_print all_gather get_rank is_main_process inference detection_performance_calc _compute_iou create_aspect_ratio_groups _compute_aspect_ratios_voc_dataset _compute_aspect_ratios_custom_dataset _compute_aspect_ratios_slow _compute_aspect_ratios_coco_dataset _compute_aspect_ratios_subset_dataset GroupedBatchSampler compute_aspect_ratios _quantize SmoothedValue MetricLogger warmup_lr_scheduler ship_data_to_cuda lazy_arg_parse resume_from_checkpoint get_lr_scheduler ship_data_to_cuda_singe_sample lucky_bunny Nestedspace get_optimizer unpickle read_json save_checkpoint write_json pickle mkdir_if_missing get_trainer _flip_coco_person_keypoints get_transform Compose ToTensor RandomHorizontalFlip main main add_argument ArgumentParser add_argument ArgumentParser add_argument ArgumentParser use_flipped get_transform ds_cls create_aspect_ratio_groups PrefetchDataLoader batch_size DistributedSampler RandomSampler get_dataset distributed BatchSampler GroupedBatchSampler SequentialSampler reshape size squeeze smooth_l1_loss numel binary_cross_entropy_with_logits float cat NormAwareEmbeddingProj resnet_backbone train eval CoordRegressor FasterRCNN_NormAware exp reshape size squeeze smooth_l1_loss numel binary_cross_entropy_with_logits cat net_fn CoordRegressor focal num_pids PixelWiseNormAwareEmbeddingProj pool_module num_features alpha_d resnet_backbone oim_momentum load_NAE_weights OIMLossSMR smr eval smr_omega_decay gamma_d num_cq_size oim_scalar train load view print load_state_dict good requires_grad_ BackboneWithFasterRCNN RCNNConvHead dist_print int setup_for_distributed format init_process_group print set_device dist_url barrier device_count rank local_rank from_buffer dumps get_world_size loads zip append tensor to empty max cat list size len get_world_size all_gather append tensor empty max range cat enumerate get_world_size ship_data_to_cuda list zip model tqdm ex_feat OrderedDict eval device append numpy cat min max asarray format record print _compute_iou average_precision_score precision_recall_curve any zip append zeros argmax range list print DataLoader SubsetSampler range len list get_height_and_width append float range len list append float range len list size append float range len list range len hasattr isinstance Subset VOCDetection CocoDetection list deepcopy sorted print compute_aspect_ratios _quantize format parse_known_args error join _ to load print start_epoch resume load_state_dict info good items list requires_grad SGD lr MultiStepLR StepLR print str bold green makedirs pop items list ndarray isinstance copy dirname mkdir_if_missing save resume_from_checkpoint apex initialize half_function hasattr convert_syncbn_model Engine convert_sync_batchnorm distributed roi_heads DistributedDataParallel forward module append ToTensor RandomHorizontalFlip resume_from_checkpoint get_data_loader load_from_json device bold dataset get_model_fn values run seed from_dict list lightgreen search_performance_calc detection_performance_calc inference to manual_seed_all format info manual_seed Nestedspace join print path to_dict STARTED use_tfboard get_optimizer get_lr_scheduler strftime is_main_process SummaryWriter init_distributed_mode debug add_event_handler close distributed get_rcnn_fg_bg_ratio mkdir_if_missing gethostname get_trainer export_to_json

# Norm-Aware Embedding for Efficient Person Search This repository hosts our code for our paper [Norm-Aware Embedding for Efficient Person Search](http://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_Norm-Aware_Embedding_for_Efficient_Person_Search_CVPR_2020_paper.pdf). ## Preparation 1. Clone this repo ```bash https://github.com/DeanChan/NAE4PS.git && cd NAE4PS ``` 2. Build environment with [conda](https://docs.anaconda.com/anaconda/install/linux/) ```bash conda env create --prefix <your_conda_env_path> -f environment.yml

Deep-Imaging-Group/RED-WGAN

['denoising']

['Denoising of 3-D Magnetic Resonance Images Using a Residual Encoder-Decoder Wasserstein Generative Adversarial Network']

preprocessing.py model7.py data.py mixdata.py model2.py model5.py MRIDataset MRIValidDataset add_rice_noise MRIDataset MRIValidDataset add_rice_noise initialize_weights Net CNN3D initialize_weights WGAN GeneratorNet DiscriminatorNet VGG19 initialize_weights WGAN GeneratorNet DiscriminatorNet VGG19 merge_test_img add_rice_noise patch_test_img generate_patch generate_noised_mri normal shape max fill_ isinstance BatchNorm3d weight modules zero_ kaiming_normal_ load affine add_rice_noise header get_data Nifti1Image save listdir makedirs str print reshape transpose get_data shape save append listdir range makedirs reshape shape vstack append range int sqrt append range len

Deep-Imaging-Group/RED-WGAN

DeepLearnXMU/NSEG

['sentence ordering']

['Graph-based Neural Sentence Ordering']

data/data.py main.py model/seq2seq.py model/generator.py override parse_args set_seeds curtime MyBatch DocIter ParallelDataset NormalField DocField GraphField DocDataset Beam decode valid_model SGRU train beam_search_pointer print_params PointerNet GRNGOB GRUCell add_argument ArgumentParser seed manual_seed_all manual_seed __dict__ Beam unsqueeze candidates list squeeze tolist new_tensor encode append range byte size new_zeros set long beam_size sort extend difference index_select step print sum model zero_grad localtime maximum_steps PointerNet cuda vectors doc doc_len load_pretrained_emb strftime load_state_dict range e_words format DocIter model_path enumerate NLLLoss load time elocs Adadelta backward print order parameters print_params equip step DocDataset early_stop model accuracy_score max open list doc doc_len tolist map from_iterable beam_search_pointer intersection append range e_words format size close set mean eval zip equal enumerate beam_size join elocs print order writetrans len

Graph based Neural Sentence Ordering ===================================================================== ### Installation The following packages are needed: - Python == 3.6 - Pytorch >= 1.0 - torchtext == 0.3 - Stanford POS tagger or Dependency Parser - Glove (100 dim) ### Dataset Format

DeepPathology/SlideRunner

['whole slide images']

['SlideRunner - A Tool for Massive Cell Annotations in Whole Slide Images']

SlideRunner/plugins/proc_macenko.py SlideRunner/plugins/proc_countdown.py SlideRunner/plugins/proc_ppc.py SlideRunner/plugins/proc_HPF.py SlideRunner/gui/style.py SlideRunner/gui/dialogs/dbmanager.py SlideRunner/plugins/proc_ObjDetResults.py SlideRunner/__main__.py SlideRunner/plugins/proc_secondaryDatabase.py SlideRunner/gui/dialogs/getCoordinates.py SlideRunner/general/pluginFinder.py SlideRunner/general/types.py SlideRunner/SlideRunner.py SlideRunner/plugins/proc_coregistration.py SlideRunner/gui/frameSlider.py SlideRunner/gui/dialogs/settings.py SlideRunner/gui/dialogs/about.py SlideRunner/processing/thumbnail.py SlideRunner/gui/mouseEvents.py tests/test_database.py SlideRunner/processing/screening.py SlideRunner/general/__init__.py SlideRunner/gui/SlideRunner_ui.py SlideRunner/gui/menu.py SlideRunner/general/SlideRunnerPlugin.py SlideRunner/gui/dialogs/question.py tests/test_screening.py SlideRunner/__init__.py SlideRunner/gui/__init__.py SlideRunner/gui/dialogs/__init__.py SlideRunner/gui/dialogs/welcomeExact.py SlideRunner/gui/zoomSlider.py SlideRunner/dataAccess/__init__.py SlideRunner/dataAccess/exact.py SlideRunner/gui/splashScreen.py SlideRunner/gui/sidebar.py SlideRunner/gui/shortcuts.py main.py SlideRunner/gui/types.py tests/test_exact.py SlideRunner/gui/toolbar.py SlideRunner/dataAccess/annotations.py SlideRunner/gui/dialogs/exactDownloadDialog.py SlideRunner/gui/dialogs/exactLinkDialog.py SlideRunner/general/dependencies.py SlideRunner/plugins/proc_coregistration_qt.py SlideRunner/gui/annotation.py SlideRunner/processing/__init__.py setup.py SlideRunner/plugins/proc_otsu.py SlideRunner/plugins/proc_MitHeatmap.py SlideRunner/dataAccess/database.py SlideRunnerUI imageReceiverThread PluginStatusReceiver SlideImageReceiverThread main AccessViolationError ExactManager get_filename_from_cd ExactImageList get_hex_color ExactAPIs ExactProcessError list_to_exactvector numeric check_qt_dependencies check_all_dependencies check_version iter_namespace PluginTypes PluginOutputType PluginConfigUpdateEntry PluginConfigUpdate PluginConfigurationType PluginConfigurationEntry getVisibleAnnotations JobDescription TablePluginConfigurationEntry PluginActionEntry NonePlugin generateMinMaxCoordsList PluginConfigUpdateFloatSlider activePlugins FilePickerDialogType StatusInformation SliderPluginConfigurationEntry FilePickerConfigurationEntry jobToQueueTuple SlideRunnerPlugin pluginJob PluginConfigUpdateComboBox ComboboxPluginConfigurationEntry AnnotationUpdatePolicy PushbuttonPluginConfigurationEntry pluginEntry removeFromPolygon addSpotAnnotation copyAllAnnotations checkAnnotationParameters addCircleAnnotation addToPolygon deleteAllFromClassOnSlide renameClass deleteImageLabel addImageLabel copyAnnotation addAreaAnnotation deleteClass addPolygonAnnotation changeClassColor frameSlider defineAnnotationMenu definePluginMenu updateOpenRecentSlide defineOpenRecentDatabase updateOpenRecentDatabase defineOpenRecent defineMenu wheelEvent removeLastPolygonPoint doubleClick releaseImage pressImage getMouseEventPosition rightClickImage leftClickImage moveImage defineShortcuts defineMenuShortcuts ClassSelectTableWidget addSidebar Ui_MainWindow splashScreen setStyle defineToolbar WandAnnotation UIMainMode ClassRowItemId ClassRowItem zoomSlider closeDlg aboutDialog DatabaseManager ExactDownloadDialog ExactLinkDialog hitClose getCoordinatesDialog YesNoAbortDialog settingsDialog changePxSlider chooseFile saveAndClose disableAndClose closeDlg welcomeExactDialog enableAndClose Plugin WSI_Matcher MatcherParameters TissueDetector Plugin Plugin Plugin Plugin normalize mypercentile quantile Plugin Plugin Plugin Plugin Plugin screeningMap thumbnail test_database test_setup test_images cleanup test_pushannos test_screening ViewingProfile ndarray dict PluginConfigUpdate pyqtSignal Database show SlideRunnerUI exceptionHook exceptionHook_threading exec_ put raise_ finish setStyle jobToQueueTuple update int ViewingProfile COLORS_CLASSES len findall len range split __import__ __import__ list IMAGE_PLUGIN list HEATMAP NONE_PLUGIN NO_OVERLAY asarray tolist zeros range len messagebox int insertNewSpotAnnotation saveLastViewport showAnnotationsInOverview slideUID getMouseEventPosition writeDebug retrieveAnnotator showImage showAnnoclass showDBentryCount getZoomValue getText showDatabaseUIelements getClassByID getColor getClassByID rgb_to_hex isValid setClassColor showDatabaseUIelements showImage fromRgb No getClassByID Yes question showDatabaseUIelements insertNewImageAnnotation showImage showImage removeImageAnnotation uid No list Yes question removeAnnotation showImage append keys showDBentryCount No getAnnotationsOfLabel addAnnotationToDatabase Yes question showImage showDBentryCount len loadIntoMemory showImage addAnnotationToDatabase showDBentryCount saveLastViewport showAnnotationsInOverview slideUID annotationUID retrieveAnnotator writeDebug exchangePolygonCoordinates showImage insertNewPolygonAnnotation showDBentryCount uid Polygon float32 slideUID xy vertices exchangePolygonCoordinates showImage array uid Polygon float32 difference xy slideUID vertices exchangePolygonCoordinates showImage about int insertNewAreaAnnotation saveLastViewport showAnnotationsInOverview min square slideUID sqrt getMouseEventPosition retrieveAnnotator showDBentryCount showImage writeDebug max showAnnoclass getZoomValue int insertNewAreaAnnotation saveLastViewport showAnnotationsInOverview min slideUID getMouseEventPosition retrieveAnnotator showDBentryCount showImage writeDebug max showAnnoclass getZoomValue pyqtSignal linkSlideToExact copyscreenshot actionAdd_annotator setRotate syncWithExact setStatusBar actions definePluginMenu action_Quit action_CloseDB action_Open setText settingsDialog setChecked menubar setMPP actionManageDatabase actionOpen manageDB setExactUser downloadSlideFromExact actionSettings statusbar QMenu connect setEnabled exportToExact QStatusBar goToCoordinate menuAction zoomOut setTitle setGeometry value partial saveDBto QRect setShortcut removeAction defineOpenRecentDatabase findAnnoByID action_setMPP savescreenshot addAction zoomMaxoptical defineOpenRecent QMenuBar addSeparator closeDatabase action_Close setCheckable actionAbout QAction actionCreate_new actionAdd_cell_class actionOpen_custom setObjectName setMenuBar zoomIn shortName menubar list QMenu setEnabled halfOpacity menuAction progressBarQueue append setTitle fullOpacity partial plugin noOpacity addAction setCheckable addSeparator setObjectName togglePlugin list value removeAction reverse openrecentactions enumerate list value openrecentdbactions removeAction reverse enumerate list addMenu updateOpenRecentSlide setEnabled updateOpenRecentDatabase list addMenu setEnabled MODE_ANNOTATE_AREA partial MODE_VIEW setUIMode MODE_ANNOTATE_WAND MODE_ANNOTATE_SPOT setEnabled MODE_ANNOTATE_POLYGON MODE_ANNOTATE_FLAG addAction setCheckable setChecked MODE_ANNOTATE_CIRCLE addMenu pos hitEscape list exec_ partial removeLastPolygonPoint Key_Enter QMenu lastAnnotationClass setShortcut mapToGlobal addAction append addPolygonAnnotation getAllClasses annotationsList addMenu asarray showImage power setZoomValue updateScrollbars getZoomValue toQImage ArrowCursor x showImage getMouseEventPosition abs max anno_pt1 screenToSlide displayedImage append ClosedHandCursor setCursor updateScrollbars fromImage square copy sqrt int setPixmap keyboardModifiers min rectangle circle showDBEntry discoveryMode getMouseEventPosition retrieveAnnotator showImage discoverUnclassified uid list screenToSlide saveLastViewport QMenu lastAnnotationClass isOpen mapToGlobal writeDebug append ClosedHandCursor setCursor addMenu getZoomValue pos addSpotAnnotation partial WandAnnotation addAction positionInAnnotationHandle exec_ addAnnotationLabel keyboardModifiers checkIfAnnotatorLabeled getAllClasses findClickAnnotation dragPoint removeFromPolygon Key_Enter ArrowCursor RETR_LIST findIntersectingAnnotation currentVP getMouseEventPosition showImage addPolygonAnnotation list uid screenToSlide QMenu lastAnnotationClass mask addAreaAnnotation mapToGlobal append setCursor addMenu pos partial addCircleAnnotation findContours setShortcut WandAnnotation addAction updatePolygonPoint exec_ CHAIN_APPROX_SIMPLE addToPolygon polygonAnnotation polygon getAllClasses pop showImage removeFromPolygon changeAnnoID Key_Enter defineAnnotator shortName removeAnnotationLabel imageOpened configurationList findIntersectingAnnotation currentVP getMouseEventPosition showImage addPolygonAnnotation setChecked numberToPosition uid list screenToSlide getAllPersons redefineScreeningLastUpper setAsCenter name simplifyPolygon QMenu lastAnnotationClass isOpen copyAnnotation removeAnnotation removePolygonPoint mapToGlobal saveTIFFfile append screeningMode addMenu changeAnnotation range getZoomValue pos showDBentryCount resetGuidedScreening addSpotAnnotation partial removeLastPolygonPoint findAllAnnotationLabels setShortcut addAction setCheckable unique addSeparator annotationsList positionInAnnotationHandle agreedClass hitEscape exec_ extendPolygonPoint addToPolygon addAnnotationLabel setAgreedAnno sendAnnoToPlugin polygonAnnotation getAllClasses findClickAnnotation len leftClickImage rightClickImage hitEscape Delete QShortcut partial nextScreeningStep deleteImageLabel connect toggleOneClass addImageLabel previousFrame setShortcut QKeySequence clickAnnoclass Backspace append deleteCurrentSelection range nextFrame value openCustomDB createNewDatabase close connect setShortcut openSlideDialog addCellClass addAnnotator openDatabase centralwidget ExtendedSelection MinimumExpanding annotationTypeTableView tab1Layout setHorizontalStretch hasHeightForWidth QWidget setText QSlider setProperty tab1widget setRowCount addWidget setToolTip QProgressBar tab2widget setEnabled setVerticalStretch QVBoxLayout addItems setStyleSheet tab3widget setSizePolicy inspectorTableView tabView QSize Horizontal QSizePolicy statisticView opacitySlider QComboBox setMinimumSize QTabWidget Maximum opacityLabel Minimum setLayout setSelectionMode setVisible tab2Layout setHeightForWidth statusLabel setHidden tab3Layout ClassSelectTableWidget annotatorComboBox setOrientation setMaximumSize Fixed QLabel stretchlabel progressBar setObjectName addTab setColumnCount QTableView show showMessage WindowStaysOnTopHint mask QPixmap setMask QSplashScreen processEvents setApplicationName black setPalette WindowText red sep HighlightedText create setWindowIcon QPalette dirname setStyleSheet AlternateBase ToolTipBase Link ToolTipText Window realpath fromRgb Highlight ButtonText print system white QIcon Text setColor Button BrightText Base MODE_ANNOTATE_AREA setUIMode iconBlinded MODE_ANNOTATE_FLAG iconQuestion setChecked setDiscoveryMode iconDrawCircle iconBack iconScreening connect setEnabled MODE_ANNOTATE_POLYGON startStopScreening iconPreviousScreen iconNextScreen addToolBar iconAnnoTN partial backToLastAnnotation iconView MODE_ANNOTATE_WAND iconRect MODE_ANNOTATE_SPOT setCheckable addAction addSeparator nextScreeningStep iconFlag QAction iconOverlay setBlindedMode iconCircle MODE_VIEW iconPolygon QIcon iconWand MODE_ANNOTATE_CIRCLE previousScreeningStep setOverlayHeatmap pyqtSignal close show time partial isActiveWindow close showMessage WindowStaysOnTopHint mask QPixmap setMask sleep QSplashScreen processEvents close setWindowModality int str QPushButton addWidget setText partial setWindowTitle exec_ QLabel connect QLineEdit dict ApplicationModal setLayout QDialog QGridLayout exec_ setIcon Question No button setWindowTitle QMessageBox Yes setStandardButtons setText Abort value text close currentIndex setValue setValue getOpenFileName setText value setText setMinimum setEchoMode QSlider addWidget setWindowTitle connect QLineEdit setCurrentIndex addItem setValue PasswordEchoOnEdit QGridLayout value Horizontal partial QComboBox setLayout setWindowModality enumerate int QPushButton exec_ setOrientation setMaximum QLabel ApplicationModal dict QDialog setValue close refreshMenu setValue close refreshMenu show QPushButton time partial move isActiveWindow close connect WindowStaysOnTopHint mask QPixmap setMask sleep QSplashScreen processEvents list IMAGE_PLUGIN Queue RGB_OVERLAY WHOLESLIDE_PLUGIN NO_OVERLAY sort len flatten eigh list exp shape int64 append rot90 range asarray arctan2 concatenate copy flip T mypercentile print reshape float32 dot cov array lstsq len RGB_IMAGE HEATMAP insertNewAreaAnnotation create getAllPersons insertNewSpotAnnotation loadIntoMemory renameClass countEntryPerClass insertNewSlide setAgreedClass insertAnnotator getAllClasses Database insertClass ExactAPIs ApiClient results destroy_annotation_type remove results imwrite ApiClient destroy_image id ExactAPIs randint imread download_image results ApiClient destroy_image destroy_annotation id ExactAPIs destroy_annotation_type imwrite id ExactAPIs insertClass str create ApiClient destroy_annotation Annotation strftime annotations terminate insertAnnotator retrieve_image_set results setAnnotationLabel insertNewPolygonAnnotation list_to_exactvector uuid4 join sync create_annotation remove ExactManager print destroy_image upload_image_to_imageset setExactPerson insertNewSlide randint loadIntoMemory array screeningMap zeros uint8 annotate

 # SlideRunner [](https://doi.org/10.1007/978-3-662-56537-7_81) *SlideRunner* is a tool for massive cell annotations in whole slide images. It has been created in close cooperation between the [Pattern Recognition Lab](https://www5.cs.fau.de), Friedrich-Alexander-Universität Erlangen-Nürnberg and the [Institute of Veterenary Pathology](http://www.vetmed.fu-berlin.de/einrichtungen/institute/we12/index.html), Freie Universität Berlin. Development is continued now at Technische Hochschule Ingolstadt. If you use the software for research purposes, please cite our paper: > M. Aubreville, C. Bertram, R. Klopfleisch and A. Maier (2018) SlideRunner - A Tool for Massive Cell Annotations in Whole Slide Images. In: Bildverarbeitung für die Medizin 2018. Springer Vieweg, Berlin, Heidelberg, 2018. pp. 309-314. [link](https://www.springerprofessional.de/sliderunner/15478976) [arXiv:1802.02347](https://arxiv.org/abs/1802.02347) Please find the authors webpage at: https://imi.thi.de ## Version 2.0.0 With so many new features, it is time to declare Version 2. While we initially wanted to declare the version on this years BVM, it has all been delayed a bit. But for a good reason: Following requests by our pathologists, SlideRunner now has support for a much broader range of image formats.

DeepSceneSeg/EfficientPS

['panoptic segmentation', 'instance segmentation', 'semantic segmentation']

['EfficientPS: Efficient Panoptic Segmentation']

mmdet/models/roi_extractors/__init__.py mmdet/ops/roi_sampling/__init__.py mmdet/__init__.py mmdet/datasets/pipelines/instaboost.py tools/cityscapes_save_predictions.py mmdet/ops/masked_conv/__init__.py mmdet/ops/norm.py mmdet/core/mask/utils.py mmdet/core/bbox/bbox_target.py mmdet/core/bbox/samplers/pseudo_sampler.py tests/test_heads.py mmdet/core/utils/misc.py mmdet/core/post_processing/bbox_nms.py mmdet/ops/grid_sampler/__init__.py mmdet/models/anchor_heads/__init__.py mmdet/ops/context_block.py mmdet/models/anchor_heads/sep_rpn_head.py mmdet/models/anchor_heads/free_anchor_retina_head.py mmdet/core/bbox/assigners/__init__.py mmdet/models/mask_heads/fcn_mask_head.py mmdet/core/bbox/assigners/base_assigner.py tests/test_roi_sampling.py mmdet/datasets/builder.py mmdet/core/evaluation/eval_hooks.py mmdet/models/losses/mse_loss.py efficientNet/geffnet/config.py mmdet/models/efficientps/two_stage.py mmdet/models/shared_heads/res_layer.py mmdet/core/bbox/geometry.py mmdet/core/bbox/assigners/assign_result.py tools/cityscapes_demo.py mmdet/utils/collect_env.py mmdet/core/utils/dist_utils.py mmdet/datasets/loader/__init__.py mmdet/models/losses/accuracy.py mmdet/models/mask_heads/grid_head.py mmdet/ops/nms/__init__.py setup.py mmdet/models/bbox_heads/double_bbox_head.py mmdet/models/builder.py mmdet/utils/__init__.py mmdet/datasets/loader/build_loader.py mmdet/core/bbox/samplers/iou_balanced_neg_sampler.py mmdet/models/registry.py mmdet/ops/activation.py mmdet/ops/roi_align/roi_align.py mmdet/datasets/__init__.py mmdet/datasets/pipelines/loading.py mmdet/models/anchor_heads/fovea_head.py mmdet/models/losses/balanced_l1_loss.py mmdet/models/efficientps/__init__.py mmdet/core/anchor/point_target.py mmdet/models/bbox_heads/convfc_bbox_head.py mmdet/models/losses/focal_loss.py mmdet/core/evaluation/class_names.py mmdet/core/evaluation/panoptic.py mmdet/datasets/pipelines/auto_augment.py mmdet/core/anchor/anchor_target.py mmdet/core/bbox/demodata.py mmdet/ops/carafe/__init__.py tools/test.py efficientNet/geffnet/activations/activations_me.py mmdet/models/efficientps/test_mixins.py efficientNet/geffnet/activations/activations.py mmdet/core/bbox/samplers/instance_balanced_pos_sampler.py mmdet/core/bbox/samplers/sampling_result.py mmdet/models/necks/two_way_fpn.py mmdet/datasets/cityscapes.py mmdet/core/evaluation/recall.py mmdet/core/optimizer/copy_of_sgd.py efficientNet/geffnet/mobilenetv3.py mmdet/core/post_processing/merge_augs.py mmdet/models/mask_heads/efficientps_semantic_head.py mmdet/ops/affine_grid/__init__.py mmdet/models/utils/weight_init.py mmdet/core/fp16/__init__.py mmdet/models/anchor_heads/ga_rpn_head.py mmdet/models/anchor_heads/ssd_head.py mmdet/models/bbox_heads/__init__.py mmdet/models/efficientps/base.py efficientNet/setup.py mmdet/models/utils/__init__.py mmdet/core/__init__.py mmdet/utils/logger.py tests/test_async.py mmdet/core/bbox/assign_sampling.py mmdet/models/anchor_heads/guided_anchor_head.py mmdet/models/mask_heads/fused_semantic_head.py mmdet/models/mask_heads/fcn_sep_mask_head.py mmdet/ops/__init__.py mmdet/utils/contextmanagers.py mmdet/apis/inference.py mmdet/models/anchor_heads/fcos_head.py mmdet/core/bbox/assigners/atss_assigner.py tests/test_forward.py mmdet/ops/affine_grid/affine_grid.py mmdet/core/fp16/hooks.py mmdet/core/bbox/samplers/base_sampler.py tools/fuse_conv_bn.py mmdet/models/anchor_heads/ga_retina_head.py tests/test_sampler.py mmdet/ops/roi_align/__init__.py mmdet/models/losses/utils.py efficientNet/geffnet/__init__.py mmdet/models/efficientps/rpn.py mmdet/datasets/pipelines/__init__.py mmdet/core/bbox/transforms.py mmdet/ops/roi_align/gradcheck.py mmdet/models/backbones/__init__.py mmdet/core/optimizer/registry.py mmdet/core/bbox/assigners/max_iou_assigner.py mmdet/models/anchor_heads/reppoints_head.py mmdet/models/mask_heads/htc_mask_head.py mmdet/ops/carafe/grad_check.py mmdet/core/bbox/assigners/point_assigner.py mmdet/models/anchor_heads/retina_head.py mmdet/models/bbox_heads/bbox_head.py mmdet/ops/roi_pool/__init__.py mmdet/ops/roi_sampling/functions.py mmdet/ops/conv_module.py mmdet/ops/conv.py tests/test_soft_nms.py mmdet/ops/nms/nms_wrapper.py mmdet/models/losses/iou_loss.py mmdet/apis/test.py efficientNet/geffnet/activations/__init__.py mmdet/core/optimizer/builder.py mmdet/ops/depthwise_separable_conv_module.py tests/test_assigner.py mmdet/ops/dcn/deform_conv.py mmdet/apis/__init__.py tests/async_benchmark.py tools/cityscapes_inference.py tools/train.py mmdet/core/evaluation/mean_ap.py mmdet/models/roi_extractors/single_level.py mmdet/ops/generalized_attention.py mmdet/core/anchor/__init__.py mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss.py mmdet/datasets/dataset_wrappers.py mmdet/ops/non_local.py mmdet/models/losses/cross_entropy_loss.py mmdet/ops/sigmoid_focal_loss/__init__.py mmdet/datasets/registry.py tools/kitti_demo.py mmdet/datasets/pipelines/formating.py configs/efficientPS_singlegpu_sample.py mmdet/models/losses/smooth_l1_loss.py mmdet/core/bbox/assigners/approx_max_iou_assigner.py mmdet/datasets/voc.py mmdet/models/necks/__init__.py mmdet/models/losses/ghm_loss.py mmdet/ops/carafe/carafe.py tests/test_nms.py mmdet/core/anchor/anchor_generator.py efficientNet/geffnet/conv2d_layers.py mmdet/core/mask/mask_target.py mmdet/utils/profiling.py configs/efficientPS_multigpu_sample.py mmdet/models/losses/__init__.py mmdet/ops/scale.py mmdet/ops/saconv.py mmdet/ops/masked_conv/masked_conv.py mmdet/models/anchor_heads/anchor_head.py mmdet/models/anchor_heads/retina_sepbn_head.py mmdet/core/post_processing/__init__.py mmdet/datasets/xml_style.py mmdet/datasets/coco.py mmdet/models/anchor_heads/rpn_head.py mmdet/models/__init__.py mmdet/datasets/pipelines/compose.py mmdet/core/bbox/samplers/__init__.py mmdet/ops/dcn/deform_pool.py mmdet/core/bbox/samplers/combined_sampler.py tools/convert_cityscapes.py efficientNet/geffnet/activations/activations_jit.py mmdet/core/bbox/samplers/ohem_sampler.py mmdet/ops/roi_pool/gradcheck.py mmdet/core/fp16/decorators.py mmdet/datasets/custom.py mmdet/datasets/wider_face.py mmdet/core/evaluation/bbox_overlaps.py mmdet/utils/flops_counter.py mmdet/ops/utils/__init__.py mmdet/ops/upsample.py mmdet/core/mask/__init__.py mmdet/ops/roi_pool/roi_pool.py tests/test_utils.py tools/convert_kitti.py mmdet/core/fp16/utils.py mmdet/models/backbones/resnet.py mmdet/utils/util_mixins.py mmdet/core/bbox/samplers/random_sampler.py mmdet/models/mask_heads/__init__.py mmdet/datasets/loader/sampler.py mmdet/datasets/pipelines/test_aug.py mmdet/ops/dcn/__init__.py mmdet/utils/registry.py mmdet/ops/carafe/setup.py mmdet/core/anchor/guided_anchor_target.py mmdet/core/bbox/__init__.py mmdet/models/efficientps/efficientPS.py mmdet/core/optimizer/__init__.py efficientNet/geffnet/version.py mmdet/models/mask_heads/maskiou_head.py mmdet/ops/grid_sampler/grid_sampler.py mmdet/ops/conv_ws.py efficientNet/geffnet/gen_efficientnet.py efficientNet/geffnet/efficientnet_builder.py mmdet/models/anchor_heads/atss_head.py tests/test_config.py mmdet/core/anchor/point_generator.py efficientNet/geffnet/model_factory.py mmdet/apis/train.py efficientNet/geffnet/helpers.py mmdet/core/evaluation/__init__.py mmdet/core/utils/__init__.py mmdet/datasets/pipelines/transforms.py mmdet/models/shared_heads/__init__.py make_cuda_ext write_version_py readme make_extension get_version parse_requirements get_git_hash get_hash find_sources set_no_jit is_no_jit layer_config_kwargs set_scriptable set_exportable is_exportable set_layer_config is_scriptable _calc_same_pad _get_padding _same_pad_arg conv2d_same get_padding_value _split_channels select_conv2d _is_static_pad Conv2dSame create_conv2d_pad CondConv2d get_condconv_initializer MixedConv2d Conv2dSameExport _ntuple drop_connect _scale_stage_depth _parse_ksize EdgeResidual resolve_act_layer round_channels initialize_weight_goog resolve_bn_args initialize_weight_default SqueezeExcite resolve_se_args ConvBnAct DepthwiseSeparableConv CondConvResidual EfficientNetBuilder make_divisible InvertedResidual decode_arch_def get_bn_args_tf _decode_block_str fbnetc_100 tf_efficientnet_lite3 efficientnet_lite1 tf_efficientnet_b2 tf_efficientnet_b5_ap tf_efficientnet_b8_ap efficientnet_b0 mixnet_xl mnasnet_small tf_efficientnet_b0_ns mnasnet_075 tf_efficientnet_cc_b1_8e mnasnet_100 efficientnet_b1 tf_efficientnet_lite1 efficientnet_b8 efficientnet_cc_b1_8e _create_model efficientnet_b5 tf_efficientnet_b6 semnasnet_100 tf_mixnet_s efficientnet_lite0 mnasnet_b1 tf_efficientnet_b3 tf_efficientnet_b5_ns tf_efficientnet_b4_ns _gen_efficientnet_edge efficientnet_cc_b0_4e tf_efficientnet_l2_ns tf_efficientnet_b7_ns spnasnet_100 mixnet_xxl tf_efficientnet_b1_ap mobilenetv2_120d tf_efficientnet_b7 tf_efficientnet_b6_ns _gen_mnasnet_small efficientnet_b7 tf_efficientnet_el efficientnet_lite2 efficientnet_b4 tf_efficientnet_lite2 tf_efficientnet_lite4 semnasnet_140 tf_efficientnet_b8 tf_efficientnet_b6_ap efficientnet_el tf_mixnet_l tf_efficientnet_b0_ap tf_efficientnet_b5 _gen_spnasnet mobilenetv2_110d efficientnet_cc_b0_8e mixnet_l efficientnet_lite4 tf_efficientnet_b0 semnasnet_050 _gen_efficientnet_condconv efficientnet_b2 tf_efficientnet_b1_ns _gen_efficientnet_lite tf_efficientnet_es mnasnet_a1 tf_efficientnet_b2_ns tf_efficientnet_b3_ns mnasnet_140 tf_efficientnet_b4 mobilenetv2_140 efficientnet_b3 efficientnet_em _gen_efficientnet tf_efficientnet_lite0 _gen_fbnetc tf_mixnet_m mnasnet_050 mixnet_m tf_efficientnet_em tf_efficientnet_b3_ap semnasnet_075 _gen_mixnet_m tf_efficientnet_b2_ap _gen_mnasnet_b1 tf_efficientnet_b7_ap tf_efficientnet_b1 _gen_mobilenet_v2 tf_efficientnet_cc_b0_4e _gen_mnasnet_a1 tf_efficientnet_cc_b0_8e tf_efficientnet_b4_ap GenEfficientNet _gen_mixnet_s efficientnet_es efficientnet_l2 mobilenetv2_100 mixnet_s efficientnet_lite3 tf_efficientnet_l2_ns_475 efficientnet_b6 load_checkpoint load_pretrained mobilenetv3_small_075 _gen_mobilenet_v3_rw MobileNetV3 mobilenetv3_large_minimal_100 tf_mobilenetv3_large_minimal_100 tf_mobilenetv3_large_075 _create_model tf_mobilenetv3_small_minimal_100 mobilenetv3_small_100 tf_mobilenetv3_small_075 tf_mobilenetv3_large_100 mobilenetv3_large_100 mobilenetv3_rw mobilenetv3_large_075 _gen_mobilenet_v3 mobilenetv3_small_minimal_100 tf_mobilenetv3_small_100 create_model Swish tanh hard_swish swish Mish sigmoid Sigmoid Tanh HardSwish hard_sigmoid Identity HardSigmoid mish SwishJit MishJit swish_jit hard_sigmoid_jit HardSwishJit hard_swish_jit HardSigmoidJit mish_jit hard_swish_jit_bwd swish_me hard_sigmoid_jit_fwd mish_jit_bwd swish_jit_fwd MishJitAutoFn SwishMe hard_sigmoid_me MishMe HardSwishJitAutoFn hard_sigmoid_jit_bwd mish_me SwishJitAutoFn swish_jit_bwd HardSigmoidMe mish_jit_fwd hard_swish_me HardSwishMe HardSigmoidJitAutoFn hard_swish_jit_fwd add_override_act_layer clear_override_act_fn add_override_act_fn clear_override_act_layer update_override_act_layer get_act_fn get_act_layer update_override_act_fn inference_detector show_result_pyplot LoadImage init_detector show_result multi_gpu_test collect_results_cpu collect_results_gpu single_gpu_test _dist_train set_random_seed batch_processor _non_dist_train train_detector parse_losses AnchorGenerator anchor_target unmap anchor_inside_flags images_to_levels anchor_target_single ga_loc_target ga_shape_target_single calc_region images_to_levels ga_shape_target PointGenerator images_to_levels point_target unmap point_target_single assign_and_sample build_assigner build_sampler bbox_target_single expand_target bbox_target ensure_rng random_boxes bbox_overlaps delta2bbox roi2bbox bbox_flip distance2bbox bbox2delta bbox_mapping bbox2result bbox_mapping_back bbox2roi ApproxMaxIoUAssigner AssignResult ATSSAssigner BaseAssigner MaxIoUAssigner PointAssigner BaseSampler CombinedSampler InstanceBalancedPosSampler IoUBalancedNegSampler OHEMSampler PseudoSampler RandomSampler SamplingResult bbox_overlaps stryker_classes get_classes imagenet_vid_classes voc_classes imagenet_det_classes coco_classes cityscapes_originalIds cityscapes_classes wider_face_classes DistEvalHook EvalHook eval_map tpfp_imagenet print_map_summary average_precision get_cls_results tpfp_default createDir save_panoptic_eval plot_iou_recall set_recall_param print_recall_summary _recalls eval_recalls plot_num_recall force_fp32 auto_fp16 Fp16OptimizerHook wrap_fp16_model patch_forward_method patch_norm_fp32 cast_tensor_type mask_target mask_target_single split_combined_polys build_optimizer CopyOfSGD register_torch_optimizers multiclass_nms merge_aug_scores merge_aug_masks merge_aug_bboxes merge_aug_proposals DistOptimizerHook allreduce_grads _allreduce_coalesced unmap tensor2imgs multi_apply build_dataset _concat_dataset CityscapesDataset CocoDataset CustomDataset RepeatDataset ConcatDataset VOCDataset WIDERFaceDataset XMLDataset build_dataloader worker_init_fn GroupSampler DistributedSampler DistributedGroupSampler Rotate ContrastTransform ColorTransform Translate BrightnessTransform enhance_level_to_value EqualizeTransform random_negative bbox2fields Shear level_to_value AutoAugment Compose DefaultFormatBundle Transpose ToTensor Collect WrapFieldsToLists to_tensor ImageToTensor ToDataContainer InstaBoost LoadImageFromFile LoadMultiChannelImageFromFiles LoadProposals LoadAnnotations MultiScaleFlipAug RandomFlip Pad Corrupt PhotoMetricDistortion MinIoURandomCrop SegRescale Resize RandomCrop Albu Normalize Expand build_shared_head build_detector build_loss build build_backbone build_roi_extractor build_head build_neck AnchorHead reduce_mean ATSSHead FCOSHead FeatureAlign FoveaHead FreeAnchorRetinaHead GARetinaHead GARPNHead FeatureAdaption GuidedAnchorHead RepPointsHead RetinaHead RetinaSepBNHead RPNHead SepRPNHead SSDHead make_res_layer ResNet Bottleneck get_activation BasicBlock BBoxHead SharedFCBBoxHead ConvFCBBoxHead DoubleConvFCBBoxHead BasicResBlock BaseDetector EfficientPS RPN MaskTestMixin BBoxTestMixin RPNTestMixin TwoStageDetector Accuracy accuracy BalancedL1Loss balanced_l1_loss binary_cross_entropy mask_cross_entropy _expand_binary_labels CrossEntropyLoss cross_entropy sigmoid_focal_loss py_sigmoid_focal_loss FocalLoss _expand_binary_labels GHMR GHMC bounded_iou_loss iou_loss IoULoss BoundedIoULoss GIoULoss giou_loss mse_loss MSELoss smooth_l1_loss SmoothL1Loss weight_reduce_loss weighted_loss reduce_loss LSFE EfficientPSSemanticHead MC DPC FCNMaskHead FCNSepMaskHead FusedSemanticHead GridHead HTCMaskHead MaskIoUHead TWOWAYFPN SingleRoIExtractor ResLayer bias_init_with_prob build_activation_layer last_zero_init ContextBlock build_conv_layer ConvModule ConvAWS2d conv_ws_2d ConvWS2d DepthwiseSeparableConvModule GeneralizedAttention NonLocal2D build_norm_layer SAConv2d Scale build_upsample_layer PixelShufflePack affine_grid _AffineGridGenerator CARAFEPack CARAFENaive CARAFENaiveFunction CARAFE CARAFEFunction DeformConvFunction ModulatedDeformConv DeformConvPack ModulatedDeformConvPack DeformConv ModulatedDeformConvFunction DeformRoIPoolingPack DeformRoIPoolingFunction ModulatedDeformRoIPoolingPack DeformRoIPooling _GridSampler grid_sample MaskedConv2dFunction MaskedConv2d nms soft_nms RoIAlign RoIAlignFunction RoIPool RoIPoolFunction ROISampling roi_sampling invert_roi_bbx SigmoidFocalLoss SigmoidFocalLossFunction collect_env add_flops_counting_methods add_flops_counter_hook_function bn_flops_counter_hook reset_flops_count gn_flops_counter_hook relu_flops_counter_hook deconv_flops_counter_hook get_model_parameters_number add_flops_mask flops_to_string params_to_string remove_flops_mask remove_batch_counter_hook_function start_flops_count add_batch_counter_variables_or_reset pool_flops_counter_hook empty_flops_counter_hook add_flops_mask_variable_or_reset add_batch_counter_hook_function get_model_complexity_info conv_flops_counter_hook remove_flops_counter_hook_function batch_counter_hook add_flops_counter_variable_or_reset is_supported_instance stop_flops_count upsample_flops_counter_hook linear_flops_counter_hook compute_average_flops_cost print_model_with_flops print_log get_root_logger profile_time build_from_cfg Registry NiceRepr test_approx_iou_assigner_with_empty_boxes test_point_assigner test_max_iou_assigner test_approx_iou_assigner_with_empty_gt test_point_assigner_with_empty_gt test_random_assign_result test_approx_iou_assigner test_point_assigner_with_empty_boxes_and_gt test_max_iou_assigner_with_empty_boxes_and_gt test_max_iou_assigner_with_empty_boxes test_max_iou_assigner_with_empty_gt test_max_iou_assigner_with_ignore test_approx_iou_assigner_with_empty_boxes_and_gt test_max_iou_assigner_with_empty_boxes_and_ignore MaskRCNNDetector AsyncTestCase AsyncInferenceTestCase test_config_build_detector _get_config_directory test_config_data_pipeline _get_config_directory _get_config_module test_cascade_forward test_faster_rcnn_forward _get_detector_cfg test_faster_rcnn_ohem_forward _demo_mm_inputs test_rpn_forward test_retina_ghm_forward test_ssd300_forward _demodata_refine_boxes test_anchor_head_loss test_bbox_head_loss test_refine_boxes test_nms_device_and_dtypes_cpu test_nms_device_and_dtypes_gpu test_random_sampler_empty_pred test_ohem_sampler _context_for_ohem test_ohem_sampler_empty_gt test_random_sampler_empty_gt test_random_sampler test_random_sample_result test_ohem_sampler_empty_pred test_soft_nms_device_and_dtypes_cpu test_params_to_string main parse_args main parse_args main parse_args _get_meta _ensure_dir _Counter main _get_images _init_counter _Worker _get_meta _ensure_dir _Counter main _get_images _init_counter _Worker main fuse_module fuse_conv_bn parse_args main parse_args main parse_args MultipleKVAction main parse_args append join listdir splitext decode _minimal_ext_cmd exists get_hash list gen_packages_items _calc_same_pad _calc_same_pad pad _is_static_pad lower _get_padding isinstance pop get_padding_value is_exportable setdefault pop isinstance create_conv2d_pad CondConv2d MixedConv2d pop items list setdefault pop get_act_layer isinstance int max div rand floor_ isdigit int dict get_act_layer startswith split int extend round zip ceil sum max append append _scale_stage_depth _decode_block_str enumerate num_experts isinstance fill_ size weight_shape out_channels Conv2d sqrt normal_ get_condconv_initializer zero_ init_weight_fn uniform_ BatchNorm2d weight Linear init_fn kaiming_uniform_ partial num_experts isinstance fill_ weight_shape Conv2d kaiming_normal_ get_condconv_initializer zero_ BatchNorm2d weight Linear pop load_pretrained as_sequential GenEfficientNet pop _gen_mnasnet_b1 _gen_mnasnet_b1 _gen_mnasnet_b1 _gen_mnasnet_b1 _gen_mnasnet_a1 _gen_mnasnet_a1 _gen_mnasnet_a1 _gen_mnasnet_a1 _gen_mnasnet_small _gen_mobilenet_v2 _gen_mobilenet_v2 _gen_mobilenet_v2 _gen_mobilenet_v2 _gen_fbnetc _gen_spnasnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet_edge _gen_efficientnet_edge _gen_efficientnet_edge _gen_efficientnet_condconv _gen_efficientnet_condconv _gen_efficientnet_condconv _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet _gen_efficientnet_edge _gen_efficientnet_edge _gen_efficientnet_edge _gen_efficientnet_condconv _gen_efficientnet_condconv _gen_efficientnet_condconv _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_efficientnet_lite _gen_mixnet_s _gen_mixnet_m _gen_mixnet_m _gen_mixnet_m _gen_mixnet_m _gen_mixnet_s _gen_mixnet_m _gen_mixnet_m load items list format print OrderedDict load_state_dict startswith sum format print load_state_dict load_state_dict_from_url filter_fn MobileNetV3 _gen_mobilenet_v3_rw _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 _gen_mobilenet_v3 dict load_checkpoint create_fn div_ sigmoid sigmoid tanh ones_like ones_like where update dict update dict get_classes isinstance model load_checkpoint warn eval build_detector fromfile to Compose cfg dict test_pipeline device seed int bool concat_list isinstance concatenate imshow_det_bboxes astype copy vstack imread bgr2rgb imshow show_result figure update show_result size ProgressBar eval save_panoptic_eval append dataset range enumerate len update get_dist_info size ProgressBar eval collect_results_cpu collect_results_gpu save_panoptic_eval append dataset range enumerate len rstrip tensor broadcast list get_dist_info mkdtemp encode append range dump format bytearray zip load join barrier extend rmtree mkdir_or_exist full list get_dist_info bytearray dumps extend tobytes shape loads all_gather zip append tensor max zeros seed manual_seed_all manual_seed items list isinstance clone get_world_size OrderedDict mean all_reduce item div_ Tensor sum dict parse_losses model log_level _non_dist_train get_root_logger _dist_train workflow MMDistributedDataParallel DistSamplerSeedHook cuda run total_epochs build_optimizer checkpoint_config work_dir build_dataset get val load_from DistEvalHook build_dataloader resume_from register_training_hooks resume optimizer DistOptimizerHook lr_config load_checkpoint register_hook dict Runner log_config Fp16OptimizerHook workflow cuda run total_epochs build_optimizer checkpoint_config work_dir build_dataset optimizer_config EvalHook get val load_from build_dataloader resume_from register_training_hooks resume optimizer lr_config load_checkpoint register_hook dict Runner log_config Fp16OptimizerHook multi_apply images_to_levels any sum range cat len append stack squeeze assign_and_sample zeros_like PseudoSampler pos_gt_bboxes size pos_weight anchor_inside_flags unmap sample new_zeros build_assigner assign pos_inds bbox2delta allowed_border neg_inds pos_bboxes assigner uint8 type new_full clamp long new_full zeros_like calc_region size sqrt log2 floor full_like item append zeros float sum long range len multi_apply images_to_levels any append sum range cat len ga_assigner build_sampler ga_sampler PseudoSampler zeros_like reshape pos_gt_bboxes size unmap build_assigner assign pos_inds sample neg_inds pos_bboxes multi_apply images_to_levels any sum range cat len assign_and_sample zeros_like PseudoSampler pos_gt_bboxes size pos_weight unmap new_zeros build_assigner assign pos_inds sample neg_inds assigner BaseAssigner isinstance BaseSampler isinstance build_sampler sampler build_assigner assign sample assigner multi_apply cat bbox2delta size new_zeros squeeze new_zeros _rand RandomState isinstance minimum astype float32 maximum from_numpy ensure_rng clamp size min max stack unsqueeze div_ float log exp clamp size repeat expand_as view_as abs log addcmul Tensor ndarray isinstance clone bbox_flip new_full new_zeros append cat enumerate cpu append unique numpy clamp minimum T astype maximum float32 zeros range items list eval is_str arange ones hstack maximum zeros sum range minimum zeros_like concatenate argsort vstack zeros bbox_overlaps range enumerate len max zeros_like concatenate argsort vstack zeros argmax bbox_overlaps enumerate len append empty starmap cumsum tuple vstack Pool get_cls_results list print_map_summary append range eps mean item zip enumerate maximum argsort any average_precision zeros len get_classes ndarray format isinstance table len is_str print_log AsciiTable append zeros range enumerate makedirs join replace createDir cityscapes_originalIds unique save append zeros sum sum sort hstack copy zeros float argmax fliplr range enumerate array isinstance min set_recall_param print_recall_summary _recalls array append zeros bbox_overlaps range len arange table insert size tolist print_log AsciiTable append array enumerate show ndarray plot isinstance xlabel tolist axis ylabel figure show ndarray plot isinstance xlabel tolist axis ylabel figure hasattr patch_norm_fp32 modules half children isinstance half patch_forward_method float forward ndarray isinstance Iterable Tensor Mapping list map cat mask_size imresize size astype maximum range new_zeros shape int32 device append to numpy clip _pair tolist append slice_list range len pop get named_modules hasattr replace endswith search copy named_parameters dict parameters append module dir getattr startswith register_module append optim pop sort size copy new_zeros expand nms_op getattr to max cat nms nms_thr sort min clone max_num zip append bbox_mapping_back cat append mean bbox_mapping_back zip Tensor isinstance average mean array list _take_tensors _flatten_dense_tensors zip _unflatten_dense_tensors OrderedDict all_reduce copy_ div_ append type values all_reduce _allreduce_coalesced get_world_size div_ uint8 transpose size astype ascontiguousarray append array range list map get deepcopy isinstance append build_dataset range len get isinstance ConcatDataset _concat_dataset build_from_cfg RepeatDataset DistributedSampler get_dist_info DistributedGroupSampler DataLoader seed Tensor ndarray isinstance isinstance all_reduce clone get_world_size div_ block Sequential build_conv_layer append range expansion topk isinstance size t eq mul_ expand_as append sum max log abs e where float weight_reduce_loss new_full size squeeze expand size weight_reduce_loss binary_cross_entropy_with_logits _expand_binary_labels float squeeze arange type_as sigmoid pow weight_reduce_loss binary_cross_entropy_with_logits _sigmoid_focal_loss weight_reduce_loss view clamp view zeros_like size min where abs max clamp min max abs where get_enum sum reduce_loss float pop activation copy Sequential isinstance constant_init dict conv_layer pop copy size view pop str setdefault norm_layer copy parameters _specify_ddp_gpu_num pop upsample copy dim range ndarray isinstance new_zeros Tensor to numpy is_cuda ndarray isinstance from_numpy cpu Tensor stack show join str format defaultdict replace list items check_output strip get_compiler_version device_count __version__ get_compiling_cuda_version platform is_available range append flops_model get_model_parameters_number input_constructor stop_flops_count add_flops_counting_methods start_flops_count compute_average_flops_cost new_empty print_model_with_flops print compute_average_flops_cost apply sum __get__ reset_flops_count apply __batch_counter__ is_supported_instance modules add_batch_counter_hook_function apply remove_batch_counter_hook_function apply add_batch_counter_variables_or_reset apply apply apply type issubclass numel shape affine prod shape affine prod groups kernel_size out_channels in_channels list kernel_size out_channels groups in_channels expand sum prod print len register_forward_hook hasattr remove hasattr is_supported_instance items issubclass hasattr register_forward_hook type is_supported_instance remove is_supported_instance hasattr is_supported_instance setFormatter basicConfig get_dist_info getLogger addHandler Formatter setLevel hasHandlers FileHandler isinstance print get_root_logger Logger log record_event monotonic Event pop get list items setdefault copy is_str isclass FloatTensor assign LongTensor MaxIoUAssigner LongTensor FloatTensor assign MaxIoUAssigner Tensor FloatTensor assign LongTensor MaxIoUAssigner LongTensor FloatTensor assign MaxIoUAssigner empty LongTensor FloatTensor assign MaxIoUAssigner Tensor empty assign empty MaxIoUAssigner LongTensor assign PointAssigner FloatTensor LongTensor assign PointAssigner FloatTensor assign PointAssigner FloatTensor FloatTensor assign LongTensor ApproxMaxIoUAssigner FloatTensor assign LongTensor ApproxMaxIoUAssigner FloatTensor assign empty ApproxMaxIoUAssigner assign empty ApproxMaxIoUAssigner random int getenv join dirname join format _get_config_directory model print build_detector import_module_from_path train_cfg test_cfg len pop join get format _get_config_directory print Compose astype float32 import_module_from_path dict train_pipeline test_pipeline randint len join _get_config_directory import_module_from_path Config deepcopy model _get_config_module train_cfg test_cfg pop _get_detector_cfg _demo_mm_inputs build_detector forward pop _get_detector_cfg _demo_mm_inputs build_detector forward pop _get_detector_cfg _demo_mm_inputs build_detector is_available forward cuda pop _get_detector_cfg _demo_mm_inputs item build_detector float forward pop _get_detector_cfg _demo_mm_inputs item build_detector float forward pop _get_detector_cfg _demo_mm_inputs item build_detector float forward T RandomState LongTensor FloatTensor rand append randint range clip Config sum AnchorHead forward loss Config _dummy_bbox_sampling forward rand get_target BBoxHead sum loss format print BBoxHead refine_bboxes _demodata_refine_boxes int random_boxes group_items astype from_numpy ensure_rng numpy randint empty long cat nms FloatTensor float64 astype float32 DoubleTensor array nms format print astype float32 skip device_count to array range LongTensor FloatTensor RandomSampler assign MaxIoUAssigner sample Tensor FloatTensor RandomSampler assign MaxIoUAssigner sample empty long LongTensor FloatTensor RandomSampler assign MaxIoUAssigner sample empty build_detector _get_detector_cfg LongTensor FloatTensor OHEMSampler _context_for_ohem assign MaxIoUAssigner sample Tensor LongTensor FloatTensor OHEMSampler _context_for_ohem assign MaxIoUAssigner sample Tensor empty LongTensor FloatTensor OHEMSampler _context_for_ohem assign MaxIoUAssigner sample Tensor empty random range FloatTensor float64 astype float32 DoubleTensor soft_nms array params_to_string assert_equal add_argument ArgumentParser config imwrite wait dilation resize init_detector device out fromarray show waitKey imshow append input fromfile parse_args imread astype inference_detector copy listdir checkpoint join uint8 convert repeat numpy array makedirs str local_rank cityscapes_originalIds save sum update format replace ProgressBar mkdir unique print zeros len alpha_composite open _get_meta list items root_dir _ensure_dir out_dir _get_images _Worker join isdir glob append listdir split sum mkdir ignoreInEval Parameter eps reshape running_mean bias sqrt weight running_var named_children isinstance fuse_conv_bn Conv2d Identity fuse_module save_checkpoint model tmpdir launcher MMDistributedDataParallel cuda format_only get_dist_info format_results build_detector build_dataset gpu_collect get dump CLASSES init_dist single_gpu_test build_dataloader wrap_fp16_model test eval evaluate load_checkpoint fuse_conv_bn multi_gpu_test MMDataParallel set_random_seed localtime autoscale_lr abspath train_detector seed strftime get_root_logger work_dir val resume_from info deepcopy deterministic gpus text dict collect_env mkdir_or_exist pipeline

# EfficientPS: Efficient Panoptic Segmentation [](https://paperswithcode.com/sota/panoptic-segmentation-on-cityscapes-val?p=efficientps-efficient-panoptic-segmentation) [](https://paperswithcode.com/sota/panoptic-segmentation-on-cityscapes-test?p=efficientps-efficient-panoptic-segmentation) [](https://paperswithcode.com/sota/panoptic-segmentation-on-mapillary-val?p=efficientps-efficient-panoptic-segmentation) [](https://paperswithcode.com/sota/panoptic-segmentation-on-kitti-panoptic-segmentationl?p=efficientps-efficient-panoptic) [](https://paperswithcode.com/sota/panoptic-segmentation-on-panoptic-segmentation-on-indian-driving?p=efficientps-efficient-panoptic) EfficientPS is a state-of-the-art top-down approach for panoptic segmentation, where the goal is to assign semantic labels (e.g., car, road, tree and so on) to every pixel in the input image as well as instance labels (e.g. an id of 1, 2, 3, etc) to pixels belonging to thing classes.  This repository contains the **PyTorch implementation** of our IJCV'2021 paper [EfficientPS: Efficient Panoptic Segmentation](https://arxiv.org/abs/2004.02307). The repository builds on [mmdetection](https://github.com/open-mmlab/mmdetection) and [gen-efficientnet-pytorch](https://github.com/rwightman/gen-efficientnet-pytorch) codebases. If you find the code useful for your research, please consider citing our paper:

DevashishPrasad/Smart-Traffic-Junction

['density estimation']

['HOG, LBP and SVM based Traffic Density Estimation at Intersection']

image-processing/main.py saverois.py save_HOG_LBP.py predictor.py classifier.py distMap sqrt uint8 float32

# Smart-Traffic-Junction This repository contains the working implementation of our research paper [link](https://arxiv.org/abs/2005.01770). The paper was presented and published at IEEE PuneCon 19 conference. We propose a simple algorithm for traffic density estimation using image processing and machine learning. > **HOG, LBP and SVM based Traffic Density Estimation at Intersection**<br> > [Devashish Prasad](https://github.com/DevashishPrasad), > [Ayan Gadpal](https://github.com/ayangadpal), > [Kshitij Kapadni](https://github.com/kshitijkapadni), > [Manish Visave](https://github.com/ManishDV), > <br> ## Dataset The dataset was created using <a href="http://www.eecs.qmul.ac.uk/~sgg/QMUL_Junction_Datasets/Junction2/Junction2.html">QMUL junction 2</a> video. We manually sorted the rois of the <a href="dataset.zip">dataset</a>.

Dictanova/term-eval

['word embeddings']

['Towards a unified framework for bilingual terminology extraction of single-word and multi-word terms']

term_eval.py precision_ranges rank_results max_precision read_gold_standard main read_result precision_ranges rank_results format print add_argument gold_standard max_precision ArgumentParser read_gold_standard parse_args float sum read_result result_file len defaultdict items sorted defaultdict list tolist reverse append enumerate dict sum dict items list min

Dictanova/term-eval

Diooooo/Reimplement-HED

['boundary detection', 'edge detection']

['Holistically-Nested Edge Detection']

main.py hed.py data_parser.py read_file_lst randomize split_pair_names DataParser _to_tensor side_branch hed cross_entropy_balanced fuse_pixel_error generator train generate_edge readlines close open convert_to_tensor weighted_cross_entropy_with_logits base_dtype _to_tensor float32 log reduce_sum reduce_mean cast clip_by_value epsilon not_equal greater float32 cast int32 side_branch Model load_weights Input compile batch_size choice train_ids validate_ids get_batch_data generator join TensorBoard EarlyStopping fit_generator CSVLogger ModelCheckpoint predict

# Reimplement-HED Reimplement Holistically-Nested Edge Detection(HED) using Keras Reference: https://github.com/lc82111/Keras_HED <br> The paper ***Holistically-Nested Edge Detection*** can be found here: https://arxiv.org/abs/1504.06375

DirtyHarryLYL/Transferable-Interactiveness-Network

['human object interaction detection']

['Transferable Interactiveness Knowledge for Human-Object Interaction Detection', 'Transferable Interactiveness Knowledge for Human-Object Interaction Detection']

lib/ult/config.py lib/ult/config_vcoco.py lib/models/test_VCOCO_D_pose_pattern_naked.py tools/Train_TIN_VCOCO.py script/Download_data.py lib/networks/TIN_VCOCO.py HICO-DET_Benchmark/config.py lib/ult/visualization.py tools/Train_TIN_HICO.py lib/models/test_HICO_pose_pattern_all_wise_pair.py HICO-DET_Benchmark/HICO_Benchmark_Binary.py lib/ult/timer.py HICO-DET_Benchmark/Generate_HICO_detection_nis.py lib/ult/apply_prior.py lib/networks/TIN_HICO.py lib/models/train_Solver_VCOCO_pose_pattern_inD_more_positive.py lib/ult/Download_data.py tools/Vcoco_lis_nis.py tools/_init_paths.py lib/ult/ult.py tools/Test_TIN_HICO.py lib/models/train_Solver_HICO_pose_pattern_inD_more_positive_coslr.py tools/Test_TIN_VCOCO.py lib/ult/vsrl_eval_output_txt.py main getSigmoid Generate_HICO_detection save_HICO get_blob test_net im_detect get_blob test_net im_detect train_net SolverWrapper train_net SolverWrapper resnet_arg_scope ResNet50 resnet_arg_scope ResNet50 apply_prior download_file_from_google_drive Timer draw_relation Augmented_HO_Neg_pose_pattern_version2 Augmented_HO_spNeg_pose_pattern_version2 bbox_trans Generate_action_30 Augmented_box Get_Next_Instance_HO_Neg_pose_pattern_version2 bb_IOU Get_Next_Instance_HO_spNeg_pose_pattern_version2 get_skeleton Get_next_sp Generate_action Get_Next_Instance_HO_Neg_HICO_pose_pattern_version2 Generate_action_HICO Get_next_sp_with_pose Augmented_HO_Neg_HICO_pose_pattern_version2 draw_bounding_boxes_HOI_PIC _draw_single_box draw_bounding_boxes draw_bounding_boxes_HOI _load_vcoco get_overlap clip_xyxy_to_image voc_ap VCOCOeval download_file_from_google_drive parse_args parse_args parse_args parse_args parse_args getSigmoid apply_prior generate_pkl add_path main getSigmoid Generate_HICO_detection save_HICO get_blob test_net im_detect train_net SolverWrapper resnet_arg_scope ResNet50 apply_prior download_file_from_google_drive Timer draw_relation Augmented_HO_Neg_pose_pattern_version2 Augmented_HO_spNeg_pose_pattern_version2 bbox_trans Generate_action_30 Augmented_box Get_Next_Instance_HO_Neg_pose_pattern_version2 bb_IOU Get_Next_Instance_HO_spNeg_pose_pattern_version2 get_skeleton Get_next_sp Generate_action Get_Next_Instance_HO_Neg_HICO_pose_pattern_version2 Generate_action_HICO Get_next_sp_with_pose Augmented_HO_Neg_HICO_pose_pattern_version2 draw_bounding_boxes_HOI_PIC _draw_single_box draw_bounding_boxes draw_bounding_boxes_HOI _load_vcoco get_overlap clip_xyxy_to_image voc_ap VCOCOeval download_file_from_google_drive parse_args getSigmoid apply_prior generate_pkl add_path main getSigmoid Generate_HICO_detection save_HICO get_blob test_net im_detect train_net SolverWrapper resnet_arg_scope ResNet50 apply_prior download_file_from_google_drive Timer draw_relation Augmented_HO_Neg_pose_pattern_version2 Augmented_HO_spNeg_pose_pattern_version2 bbox_trans Generate_action_30 Augmented_box Get_Next_Instance_HO_Neg_pose_pattern_version2 bb_IOU Get_Next_Instance_HO_spNeg_pose_pattern_version2 get_skeleton Get_next_sp Generate_action Get_Next_Instance_HO_Neg_HICO_pose_pattern_version2 Generate_action_HICO Get_next_sp_with_pose Augmented_HO_Neg_HICO_pose_pattern_version2 draw_bounding_boxes_HOI_PIC _draw_single_box draw_bounding_boxes draw_bounding_boxes_HOI _load_vcoco get_overlap clip_xyxy_to_image voc_ap VCOCOeval download_file_from_google_drive parse_args getSigmoid apply_prior generate_pkl add_path int list items join remove exists print tolist min zfill getSigmoid savemat append range len load print save_HICO makedirs open print float Generate_HICO_detection reshape astype float32 zfill shape imread DATA_DIR str print reshape get_blob append test_image_HO range len seed int toc format dump RNG_SEED iglob print average_time tic DATA_DIR im_detect open max test_image_H rstrip join remove ConfigProto makedirs get get_confirm_token save_response_content Session copy zeros bbox_trans minimum maximum concatenate reshape float64 min astype floor randint max zeros reshape zeros reshape zeros reshape line tuple zeros range len int min zeros float round range zeros bbox_trans get_skeleton concatenate Augmented_HO_Neg_pose_pattern_version2 reshape astype float32 zfill shape imread DATA_DIR len int list concatenate Augmented_box reshape min extend Generate_action sample zeros empty array range len Augmented_HO_spNeg_pose_pattern_version2 reshape astype float32 zfill shape imread DATA_DIR len int list concatenate Augmented_box reshape min extend copy Generate_action sample zeros empty array range len Augmented_HO_Neg_HICO_pose_pattern_version2 reshape astype float32 zfill shape imread DATA_DIR int list concatenate Augmented_box reshape min extend Generate_action_HICO sample zeros empty range len line Draw text rectangle ceil getsize fromarray int uint8 copy _draw_single_box round array range fromarray uint8 copy _draw_single_box round array enumerate fromarray uint8 copy _draw_single_box round array enumerate T print reshape range len minimum maximum minimum maximum concatenate size maximum sum range add_argument ArgumentParser max concatenate print reshape getSigmoid apply_prior append empty array range len insert

# TIN: Transferable Interactiveness Network #### **News**: (2022.12.19) HAKE 2.0 is accepted by TPAMI! (2022.11.19) We release the interactive object bounding boxes & classes in the interactions within AVA dataset (2.1 & 2.2)! [HAKE-AVA](https://github.com/DirtyHarryLYL/HAKE-AVA), [[Paper]](https://arxiv.org/abs/2211.07501). BTW, we also release a CLIP-based human body part states recognizer in [CLIP-Activity2Vec](https://github.com/DirtyHarryLYL/HAKE-Action-Torch/tree/CLIP-Activity2Vec)! (2022.07.29) Our new work PartMap (ECCV'22) is released! [Paper](https://github.com/enlighten0707/Body-Part-Map-for-Interactiveness/blob/main), [Code](https://github.com/DirtyHarryLYL/HAKE-Action-Torch) (2022.04.23) Two new works on HOI learning are releassed! [Interactiveness Field](https://arxiv.org/abs/2204.07718) (CVPR'22) and a new HOI metric [mPD](https://arxiv.org/abs/2202.09492) (AAAI'22). (2022.02.14) We release the human body part state labels based on AVA: [HAKE-AVA](https://github.com/DirtyHarryLYL/HAKE-AVA). (2021.2.7) Upgraded [HAKE-Activity2Vec](https://github.com/DirtyHarryLYL/HAKE-Action-Torch/tree/Activity2Vec) is released! Images/Videos --> human box + ID + skeleton + part states + action + representation. [[Description]](https://drive.google.com/file/d/1iZ57hKjus2lKbv1MAB-TLFrChSoWGD5e/view?usp=sharing) <p align='center'> <img src="https://github.com/DirtyHarryLYL/HAKE-Action-Torch/blob/Activity2Vec/demo/a2v-demo.gif", height="400"> </p>

DnanaDev/CRNN_for_OCR

['optical character recognition', 'scene text recognition']

['An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition']

CRNN Model_keras.py CRNN Model_keras_validation.py encode_to_labels ctc_lambda_func LossHistory encode_to_labels ctc_lambda_func append index enumerate

# A-CRNN-model-for-Text-Recognition-in-Keras To understand the algorithm used in the model follow these blogs: 1. https://theailearner.com/2019/05/29/creating-a-crnn-model-to-recognize-text-in-an-image-part-1/ 2. https://theailearner.com/2019/05/29/creating-a-crnn-model-to-recognize-text-in-an-image-part-2/ # Implementation for the following paper : An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition <br> https://arxiv.org/abs/1507.05717 ## Reference for CTC : https://towardsdatascience.com/intuitively-understanding-connectionist-temporal-classification-3797e43a86c ## Dataset: MjSynth - https://www.robots.ox.ac.uk/~vgg/data/text/

Dodiom/dodiom

['machine translation']

['Gamified Crowdsourcing for Idiom Corpora Construction']

src/bot/handlers/feedback.py src/nlp/turkish/translation.py src/bot/handlers/start.py src/api/review.py src/bot/helpers/keyboard_helper.py src/nlp/english/parser.py src/nlp/italian/parser.py src/models.py src/bot/handlers/help.py src/bot/handlers/review.py src/bot/helpers/feedback_helper.py src/nlp/turkish/parser.py src/initial_data.py src/bot/handlers/language.py src/bot/helpers/general.py src/bot/helpers/submission_scores.py src/api/achievements.py src/bot/handlers/announcements.py src/bot/helpers/scoreboard.py src/i18n_tokens.py src/bot/handlers/message.py src/config.py src/database.py src/bot/helpers/user_helper.py src/i18n.py src/api/submission.py src/bot/stickers.py src/bot/main.py src/api/mwe.py src/nlp/parser.py src/nlp/turkish/lemma.py src/bot/handlers/achievements.py src/bot/handlers/scoreboard.py src/log.py src/bot/helpers/tip_helper.py src/test/test_parser.py src/nlp/english/translation.py src/bot/handlers/submit.py src/bot/handlers/stats.py src/bot/helpers/notification_manager.py src/api/user.py src/nlp/language_helper.py src/main.py src/cron.py src/bot/handlers/email.py src/bot/helpers/state_helper.py src/nlp/italian/translation.py src/bot/handlers/todays_mwe.py src/nlp/parsing.py src/nlp/italian/lemma.py src/bot/handlers/report.py Config award_champion run_scheduled_jobs schedule_jobs send_game_starting_message_to_all clear_scores_for_today end_of_day_job unmute_everyone Database Language get_random_congrats_message Token load_initial_data OneLineExceptionFormatter Review FeedbackData ReviewCategory SubmissionCategory MweCategory Submission User Achievement Mwe AchievementType get_user_achievements user_has_achievement award_achievement get_date_mwe get_todays_mwe add_mwe add_review get_category_score add_submission _compound_interest update_user get_all_users mute_user add_user unmute_user get_user add_user_with_id MWExpress print_unlocked_achievements get_level achievements_handler print_locked_achievements send_game_started_again_with_awards review_happy_hour claim_email_announcement send_i_need_x_examples stats ask_for_email _clear_context check_email_and_ask_confirmation main_email_handler email_handler confirm_email feedback_handler help_handler language_change_handler change_user_language language_update_handler sticker _safe_delete_context_data message flag_submission ban_user _get_word_list_str_from_submission main_review_handler _review_answer_handler _send_submission_to_review get_submissions_to_review _process_review_achievements _safe_delete_context_data scoreboard_handler start stats submit_category_handler _clear_context main_submit_handler start_submit_handler submission_contains_todays_mwe _safe_delete_context_data _get_word_list_str_from_mwe submit_message_handler todays_mwe_handler send_feedback_url_to_user create_feedback_data_for_user send_action Keyboard NotificationManager NotificationType SentNotification ScoreBoard UserScore clear_state State set_state get_state SubmissionScores send_hint_message _get_submission_category_count get_user_from_update reply_to reply_html send_message_to_user lowercase turkish_lowercase Parser Parsed EnglishParser ItalianLemmatizer ItalianParser findPos check checkSuffixValidation turkish_lemmatize TurkishParser EnglishParserTests get bot language send_message_to_user get_all_users id TODAYS_MWE_REPLY_TEXT send_sticker info get_todays_mwe sleep GAME_STARTED unmute_everyone award_champion clear clear_scores_for_today get scoreboards bot CHAMPION_ACH_CONGRATS_MSG CHAMPION send_message_to_user TODAYS_WINNER_WITH_EMAIL id moderator email_enabled email TODAYS_WINNER_WITHOUT_EMAIL send_sticker info user_id send_message get_user award_achievement clear commit get_session get_all_users info get_all_users get_session commit info do run_pending sleep auto auto auto Column auto ForeignKey relationship Column auto ForeignKey relationship Column Column ForeignKey relationship UUID Column auto ForeignKey relationship Enum Column append type all get_session get_session commit get_session Achievement add get_session Mwe commit get_session add Review get_session commit add language commit sorted get_category_score all get_session Submission POSITIVE_TOGETHER add NEGATIVE_TOGETHER NEGATIVE_SEPARATED POSITIVE_SEPARATED get_mwe_indices _compound_interest round get_session count commit User get_session add commit User get_session add get_session get_session commit get_user get_session commit get_user get_session commit get_session get USER_DAILY_PLAY_DETAILS_MESSAGE count get_session print_unlocked_achievements score get_level reply_text UNLOCKED_ACHIEVEMENTS CHAMP_BUT_NO_EMAIL reply_html print_locked_achievements get_user_achievements LOCKED_ACHIEVEMENTS date LEVEL_MESSAGE replace send_message_to_user get_all_users sleep bot get_user_from_update start_review_happy_hour send_review_worth_more get_user_from_update get send_message_to_user get_all_users get_user_from_update GAME_STARTED_AGAIN_ANNOUNCEMENT id send_sticker info sleep bot get get_all_users get_user_from_update became_champion id CHAMP_BUT_NO_EMAIL send_message send_sticker info sleep ADDING_EMAIL ask_for_email check_email_and_ask_confirmation set_state confirm_email get_user_from_update main_email_handler get ADD_EMAIL_START reply_html info get CONFIRM_EMAIL text INVALID_EMAIL reply_html info reply_to get update_user EMAIL_SET EMAIL_CANCELLED text _clear_context ENTER_VALID_COMMAND moderator reply_html send_message info main reply_to clear_state get FEEDBACK_URL main reply_to FEEDBACK_MESSAGE get commit get_session DISCLAIMER id send_sticker HELP_MESSAGE main reply_to get language CHANGING_LANGUAGE language_selection SELECT_LANGUAGE set_state info reply_to commit get_session info get language language_selection ENGLISH TURKISH clear_state change_user_language LANGUAGE_CHANGE_SUCCESSFUL PLEASE_SELECT_VALID_LANGUAGE info main reply_to get_user_from_update banned scoreboard_handler reply_text ENTER_VALID_COMMAND id main_email_handler ban_user open main_review_handler get_state language_update_handler flag_submission game_stopped SURVEY_MESSAGE send_photo get replace language_change_handler USER_IS_BANNED_MESSAGE submission_contains_todays_mwe help_handler startswith reply_sticker info todays_mwe_handler feedback_handler achievements_handler int text main_submit_handler GAME_TEMPORARILY_STOPPED WELCOME_MESSAGE_8 file_id id send_sticker info commit get_session id send_message first commit all get_session id send_message get_user sorted all get_session get hour time GAME_HOURS_FINISHED _review_answer_handler clear_state _send_submission_to_review id REVIEWING _safe_delete_context_data set_state unmute_user reply_to clear_state id get_todays_mwe review_keyboard update_user sorted _safe_delete_context_data reply_html sleep REVIEW_QUESTION_POSITIVE REVIEW_HELP_MESSAGE_1 get language parse value NO_SUBMISSIONS reversed REVIEW_QUESTION_NEGATIVE NO_SUB_LEFT_TO_REVIEW unmute_user main REVIEW_HELP_MESSAGE_2 reply_to mwe_indices get_submissions_to_review mwe clear_state id PLEASE_ENTER_VALID_REVIEW moderator send_message mute_user REVIEW_CANCELLED SKIP review_keyboard REPORT_SUBMISSION_REPLY _send_submission_to_review _safe_delete_context_data get language user LIKE iterate unmute_user main reply_to add_review DISLIKE send_someone_liked_your_example _process_review_achievements THANKS_FOR_REVIEW language get REVIEW_LVL_3_ACH_CONGRATS_MSG get_session REVIEW_LVL_4 REVIEW_LVL_5_ACH_CONGRATS_MSG REVIEW_LVL_3 REVIEW_LVL_2 reply_html REVIEW_LVL_1_ACH_CONGRATS_MSG REVIEW_LVL_5 reply_sticker get_todays_mwe REVIEW_LVL_2_ACH_CONGRATS_MSG REVIEW_LVL_4_ACH_CONGRATS_MSG REVIEW_LVL_1 award_achievement count language get get_session send_to_user SURVEY_MESSAGE reply_text CHAMP_BUT_NO_EMAIL TODAYS_TARGET get_todays_mwe count get_user_from_update clear_state reply_text id open WELCOME_MESSAGE_1 send_sticker sleep send_photo get WELCOME_MESSAGE_5 DISCLAIMER WELCOME_MESSAGE_4 info reply_to WELCOME_MESSAGE_6 main remove WELCOME_MESSAGE_2 WELCOME_MESSAGE_8 WELCOME_MESSAGE_3 int all get_session get_user_from_update reply_text set mean round info date count get hour time GAME_HOURS_FINISHED submit_category_handler reply_to _clear_context start_submit_handler id set_state mute_user unmute_user SUBMISSION submit_message_handler language get update_user PLEASE_ENTER_EXAMPLE SUBMISSION_HELP_MESSAGE_1 _get_word_list_str_from_mwe reply_html sleep get_todays_mwe get language join get_sentence_count parse name text get_mwe_tokens PLEASE_ENTER_ONE_SENTENCE reply_html submission_category get_todays_mwe reply_to DOES_WORDS_FORM_SPECIAL_MEANING SUBMISSION_DOES_NOT_CONTAIN_MWE get_todays_mwe language parse SUB_LVL_2_ACH_CONGRATS_MSG SUB_LVL_5_ACH_CONGRATS_MSG SUB_LVL_2 EARLY_BIRD _clear_context SUB_LVL_1 SUB_LVL_1_ACH_CONGRATS_MSG id SUB_LVL_4_ACH_CONGRATS_MSG get_todays_mwe ENTER_VALID_MWE_CATEGORY count SUB_LVL_3_ACH_CONGRATS_MSG FIRST_SUB_ACH_CONGRATS_MSG FIRST_SUBMISSION reply_html add_submission sleep EARLY_BIRD_ACH_CONGRATS_MSG award_achievement get language SUB_LVL_5 SUB_LVL_3 reply_sticker unmute_user main reply_to iterate start_time send_hint_message get_session THANKS_FOR_SUBMISSION now SUB_LVL_4 SUBMISSION_CANCELLED submission_category combine _safe_delete_context_data get hour time GAME_HOURS_FINISHED language update_user reply_text reply_html sleep get_todays_mwe reply_to commit FeedbackData get_session reviews now add submissions len get reply_text id create_feedback_data_for_user FEEDBACK_URL FEEDBACK_MESSAGE auto auto set_state NONE info info set_state NONE get_session get HINT_MESSAGE_1 HINT_MESSAGE_3 reply_text HINT_MESSAGE_2 choice reply_html append HINT_MESSAGE_4 commit get_session id username get_user send_message reply_text info reply_text info reversed startswith insert check append range len append range len append findPos lower

# Dodiom Code for the Telegram bot [Dodiom](https://t.me/mwetest_bot). ## Setup Create a new bot using [BotFather](https://t.me/botfather), get its token and put it in `docker/english/docker-compose.yml`, also put "1" for moderator id there (you can change this with your own Telegram ID after you find it) and then run ```sh cd docker/english docker-compose build # might take some time docker-compose up -d docker-compose down docker-compose up -d # restart second time for initial data changes to be applied

Dong-JinKim/ActionCooccurrencePriors

['human object interaction detection']

['Detecting Human-Object Interactions with Action Co-occurrence Priors', 'ACP++: Action Co-occurrence Priors for Human-Object Interaction Detection', 'Detecting Human-Object Interactions with Action Co-occurrence Priors']

find_correlation.py utils/io.py exp/hoi_classifier/data/box_features.py data/hico/split_ids.py exp/hoi_classifier/data/assign_pose_to_human_candidates.py exp/hico_eval/sample_complexity_analysis.py data/hico/hoi_cls_count.py exp/hoi_classifier/vis/top_boxes_per_hoi.py exp/hoi_classifier/vis/faster_rcnn_aps.py exp/hoi_classifier/data/hoi_candidates.py exp/hoi_classifier/data/label_hoi_candidates.py exp/hoi_classifier/models/verb_given_object_appearance.py utils/losses.py utils/constants.py exp/hoi_classifier/data/cache_box_features.py exp/run_template.py exp/detect_coco_objects/select_confident_boxes.py exp/hoi_classifier/data/write_faster_rcnn_feats_to_hdf5.py data/hico/hico_constants.py exp/detect_coco_objects/prepare_data_for_faster_rcnn.py exp/hoi_classifier/vis/vis_object_aps_per_interaction.py exp/hoi_classifier/models/verb_given_human_appearance.py exp/hoi_classifier/data/features_dataset.py exp/detect_coco_objects/evaluate_boxes.py exp/hoi_classifier/eval.py exp/hoi_classifier/train.py utils/model.py data/coco_classes.py exp/hico_eval/compute_map.py utils/bbox_utils.py exp/hoi_classifier/vis/vis_human_pose.py utils/pytorch_layers.py exp/hoi_classifier/models/verb_given_human_pose.py exp/detect_coco_objects/run.py exp/hoi_classifier/models/scatter_verbs_to_hois.py exp/hoi_classifier/vis/vis_interaction_aps_per_object.py exp/hoi_classifier/models/verb_given_boxes_and_object_label.py utils/argparse_utils.py exp/experimenter.py exp/hoi_classifier/data/pose_features.py utils/html_writer.py exp/hoi_classifier/data/cache_pose_features.py exp/hoi_classifier/run.py exp/hoi_classifier/models/hoi_classifier_model.py data/hico/mat_to_json.py HicoConstants bin_hoi_ids main ConvertMat2Json main split list_exps exp_do_something evaluate_boxes box_recall evaluate_boxes_and_labels box_label_recall prepare_hico exp_select_and_evaluate_confident_boxes_in_hico exp_detect_coco_objects_in_hico select_det_ids select_dets select compute_ap eval_hoi compute_pr compute_normalized_pr load_gt_dets match_hoi main main compute_mAP main eval_model exp_top_boxes_per_hoi exp_eval exp_cache_pose_feats exp_assign_pose_to_human_cand exp_gen_and_label_hoi_cand exp_train exp_cache_box_feats train_model main eval_model assign_pose count_keypoints_in_box main get_pose_box BoxFeatures main compute_box_feats main Features FeatureConstants HoiCandidatesGenerator generate load_gt_dets assign match_hoi PoseFeatures main MTL HoiClassifier HoiClassifierConstants ScatterClusterToHois ScatterVerbsToHois ScatterVerbsToHoisConstants VerbGivenBoxesAndObjectLabel VerbGivenBoxesAndObjectLabelConstants VerbGivenHumanAppearance VerbGivenHumanAppearanceConstants VerbGivenHumanPose VerbGivenHumanPoseConstants VerbGivenObjectAppearance VerbGivenObjectAppearanceConstants create_html get_gt_boxes get_gt_hois select_best_boxes_across_dataset main vis_keypts main main main str_to_bool manage_required_args vis_sub_obj_bboxes compute_area_batch join_bboxes_by_line add_bbox compute_iou compute_area compute_iou_batch vis_human_keypts vis_bboxes vis_bbox ExpConstants Constants save_constants HtmlWriter load_pickle_object load_json_object read dump_json_object load_mat_object dump_pickle_object deserialize_object write load_yaml_object mkdir_if_not_exists dumps_json_object JsonSerializableClass serialize_object WritableToFile NumpyAwareJSONEncoder FocalLoss Model MLP adjust_learning_rate get_activation Identity create_mlp append items list load_json_object join anno_list_json dump_json_object proc_dir HicoConstants tqdm bin_hoi_ids int sample set append len list items print len split print parse_args exp print enumerate compute_iou len zip enumerate compute_iou len join load_json_object anno_list_json list dump_json_object concatenate print File tolist iou_thresh tqdm exp_dir append box_recall keys enumerate hoi_list_json join load_json_object anno_list_json list dump_json_object concatenate print File tolist iou_thresh box_label_recall tqdm exp_dir append keys enumerate join load_json_object anno_list_json dump_json_object proc_dir print mkdir_if_not_exists dict exp_dir images_dir to_json enumerate len ExpConstants prepare_hico HicoConstants evaluate_boxes evaluate_boxes_and_labels select HicoConstants ExpConstants arange min astype compute_area append array range background_score_thresh select_det_ids max_humans max_objects_per_class object_score_thresh concatenate max_background human_score_thresh append zeros array enumerate load join load_json_object anno_list_json print faster_rcnn_boxes File mkdir_if_not_exists select_dets close tqdm exp_dir create_dataset to_json create_group compute_iou enumerate isnan any max arange cumsum array nan cumsum sum array nan compute_ap save ylabel match_hoi ylim title savefig append format compute_pr close xlim join print xlabel File figure fill_between step print join load_json_object append starmap load_gt_dets num_processes mkdir_if_not_exists close set out_dir append parse_args Pool sorted compute_mAP bin_to_hoi_ids_json keys cuda FloatTensor SequentialSampler create_group load_json_object concatenate close masked_fill_ eval float load join print File tqdm t exp_dir hoi_classifier create_dataset numpy load model_pth Features eval_model Model load_state_dict cuda join print getcwd subset HicoConstants manage_required_args gen_hoi_cand exp_dir assign generate parse_args ExpConstants label_hoi_cand join subset HicoConstants manage_required_args exp_dir main parse_args ExpConstants join proc_dir subset HicoConstants manage_required_args exp_dir main parse_args ExpConstants join subset HicoConstants manage_required_args exp_dir main parse_args ExpConstants join verb_given_boxes_and_object_label verb_given_object_appearance getcwd Constants rcnn_det_prob manage_required_args exp_dir HoiClassifierConstants imgs_per_batch verb_given_human_pose FeatureConstants verb_given_appearance main parse_args ExpConstants verb_given_human_appearance join verb_given_boxes_and_object_label verb_given_object_appearance getcwd Constants model_num rcnn_det_prob manage_required_args exp_dir model_dir HoiClassifierConstants verb_given_human_pose FeatureConstants verb_given_appearance main parse_args ExpConstants verb_given_human_appearance join verb_given_boxes_and_object_label getcwd Constants model_num rcnn_det_prob manage_required_args exp_dir model_dir HoiClassifierConstants verb_given_human_pose FeatureConstants verb_given_appearance main parse_args ExpConstants data zero_grad model_dir save cuda FloatTensor criterion_softmax Adam matmul chain sum CrossEntropyLoss range state_dict load_json_object format eval_model LongTensor mean float BCELoss num_epochs enumerate load join backward print Variable RandomSampler t parameters hoi_classifier train step criterion Variable size RandomSampler manual_seed BCELoss train_model log_dir exp_dir model_dir save_constants to_txt amin array amax zeros enumerate compute_iou compute_area str assign_pose hoi_cand_hdf5 range create_group split_ids_json num_keypoints int File human_pose_dir create_dataset compute_features BoxFeatures array tile compute_box_feats PoseFeatures rpn_id_to_pose_h5py_to_npy compute_pose_feats array tile human_cands_pose_hdf5 load_json_object join value create_group split_ids_json print File mkdir_if_not_exists close tqdm exp_dir save_constants create_dataset selected_dets_hdf5 HoiCandidatesGenerator predict enumerate set hoi_list_json join load_json_object anno_list_json split_ids_json print load_gt_dets File mkdir_if_not_exists hoi_cand_hdf5 close match_hoi tqdm exp_dir save_constants create_dataset zeros range faster_rcnn_boxes concatenate len append zeros enumerate vis_human_keypts tile list sorted reshape get_gt_boxes zfill tqdm append zeros num_to_vis keys range enumerate vis_sub_obj_bboxes join deepcopy list add_element mkdir_if_not_exists close zfill get_gt_hois HtmlWriter tqdm tile imread keys imsave enumerate vis_keypts hoi_list_json create_html human_pose_feats_hdf5 select_best_boxes_across_dataset pred_hoi_dets_h5py images_dir vis_human_keypts human_pose_feats_h5py add hoi_cand_h5py imread imsave enumerate reshape Box plot getcwd Scatter Layout join sorted print exit set getattr choices append help polygon polygon_perimeter set_color min max compute_area zeros logical_and minimum compute_area_batch stack maximum min copy set_color polygon polygon_perimeter max range copy vis_bbox min copy line_aa max range circle vis_bboxes join_bboxes_by_line zip line_aa circle range copy join items list print to_json decompress read loads compress write dumps encode compress write dumps dumps mkdir exists makedirs get_activation MLP param_groups

# Action Co-occurrence Priors for HOI Detection Official code for our ECCV 2020 paper, **[Detecting Human-Object Interactions with Action Co-occurrence Priors](https://sites.google.com/view/action-cooccurrence)**. Done by Dong-Jin Kim, Xiao Sun, Jinsoo Choi, Stephen Lin, and In So Kweon. We Introduce novel "action co-occurrence priors" to improve state-of-the-art performance of Human-Object Interaction (HOI) detectors. <p align="center"><img src='imgs/ACP_teaser.png' width="60%" ></p> The figure shows the marginal/conditional probability values computed from the distribution of the training label. Intuitively, detection of rarely labeled HOIs (operate-hair dryer) can be facilitated by detection of commonly co-occurring HOIs (hold-hair dryer). Also, non-detection of rare HOIs (blow-cake) can be aided by detection of incompatible HOIs (cut-cake). We leverage this intuition as a prior to learn an HOI detector effective on long-tailed datasets. <img src='imgs/ACP_matrix.png'> Examples of co-occurrence matrices constructed for several objects. Along the Y-axis is the given action, and the X-axis enumerates conditional actions. Each element represents the conditional probability that an action occurs when another action is happening. # Requirements

DongHande/MCGL

['graph learning', 'data augmentation']

['Data Augmentation View on Graph Convolutional Network and the Proposal of Monte Carlo Graph Learning']

plot/comparison_clean.py plot/comparison_noisy.py train_GCN.py train_MC_base.py plot/deep_GCO.py ps.py models.py layers.py data/coauthorship/sparsegraph.py utils.py data/coauthorship/io.py Avelayer Conlayer Linear graph_ave GCN MLP parse_args train test train test get_noise_rate load_data_ms_academic sparse_mx_to_torch_sparse_tensor accuracy reduce_noise parse_index_file load_data adj_to_graph graph_sample normalize_features write_file networkx_to_sparsegraph load_from_npz load_dataset remove_self_loops SparseGraph create_subgraph largest_connected_components plot_points_and_lines plot_boundary_and_set_ticklabels create_and_split points plot_row split_community plot_all MCGL create_and_split plot_row GCN plot_row create_and_split plot_all GCN ones size matmul div is_available cuda add_argument ArgumentParser time model backward print nll_loss zero_grad accuracy eval step model print nll_loss accuracy eval graph_ave batch_size tsdep trdep graph_sample range graph_ave tsdep range append int strip open todense LongTensor FloatTensor sparse_mx_to_torch_sparse_tensor adj_matrix concatenate labels attr_matrix choice eye load_dataset normalize_features array range append len from_dict_of_lists tuple parse_index_file vstack max list todense FloatTensor tolist normalize_features range format lil_matrix LongTensor adjacency_matrix print sort tolil min sparse_mx_to_torch_sparse_tensor eye zeros array len diags flatten dot sum array sum type_as double data Size astype float32 from_numpy shape int64 append T numpy range append int range len LongTensor concatenate FloatTensor ones transpose size append numpy transpose numpy Path isinstance exists update data list remove all lil_matrix items tocsr convert_node_labels_to_integers nodes set keys adjacency_matrix zeros array sorted bincount adj_matrix connected_components setdiag tocsr format tolil warn sum append uniform range pi append sqrt power range max seed list transpose points range normal concatenate choice set split_community sqrt zeros sort min eye amin array amax len plot set_xlim scatter range set_ylim plot set_xticklabels set_yticklabels set_yticks cos pi sqrt set_xticks twinx linspace sin set_ylim str list plot_points_and_lines ones text plot_boundary_and_set_ticklabels add_subplot divide matmul set sqrt create_and_split append array range random ones divide matmul copy sqrt range find_one set plot_all MCGL GCN plot set_yticklabels set_yticks set_xlim scatter set_xticks twinx range set_ylim text

# Data Augmentation View on Graph Convolutional Network and the Proposal of Monte Carlo Graph Learning This repository is the official implementation of [Data Augmentation View on Graph Convolutional Network and the Proposal of Monte Carlo Graph Learning](https://arxiv.org/abs/2006.13090). ## Overview In this project, we provide implementations of visual analysis and three models -- GCN, GCN* and MCGL-UM. The repository is organised as follows: - `data/` contains the necessary dataset files for CORA, PubMed, CiteSeer and MS Academic. - `plot/` contains the visual analysis on synthetic clean and noisy graphs.

DongjuSin/computer-vision-proj

['semantic segmentation']

['FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation']

encoding/parallel.py encoding/dilated/resnet.py encoding/utils/metrics.py encoding/functions/encoding.py encoding/models/encnet.py encoding/nn/syncbn.py scripts/prepare_pcontext.py encoding/models/psp.py encoding/functions/__init__.py encoding/models/model_store.py encoding/datasets/__init__.py encoding/nn/__init__.py encoding/models/fcn.py encoding/datasets/pascal_voc.py encoding/utils/files.py encoding/utils/lr_scheduler.py encoding/datasets/cityscapes.py encoding/models/__init__.py experiments/segmentation/loss.py experiments/segmentation/test.py scripts/prepare_pascal.py encoding/utils/__init__.py scripts/prepare_cityscapes.py encoding/datasets/base.py encoding/models/model_zoo.py encoding/lib/gpu/setup.py experiments/segmentation/option.py encoding/datasets/ade20k.py encoding/datasets/pascal_aug.py encoding/datasets/coco.py encoding/lib/cpu/setup.py encoding/lib/__init__.py experiments/segmentation/test_fps_params.py experiments/segmentation/train.py encoding/models/base.py scripts/prepare_coco.py scripts/prepare_ade20k.py encoding/utils/pallete.py encoding/dilated/__init__.py encoding/nn/comm.py encoding/__init__.py encoding/nn/customize.py encoding/nn/encoding.py setup.py encoding/functions/syncbn.py encoding/datasets/pcontext.py encoding/models/deeplabv3.py create_version_file develop install CallbackContext allreduce AllReduce DataParallelModel _criterion_parallel_apply execute_replication_callbacks Reduce DataParallelCriterion ADE20KSegmentation _get_ade20k_pairs BaseDataset test_batchify_fn get_city_pairs CitySegmentation COCOSegmentation VOCAugSegmentation VOCSegmentation ContextSegmentation get_segmentation_dataset ResNet resnet50 Bottleneck resnet152 conv3x3 resnet34 resnet18 BasicBlock resnet101 scaled_l2 _scaled_l2 aggregate _aggregate _batchnormtrain batchnormtrain _sum_square sum_square module_inference MultiEvalModule BaseNet flip_image resize_image pad_image crop_image ASPP_Module DeepLabV3 get_deeplab AsppPooling ASPPConv DeepLabV3Head EncHead get_encnet_resnet101_ade EncNet get_encnet_resnet50_ade EncModule get_encnet_resnet101_pcontext get_encnet_resnet152_ade get_encnet get_encnet_resnet50_pcontext get_fcn_resnet50_pcontext FCNHead get_fcn_resnet50_ade FCN get_fcn get_model_file short_hash purge pretrained_model_list get_model PSPHead PSP get_psp_resnet50_ade get_psp get_segmentation_model SyncMaster FutureResult SlavePipe SeparableConv2d Normalize SegmentationLosses PyramidPooling Mean JPU Encoding BatchNorm3d SharedTensor _SyncBatchNorm BatchNorm1d BatchNorm2d download save_checkpoint mkdir check_sha1 LR_Scheduler SegmentationMetric batch_pix_accuracy batch_intersection_union get_mask_pallete _get_voc_pallete ImageBasedCrossEntropyLoss2d CrossEntropyLoss2d get_loss JointEdgeSegLoss Options test Trainer parse_args download_ade download_city parse_args parse_args install_coco_api download_coco parse_args download_voc download_aug install_pcontext_api parse_args download_ade print join join isinstance _worker len start is_grad_enabled append range Lock list hasattr __data_parallel_replicate__ modules enumerate len print join get_path_pairs list isinstance zip print join get_path_pairs load_url ResNet load_state_dict load_url ResNet load_state_dict load ResNet load_state_dict get_model_file load ResNet load_state_dict get_model_file load ResNet load_state_dict get_model_file flip_image evaluate size resize_ pad array range norm_layer Sequential ReLU Conv2d NUM_CLASS DeepLabV3 load EncNet get_model_file load_state_dict NUM_CLASS load get_model_file load_state_dict NUM_CLASS FCN get join remove format print check_sha1 expanduser download exists makedirs join remove endswith expanduser listdir lower load PSP get_model_file load_state_dict NUM_CLASS copyfile save makedirs get join isdir print dirname abspath expanduser makedirs sha1 makedirs sum max astype dtype max histogram astype fromarray lower astype putpalette range joint_edgeseg_loss cuda img_wt_loss num_class model save_folder get_mask_pallete DataLoader get_segmentation_model save dataset cuda get_segmentation_dataset load_state_dict format model_zoo Compose eval resume zip enumerate load join SegmentationMetric print tqdm get_model makedirs add_argument ArgumentParser join download mkdir print join mkdir download mkdir rmtree system join download mkdir join download mkdir basename move rmtree system

# FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation [[Project]](http://wuhuikai.me/FastFCNProject/) [[Paper]](http://wuhuikai.me/FastFCNProject/fast_fcn.pdf) [[arXiv]](https://arxiv.org/abs/1903.11816) [[Home]](http://wuhuikai.me) [](https://paperswithcode.com/sota/semantic-segmentation-pascal-context?p=fastfcn-rethinking-dilated-convolution-in-the) Official implementation of **FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation**. A **Faster**, **Stronger** and **Lighter** framework for semantic segmentation, achieving the state-of-the-art performance and more than **3x** acceleration. ``` @inproceedings{wu2019fastfcn, title = {FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation}, author = {Wu, Huikai and Zhang, Junge and Huang, Kaiqi and Liang, Kongming and Yu Yizhou}, booktitle = {arXiv preprint arXiv:1903.11816},

DonkeyShot21/uis-rnn-sml

['speaker diarization']

['Supervised online diarization with sample mean loss for multi-domain data']

demo.py uisrnn/utils.py uisrnn/uisrnn.py uisrnn/__init__.py uisrnn/evals.py uisrnn/arguments.py uisrnn/loss_func.py main diarization_experiment str2bool parse_arguments compute_sequence_match_accuracy get_list_inverse_index regularization_loss weighted_mse_loss sigma2_prior_loss CoreRNN BeamState UISRNN enforce_cluster_id_uniqueness output_result concatenate_training_data estimate_transition_bias sample_permuted_segments resize_sequence estimate_crp_alpha Logger pack_sequence generate_random_string compute_sequence_match_accuracy save share_memory Pool list tolist imap append range output_result SummaryWriter close enumerate load items print fit UISRNN epochs add_scalar parse_arguments diarization_experiment parse_known_args parse_args add_argument ArgumentParser dict enumerate sorted linear_sum_assignment get_list_inverse_index set zip zeros sum len float mm diag view ndarray isinstance tolist append generate_random_string enforce_cluster_id_uniqueness list concatenate shuffle shape zip enumerate permutation concatenate append range len sample_permuted_segments where unique append range len zeros_like sort pack_padded_sequence randint choice mean zeros to range len rnn_dropout format learning_rate batch_size sigma_alpha mean rnn_depth rnn_hidden_size crp_alpha zip sigma_beta regularization_weight range len sum range len

# Better, Faster, Stronger UIS-RNN This repository implements some useful features on top of the original [UIS-RNN repository](https://github.com/google/uis-rnn). Some of them are described in the following paper: [Supervised Online Diarization with Sample Mean Loss for Multi-Domain Data](https://arxiv.org/abs/1911.01266). Here is a list: * **Sample Mean Loss (SML)**, a loss function that improves performance and training efficiency. To learn more about it you can read our paper. * **Estimation of** `crp_alpha`, a parameter of the distance dependent Chinese Restaurant Process (ddCRP) that determines the probability of switching to a new speaker. Again, more info in our paper. * **Parallel prediction** using `torch.multiprocessing`, that mitigates the issue with slow decoding and enables higher GPU usage. * **Tensorboard** logging, for visualizing training. Here is a diagram of the Sample Mean Loss: <p align="center"> <img src="./resources/SML_diag.png" width="450">

DoodleJZ/ParsingAll

['semantic parsing']

['Parsing All: Syntax and Semantics, Dependencies and Spans']

src/Evaluator/dep_eval.py srl_scripts/filter_conll2012_data.py srl_scripts/cache_elmo.py src/pretrained_bert/tokenization.py src/Decoder/uniform_decoder.py src/utils_io.py src/Evaluator/conll09_utils.py src/Datareader/srlspan_reader.py src/transliterate.py src/pretrained_bert/file_utils.py srl_scripts/conll05_to_json.py srl_scripts/conll09_to_json.py src/Evaluator/pos_eval.py srl_scripts/ontonotes5_to_json.py src/trees.py src/Zparser.py src/pretrained_bert/convert_tf_checkpoint_to_pytorch.py srl_scripts/json_to_conll09.py src/makehp.py src/vocabulary.py src/Evaluator/evaluate.py src/utils.py src/Evaluator/srl_eval.py src/pretrained_bert/modeling.py src/pretrained_bert/__main__.py srl_scripts/get_char_vocab.py src/Datareader/syndep_reader.py srl_scripts/make_conll09_gold_props.py srl_scripts/filter_embeddings.py src/main.py src/Datareader/srldep_reader.py src/pretrained_bert/optimization.py src/pretrained_bert/__init__.py format_elapsed count_wh torch_load make_align correct_sent make_hparams main align_sent run_train span_miss_verb run_test HParams arabic hebrew InternalParseNode LeafParseNode load_trees ParseNode LeafTreebankNode InternalTreebankNode TreebankNode load_embedding_dict Vocabulary SrlDep read_srldep CoNLLXReader SrlInstance CoNLLXReader read_srlspan read_syndep DependencyInstance CoNLLXReader save_conll merge_data read_conll eval is_punctuation is_uni_punctuation FScore evalb eval compute_srl_f1 compute_unlabeled_span_f1 print_to_span_conll read_gold_predicates split_example_for_eval _print_f1 compute_span_f1 print_sentence_to_conll print_dependency_conll compute_dependency_f1 evaluate_retrieval convert_tf_checkpoint_to_pytorch cached_path s3_etag http_get s3_request s3_get read_set_from_file get_from_cache filename_to_url url_to_filename split_s3_path get_file_extension BertPreTrainingHeads BertForQuestionAnswering BertEncoder PreTrainedBertModel BertSelfAttention BertForMaskedLM BertOnlyMLMHead BertOnlyNSPHead BertEmbeddings BertOutput BertPredictionHeadTransform BertAttention BertPooler gelu BertForMultipleChoice BertConfig BertLayer BertForTokenClassification BertModel BertForNextSentencePrediction BertIntermediate BertForSequenceClassification BertForPreTraining swish BertLMPredictionHead BertSelfOutput warmup_cosine warmup_constant warmup_linear BertAdam BasicTokenizer WordpieceTokenizer load_vocab whitespace_tokenize _is_whitespace _is_control BertTokenizer _is_punctuation main cache_dataset build_elmo DocumentState flatten normalize_word minimize_partition handle_line handle_bit conll09_to_json convert_sent_dict filter_data get_char_vocab json_to_conll09 make_gold_props get_doc_key DocumentState flatten normalize_word minimize_partition handle_line handle_bit use_cuda divmod int time format pop InternalParseNode children isinstance print extend reversed enumerate print dump open enumerate append items zip enumerate print zip from_spec children ReduceLROnPlateau seed srlspan_train_ptb_path print_vocabs Adam CHAR_STOP_WORD load_state_dict format_elapsed max_len_dev tag load_trees manual_seed enumerate time learning_rate CHAR_START_WORD print_vocabulary model_name step punctuation check_dev zero_grad max make_align word CHAR_STOP_SENTENCE srlspan_dev_ptb_path ceil Vocabulary learning_rate_warmup_steps shuffle set srldep_dev_ptb_path float isinstance index CHAR_UNK train read_syndep count_wh START tuple read_srldep clip_grad_norm_ CHAR_START_SENTENCE checks_per_epoch count chr syndep_train_ptb_path split_batch parse_batch range set_from_args synconst_train_ptb_path label type pop correct_sent print extend synconst_dev_ptb_path numpy InternalParseNode batch_size sorted torch_load max_len_train schedule_lr freeze STOP TAG_UNK UNK step_decay format subbatch_max_tokens numpy_seed reversed align_sent syndep_dev_ptb_path srldep_train_ptb_path ChartParser items int read_srlspan backward randint len from_spec read_syndep synconst_test_ptb_path read_srldep srlspan_test_brown_path eval_batch_size torch_load use_gold_predicate parse_batch compute_dependency_f1 evalb evalb_dir range format_elapsed compute_srl_f1 format srlspan_test_ptb_path load_trees eval srldep_test_ptb_path syndep_test_ptb_path time read_srlspan print extend model_path_base srldep_test_brown_path len callback populate_arguments add_argument make_hparams add_parser ArgumentParser parse_args set_defaults add_subparsers join process_NONE helper split append enumerate print dict items list getNext syndep_heads gold_verb print length words close pred_pos append gold_srl CoNLLXReader srl_label_start max list items getNext syndep_heads gold_verb srlabel print length words close gold_pos pred_pos append gold_srl CoNLLXReader len getNext print length words close types append postags heads CoNLLXReader append split append range len match print range len join list format cleanup name print leaves FScore nan zip exists TemporaryDirectory run zip append sum enumerate update items list print set zip len print format update Counter _print_f1 range len update Counter _print_f1 range len items list format print zip update items list print_to_span_conll format str read_gold_predicates print float Counter getpid zip _print_f1 Popen write range len strip close open append split sorted print_sentence_to_conll close open keys enumerate sorted print_sentence_to_conll close open keys range enumerate abspath save from_json_file str transpose from_numpy getattr list_variables append state_dict format zip load_variable join int print BertForPreTraining fullmatch any split encode hexdigest sha256 str join isinstance str urlparse isinstance exists path netloc urlparse startswith resource split_s3_path Object resource split_s3_path download_fileobj get update write close tqdm iter_content len get str s3_etag join isinstance url_to_filename startswith head makedirs set OrderedDict strip split category category startswith startswith category ord pop print convert_tf_checkpoint_to_pytorch Module placeholder stack string int32 elmo_module append add pop find pop max int const_stack assert_finalizable handle_bit text normalize_word ner_stack finalize split append ner_buffer sum finalize_sentence enumerate const_buffer len print format append range len append split print format list sorted format print set len append split append split assert_empty group startswith get_doc_key match

# ParsingAll ## Contents 1. [Requirements](#Requirements) 2. [Training](#Training) ## Requirements * Python 3.6 or higher. * Cython 0.25.2 or any compatible version. * [PyTorch](http://pytorch.org/) 1.0.0+. * [EVALB](http://nlp.cs.nyu.edu/evalb/). Before starting, run `make` inside the `EVALB/` directory to compile an `evalb` executable. This will be called from Python for evaluation. * [pytorch-transformers](https://github.com/huggingface/pytorch-transformers) PyTorch 1.0.0+ or any compatible version.

DoubleML/doubleml-for-py

['causal inference']

['Double/Debiased Machine Learning for Treatment and Causal Parameters']

doubleml/tests/test_pliv.py doubleml/tests/test_irm_no_cross_fit.py doubleml/tests/test_plr_set_smpls_externally.py doubleml/_utils.py doubleml/tests/test_pliv_partial_x.py doubleml/tests/_utils_dml_cv_predict.py doubleml/tests/test_doubleml_set_sample_splitting.py doubleml/tests/test_pliv_partial_xz_tune.py doubleml/tests/test_doubleml_model_defaults.py doubleml/tests/test_plr.py doubleml/tests/test_iivm_classifier.py doubleml/tests/test_pliv_partial_z.py doubleml/tests/_utils_pliv_partial_x_manual.py doubleml/__init__.py doubleml/tests/test_irm_tune.py doubleml/tests/test_datasets.py doubleml/tests/test_irm_classifier.py doubleml/tests/_utils_iivm_manual.py doubleml/tests/test_iivm_no_cross_fit.py doubleml/double_ml_pliv.py doubleml/tests/test_plr_classifier.py doc/conf.py doubleml/tests/test_plr_reestimate_from_scores.py doubleml/tests/test_plr_no_cross_fit.py doubleml/tests/test_doubleml_exceptions.py doubleml/tests/test_dml_data.py doubleml/double_ml_irm.py doubleml/tests/_utils_plr_manual.py doubleml/double_ml.py doubleml/tests/test_package.py doubleml/tests/test_pliv_tune.py doubleml/tests/test_irm.py doubleml/tests/test_doubleml_return_types.py doubleml/tests/conftest.py doubleml/double_ml_iivm.py doubleml/double_ml_data.py doubleml/tests/test_pliv_partial_z_tune.py doubleml/tests/test_plr_multi_treat.py doubleml/tests/_utils.py doubleml/tests/_utils_boot.py doubleml/tests/_utils_cluster.py doubleml/tests/test_pliv_partial_xz.py doubleml/tests/test_iivm_subgroups.py doubleml/tests/_utils_pliv_partial_xz_manual.py doubleml/_utils_resampling.py doubleml/tests/test_pliv_no_cross_fit.py doubleml/tests/_utils_pliv_partial_z_manual.py doubleml/tests/test_doubleml_scores.py doubleml/datasets.py doubleml/double_ml_plr.py doubleml/tests/test_irm_with_missings.py doubleml/tests/test_plr_set_ml_nuisance_pars.py doubleml/tests/test_cv_predict.py doubleml/tests/test_iivm.py doubleml/tests/_utils_irm_manual.py setup.py doubleml/tests/test_iivm_tune.py doubleml/tests/test_multiway_cluster.py doubleml/tests/test_plr_tune.py doubleml/tests/test_pliv_partial_x_tune.py doubleml/tests/test_plr_rep_cross.py doubleml/tests/_utils_pliv_manual.py make_irm_data make_iivm_data _make_pliv_data fetch_401K fetch_bonus _m _g make_pliv_multiway_cluster_CKMS2021 make_plr_turrell2018 make_pliv_CHS2015 make_plr_CCDDHNR2018 DoubleML DoubleMLClusterData DoubleMLData DoubleMLIIVM DoubleMLIRM DoubleMLPLIV DoubleMLPLR _check_smpl_split_tpl _get_cond_smpls _check_all_smpls _draw_weights _check_finite_predictions _assure_2d_array _check_is_partition _fit _dml_tune _dml_cv_predict _check_smpl_split DoubleMLClusterResampling DoubleMLResampling ResampleNoSplit learner test_dml_irm_boot score trimming_threshold dml_procedure test_dml_irm_se dml_irm_fixture test_dml_irm_coef learner test_dml_plr_se test_dml_plr_ols_manual_coef score test_dml_plr_boot test_dml_plr_ols_manual_se dml_plr_ols_manual_fixture dml_procedure test_dml_plr_coef dml_plr_fixture test_dml_plr_ols_manual_boot generate_data1 generate_data_iivm_binary generate_data2 generate_data_pliv_partialXZ generate_data_iv generate_data_bivariate generate_data_irm_binary generate_data_pliv_partialZ make_data_pliv_partialZ generate_data_cv_predict _m generate_data_irm generate_data_irm_w_missings _m2 _g generate_data_iivm generate_data_pliv_partialX generate_data_toeplitz test_cv_predict params cv_predict_fixture cross_fit test_fetch_bonus_poly test_make_pliv_CHS2015_return_types test_make_plr_turrell2018_return_types test_fetch_401K_poly test_make_plr_CCDDHNR2018_return_types test_fetch_bonus_return_types test_fetch_401K_return_types test_make_irm_data_return_types test_make_iivm_data_return_types test_make_pliv_multiway_cluster_CKMS2021_return_types test_make_pliv_data_return_types test_dml_data_y test_dml_data_no_instr test_d_cols_setter test_add_vars_in_df test_x_cols_setter_defaults test_x_cols_setter test_dml_data_x test_disjoint_sets test_duplicates test_obj_vs_from_arrays test_z_cols_setter test_cluster_cols_setter test_dml_data_w_missings test_use_other_treat_as_covariate test_dml_datatype dml_data_fixture test_x_cols_setter_defaults_w_cluster test_y_col_setter test_dml_data_d test_doubleml_exception_scores test_doubleml_exception_confint test_doubleml_exception_resampling _DummyNoClassifier test_doubleml_cluster_not_yet_implemented test_doubleml_exception_fit test_doubleml_exception_trimming_rule test_doubleml_exception_set_ml_nuisance_params test_doubleml_exception_subgroups test_doubleml_exception_tune test_doubleml_exception_smpls LogisticRegressionManipulatedPredict test_doubleml_exception_get_params test_doubleml_exception_bootstrap test_doubleml_exception_no_cross_fit test_doubleml_exception_and_warning_learner LassoWithNanPred test_doubleml_exception_dml_procedure test_doubleml_exception_data LassoWithInfPred test_doubleml_nan_prediction test_doubleml_exception_learner _DummyNoGetParams test_doubleml_warning_crossfitting_onefold test_doubleml_exception_p_adjust _DummyNoSetParams test_pliv_defaults test_plr_defaults test_irm_defaults _assert_resampling_default_settings test_iivm_defaults test_property_types_and_shapes test_return_types test_pliv_callable_vs_str_score test_plr_callable_vs_str_score test_iivm_callable_vs_str_score test_irm_callable_vs_pred_export test_pliv_callable_not_implemented test_linear_score test_iivm_callable_vs_pred_export test_irm_callable_vs_str_score test_plr_callable_vs_pred_export test_doubleml_set_sample_splitting_invalid_sets test_doubleml_set_sample_splitting_tuple test_doubleml_draw_vs_set test_doubleml_set_sample_splitting_all_tuple _assert_smpls_equal _assert_resampling_pars test_doubleml_set_sample_splitting_all_list learner test_dml_iivm_se score trimming_threshold test_dml_iivm_coef dml_procedure test_dml_iivm_boot dml_iivm_fixture learner dml_iivm_classifier_fixture test_dml_iivm_se score trimming_threshold test_dml_iivm_coef dml_procedure test_dml_iivm_boot learner test_dml_iivm_no_cross_fit_coef dml_iivm_no_cross_fit_fixture score test_dml_iivm_no_cross_fit_boot n_folds test_dml_iivm_no_cross_fit_se learner test_dml_iivm_subgroups_coef test_dml_iivm_subgroups_se score trimming_threshold dml_procedure test_dml_iivm_subgroups_boot dml_iivm_subgroups_fixture subgroups test_dml_iivm_subgroups learner_m test_dml_iivm_se score test_dml_iivm_coef tune_on_folds dml_procedure learner_g test_dml_iivm_boot get_par_grid learner_r subgroups dml_iivm_fixture learner test_dml_irm_boot score trimming_threshold dml_procedure test_dml_irm_se test_dml_irm_coef dml_irm_classifier_fixture learner test_dml_irm_no_cross_fit_boot score n_folds test_dml_irm_no_cross_fit_coef test_dml_irm_no_cross_fit_se dml_irm_no_cross_fit_fixture learner_m test_dml_irm_boot score tune_on_folds dml_procedure learner_g test_dml_irm_se get_par_grid dml_irm_fixture test_dml_irm_coef learner_xgboost score learner_sklearn trimming_threshold test_dml_irm_w_missing_coef dml_procedure test_dml_irm_w_missing_se test_dml_irm_w_missing_boot dml_irm_w_missing_fixture test_irm_exception_with_missings learner dml_pliv_multiway_cluster_fixture dml_pliv_oneway_cluster_fixture dml_pliv_multiway_cluster_old_vs_new_fixture dml_procedure test_dml_pliv_multiway_cluster_coef test_dml_pliv_oneway_cluster_coef test_dml_pliv_multiway_cluster_old_vs_new_coef test_dml_plr_cluster_with_index_coef dml_plr_cluster_with_index test_dml_plr_cluster_with_index_se test_dml_pliv_oneway_cluster_se test_dml_pliv_multiway_cluster_se test_version_is_string learner test_dml_pliv_coef test_dml_pliv_boot score test_dml_pliv_se dml_pliv_fixture dml_procedure learner score test_dml_pliv_no_cross_fit_boot dml_pliv_no_cross_fit_fixture test_dml_pliv_no_cross_fit_coef n_folds test_dml_pliv_no_cross_fit_se learner test_dml_pliv_coef test_dml_pliv_boot score test_dml_pliv_se dml_procedure dml_pliv_partial_x_fixture learner test_dml_pliv_coef test_dml_pliv_boot score test_dml_pliv_se dml_pliv_partial_xz_fixture dml_procedure test_dml_pliv_coef test_dml_pliv_boot learner_m score test_dml_pliv_se tune_on_folds dml_procedure dml_pliv_partial_xz_fixture learner_g get_par_grid learner_r test_dml_pliv_coef test_dml_pliv_boot learner_m score test_dml_pliv_se tune_on_folds dml_procedure learner_g get_par_grid learner_r dml_pliv_partial_x_fixture learner test_dml_pliv_coef test_dml_pliv_boot score test_dml_pliv_se dml_procedure dml_pliv_partial_z_fixture test_dml_pliv_coef test_dml_pliv_boot score test_dml_pliv_se tune_on_folds dml_procedure dml_pliv_partial_z_fixture get_par_grid learner_r test_dml_pliv_coef test_dml_pliv_boot learner_m score test_dml_pliv_se tune_on_folds dml_procedure dml_pliv_fixture learner_g get_par_grid learner_r learner dml_plr_binary_classifier_fixture score test_dml_plr_binary_classifier_se test_dml_plr_binary_classifier_coef dml_procedure test_dml_plr_binary_classifier_boot learner n_rep test_dml_plr_multitreat_coef score dml_procedure dml_plr_multitreat_fixture test_dml_plr_multitreat_boot test_dml_plr_multitreat_se idx learner n_rep test_dml_plr_no_cross_fit_boot score dml_plr_rep_no_cross_fit_fixture test_dml_plr_no_cross_fit_se tune_on_folds test_dml_plr_no_cross_fit_tune_coef test_dml_plr_rep_no_cross_fit_se n_folds test_dml_plr_rep_no_cross_fit_coef dml_plr_no_cross_fit_tune_fixture test_dml_plr_no_cross_fit_tune_boot test_dml_plr_no_cross_fit_tune_se dml_plr_no_cross_fit_fixture test_dml_plr_rep_no_cross_fit_boot test_dml_plr_no_cross_fit_coef learner test_dml_plr_se dml_plr_reestimate_fixture n_rep score dml_procedure test_dml_plr_coef learner test_dml_plr_se n_rep score test_dml_plr_boot dml_procedure test_dml_plr_coef dml_plr_fixture test_dml_plr_se score test_dml_plr_boot dml_procedure test_dml_plr_coef dml_plr_fixture learner dml_plr_smpls_fixture test_dml_plr_se n_rep score dml_procedure test_dml_plr_coef test_dml_plr_se learner_m score test_dml_plr_boot tune_on_folds dml_procedure learner_g test_dml_plr_coef get_par_grid dml_plr_fixture fit_predict_proba tune_grid_search draw_smpls fit_predict draw_weights boot_manual est_one_way_cluster_dml2 est_two_way_cluster_dml2 DoubleMLMultiwayResampling var_two_way_cluster var_one_way_cluster _dml_cv_predict_ut_version fit_iivm fit_nuisance_iivm var_iivm compute_iivm_residuals iivm_dml2 iivm_orth iivm_dml1 boot_iivm tune_nuisance_iivm boot_iivm_single_split irm_orth fit_irm compute_iivm_residuals irm_dml1 tune_nuisance_irm fit_nuisance_irm var_irm boot_irm_single_split boot_irm irm_dml2 fit_pliv fit_nuisance_pliv pliv_dml2 tune_nuisance_pliv var_pliv boot_pliv_single_split compute_pliv_residuals pliv_orth pliv_dml1 boot_pliv pliv_partial_xz_dml1 var_pliv_partial_xz boot_pliv_partial_xz_single_split pliv_partial_xz_dml2 fit_nuisance_pliv_partial_xz compute_pliv_partial_xz_residuals boot_pliv_partial_xz pliv_partial_xz_orth tune_nuisance_pliv_partial_xz fit_pliv_partial_xz compute_pliv_partial_x_residuals var_pliv_partial_x boot_pliv_partial_x_single_split fit_pliv_partial_x pliv_partial_x_dml2 tune_nuisance_pliv_partial_x pliv_partial_x_dml1 pliv_partial_x_orth boot_pliv_partial_x fit_nuisance_pliv_partial_x var_pliv_partial_z boot_pliv_partial_z compute_pliv_partial_z_residuals fit_pliv_partial_z tune_nuisance_pliv_partial_z pliv_partial_z_dml1 fit_nuisance_pliv_partial_z pliv_partial_z_dml2 boot_pliv_partial_z_single_split pliv_partial_z_orth fit_plr_multitreat fit_nuisance_plr_classifier boot_plr plr_orth var_plr tune_nuisance_plr fit_plr fit_nuisance_plr boot_plr_multitreat plr_dml2 fit_plr_single_split boot_plr_single_split compute_plr_residuals plr_dml1 read_stata copy get_feature_names reset_index replace toarray concat log PolynomialFeatures DataFrame fit_transform read_csv fit get exp multivariate_normal standard_normal divide toeplitz zeros DataFrame column_stack get make_spd_matrix multivariate_normal _g standard_normal dot _m zeros DataFrame column_stack exp multivariate_normal multiply standard_normal pi dot toeplitz uniform sqrt zeros DataFrame column_stack multivariate_normal dot toeplitz binomial zeros DataFrame array column_stack make_spd_matrix multivariate_normal _g standard_normal dot _m zeros DataFrame column_stack multivariate_normal transpose hstack dot toeplitz eye zeros DataFrame array column_stack get normal arange multivariate_normal reset_index concat matmul toeplitz repeat tile power array zeros reshape zeros concatenate append list _check_smpl_split append _check_smpl_split_tpl list sort array set fit list asarray set_params isinstance parallel pred_fun cross_val_predict clone LabelEncoder _check_is_partition Parallel getattr nan full fit_transform enumerate append len list RandomizedSearchCV GridSearchCV fit append KFold exponential normal sqrt power concatenate seed from_arrays bootstrap boot_coef DoubleMLIRM fit_irm len clone boot_t_stat draw_smpls boot_irm fit seed bootstrap boot_plr boot_coef DoubleMLData tolist clone len fit_plr DoubleMLPLR draw_smpls boot_t_stat values fit boot_coef DoubleMLPLR values seed list ones tolist boot_t_stat append bootstrap plr_dml1 DoubleMLData plr_dml2 set_sample_splitting LinearRegression int boot_plr clone dot fit seed param make_plr_turrell2018 seed param make_plr_turrell2018 seed make_spd_matrix param multivariate_normal _g standard_normal dot _m _m2 zeros DataFrame array column_stack seed normal param multivariate_normal delete dot toeplitz zeros DataFrame array column_stack seed param make_pliv_CHS2015 seed param make_irm_data seed make_spd_matrix param exp multivariate_normal _g standard_normal dot binomial zeros seed make_iivm_data param seed make_spd_matrix param exp multivariate_normal _g standard_normal dot binomial zeros seed param make_pliv_CHS2015 seed param make_pliv_CHS2015 seed make_data_pliv_partialZ param multivariate_normal hstack dot toeplitz eye zeros DataFrame array column_stack seed make_classification param make_regression seed param arange make_irm_data size choice nan arange LogisticRegression _dml_cv_predict Lasso _dml_cv_predict_ut_version train_test_split len isnan fetch_401K fetch_bonus fetch_bonus x_cols len seed make_plr_CCDDHNR2018 seed make_plr_turrell2018 seed make_irm_data seed make_iivm_data seed _make_pliv_data seed make_pliv_CHS2015 seed make_pliv_multiway_cluster_CKMS2021 seed from_arrays DoubleMLData tolist values seed from_arrays _make_pliv_data DoubleMLData copy drop rename make_pliv_multiway_cluster_CKMS2021 make_pliv_CHS2015 make_plr_CCDDHNR2018 seed DoubleMLData smpls fit DoubleMLPLR Lasso set_sample_splitting make_plr_CCDDHNR2018 seed from_arrays make_plr_CCDDHNR2018 DoubleMLData DataFrame arange tile DoubleMLClusterData DataFrame arange tile seed x_cols values make_plr_CCDDHNR2018 seed DoubleMLData values make_plr_CCDDHNR2018 seed DoubleMLData values make_plr_CCDDHNR2018 seed DoubleMLClusterData values make_plr_CCDDHNR2018 seed DoubleMLData values make_plr_CCDDHNR2018 seed DoubleMLData set_x_d make_plr_CCDDHNR2018 seed DataFrame arange tile seed make_pliv_multiway_cluster_CKMS2021 make_plr_CCDDHNR2018 zeros inf from_arrays copy copy DoubleMLPLIV DoubleMLPLR DoubleMLPLR fit confint DoubleMLPLR bootstrap fit p_adjust DoubleMLPLR bootstrap fit LogisticRegressionManipulatedPredict LogisticRegression DoubleMLClusterData smpls copy DoubleMLPLIV fit _assert_resampling_default_settings _assert_resampling_default_settings _assert_resampling_default_settings _assert_resampling_default_settings DataFrame p_adjust isinstance all array y d squeeze plr_score predictions y d irm_score squeeze predictions iivm_score y d squeeze predictions smpls Lasso set_sample_splitting DoubleMLPLIV _score_elements fit seed _partialXZ Lasso _partialX _partialZ _score_elements make_pliv_CHS2015 apply_cross_fitting _assert_smpls_equal smpls enumerate smpls _assert_smpls_equal set_sample_splitting smpls _assert_smpls_equal set_sample_splitting array smpls _assert_smpls_equal set_sample_splitting seed smpls DoubleMLPLR _assert_resampling_pars set_sample_splitting seed fit_iivm bootstrap DoubleMLIIVM boot_coef DoubleMLData tolist clone len boot_t_stat draw_smpls boot_iivm values fit seed fit_iivm from_arrays bootstrap DoubleMLIIVM boot_coef len clone boot_t_stat draw_smpls boot_iivm fit seed fit_iivm bootstrap DoubleMLIIVM boot_coef DoubleMLData tolist clone len boot_t_stat draw_smpls boot_iivm values fit seed fit_iivm bootstrap DoubleMLIIVM boot_coef DoubleMLData tolist clone len boot_t_stat draw_smpls boot_iivm values fit tune_nuisance_iivm tune seed from_arrays bootstrap boot_coef DoubleMLIRM fit_irm len clone boot_t_stat draw_smpls boot_irm fit seed from_arrays bootstrap boot_coef DoubleMLIRM fit_irm len clone boot_t_stat draw_smpls boot_irm fit tune_nuisance_irm tune seed from_arrays bootstrap boot_coef DoubleMLIRM fit_irm len clone boot_t_stat draw_smpls boot_irm fit seed DoubleMLIRM from_arrays clone seed set_index DoubleMLData clone DoubleMLMultiwayResampling set_sample_splitting split_samples DoubleMLPLIV fit fit_pliv y est_two_way_cluster_dml2 z seed multiply len DoubleMLPLIV range d smpls sqrt nan unique compute_pliv_residuals power var_two_way_cluster min clone median ravel full x fit seed fit_pliv y d multiply est_one_way_cluster_dml2 clone smpls ravel z sqrt compute_pliv_residuals var_one_way_cluster DoubleMLPLIV x fit seed DoubleMLClusterData reset_index DoubleMLData tolist clone DoubleMLPLR fit seed fit_pliv bootstrap boot_coef DoubleMLData tolist clone len boot_pliv boot_t_stat draw_smpls DoubleMLPLIV values fit seed fit_pliv bootstrap boot_coef DoubleMLData tolist clone len boot_pliv boot_t_stat draw_smpls DoubleMLPLIV values fit seed y d bootstrap boot_coef len clone fit_pliv_partial_x z boot_t_stat draw_smpls _partialX boot_pliv_partial_x x fit seed y d bootstrap boot_coef len clone z boot_t_stat _partialXZ draw_smpls boot_pliv_partial_xz fit_pliv_partial_xz x fit tune_nuisance_pliv_partial_xz tune tune_nuisance_pliv_partial_x tune seed bootstrap boot_pliv_partial_z boot_coef fit_pliv_partial_z DoubleMLData tolist clone len boot_t_stat draw_smpls _partialZ values fit tune_nuisance_pliv_partial_z tune tune_nuisance_pliv tune seed y d bootstrap boot_plr boot_coef len clone fit_plr DoubleMLPLR draw_smpls boot_t_stat Lasso x fit seed fit_plr_multitreat bootstrap boot_coef DoubleMLData tolist clone len DoubleMLPLR draw_smpls boot_plr_multitreat boot_t_stat values fit seed bootstrap boot_plr boot_coef DoubleMLData tolist clone len fit_plr DoubleMLPLR draw_smpls boot_t_stat values fit boot_coef fit_nuisance_plr DoubleMLPLR draw_smpls values seed list tolist boot_t_stat append range plr_dml1 bootstrap DoubleMLData sqrt power zeros boot_plr fit clone median len y boot_coef DoubleMLPLR draw_smpls seed tolist fit_plr boot_t_stat tune bootstrap d DoubleMLData tune_nuisance_plr Lasso boot_plr fit clone x len seed DoubleMLData tolist clone DoubleMLPLR nan _est_causal_pars_and_se fit Lasso set_ml_nuisance_params seed DoubleMLData tolist clone smpls DoubleMLPLR set_sample_splitting fit y d tune_nuisance_plr tune x append range KFold set_params intersect1d append predict enumerate append set_params intersect1d enumerate GridSearchCV fit len intersect1d enumerate KFold exponential normal sqrt power multiply divide mean zeros range unique len unique len unique len min unique len arange LabelEncoder Parallel vstack ndarray _fit_and_predict _check_is_permutation append fit_transform parallel range issparse asarray concatenate nan empty set_params isinstance clone _num_samples full len fit_nuisance_iivm list median iivm_dml1 iivm_dml2 power sqrt append zeros range len ones_like zeros_like clone is_classifier append fit_predict_proba fit_predict tune_grid_search nan full_like enumerate var_iivm compute_iivm_residuals len mean sqrt iivm_orth zeros enumerate var_iivm compute_iivm_residuals sqrt iivm_orth len mean power multiply divide mean multiply divide draw_weights list hstack append boot_iivm_single_split range len multiply compute_iivm_residuals divide mean boot_manual median list power irm_dml1 fit_nuisance_irm sqrt irm_dml2 append zeros range len list zeros_like clone mean is_classifier append fit_predict_proba fit_predict enumerate tune_grid_search irm_orth compute_iivm_residuals len mean sqrt var_irm zeros enumerate irm_orth compute_iivm_residuals sqrt var_irm len mean power multiply divide mean multiply divide draw_weights list hstack append boot_irm_single_split range len multiply compute_iivm_residuals divide mean boot_manual median list fit_nuisance_pliv pliv_dml2 power sqrt pliv_dml1 append zeros range len fit_predict tune_grid_search nan full_like enumerate len mean sqrt var_pliv compute_pliv_residuals pliv_orth zeros enumerate sqrt var_pliv compute_pliv_residuals pliv_orth len mean power multiply mean multiply draw_weights list hstack boot_pliv_single_split append range len mean multiply boot_manual compute_pliv_residuals median list pliv_partial_xz_dml1 pliv_partial_xz_dml2 fit_nuisance_pliv_partial_xz power sqrt append zeros range len set_params hstack predict append enumerate fit_predict fit tune_grid_search GridSearchCV hstack len fit enumerate predict KFold nan full_like enumerate var_pliv_partial_xz len compute_pliv_partial_xz_residuals mean sqrt pliv_partial_xz_orth zeros enumerate var_pliv_partial_xz compute_pliv_partial_xz_residuals sqrt pliv_partial_xz_orth len mean power multiply mean multiply draw_weights list boot_pliv_partial_xz_single_split hstack append range len mean multiply compute_pliv_partial_xz_residuals boot_manual median list power pliv_partial_x_dml2 sqrt pliv_partial_x_dml1 append zeros range fit_nuisance_pliv_partial_x len list zeros_like predict append range fit_predict enumerate list tune_grid_search range append nan full_like enumerate compute_pliv_partial_x_residuals var_pliv_partial_x len mean sqrt zeros pliv_partial_x_orth enumerate compute_pliv_partial_x_residuals var_pliv_partial_x sqrt pliv_partial_x_orth len mean power multiply mean multiply draw_weights list boot_pliv_partial_x_single_split hstack append range len compute_pliv_partial_x_residuals multiply mean boot_manual median list pliv_partial_z_dml1 power fit_nuisance_pliv_partial_z sqrt pliv_partial_z_dml2 append zeros range len hstack fit_predict tune_grid_search hstack nan full_like enumerate var_pliv_partial_z compute_pliv_partial_z_residuals len mean sqrt zeros pliv_partial_z_orth enumerate var_pliv_partial_z compute_pliv_partial_z_residuals sqrt pliv_partial_z_orth len mean power multiply mean multiply draw_weights list hstack append boot_pliv_partial_z_single_split range len compute_pliv_partial_z_residuals multiply mean boot_manual list hstack power sqrt array nan fit_plr_single_split append median full range len median list power sqrt fit_plr_single_split append zeros range len fit_nuisance_plr_classifier fit_nuisance_plr plr_dml2 is_classifier plr_dml1 clone fit_predict clone fit_predict_proba fit_predict tune_grid_search nan full_like enumerate len plr_orth mean sqrt var_plr zeros compute_plr_residuals enumerate plr_orth var_plr sqrt compute_plr_residuals len mean power multiply mean multiply draw_weights list hstack boot_plr_single_split append range len draw_weights list hstack nan boot_plr_single_split append full range len mean multiply compute_plr_residuals boot_manual

# DoubleML - Double Machine Learning in Python <a href="https://docs.doubleml.org"><img src="https://raw.githubusercontent.com/DoubleML/doubleml-for-py/main/doc/logo.png" align="right" width = "120" /></a> [](https://github.com/DoubleML/doubleml-for-py/actions?query=workflow%3Abuild) [](https://badge.fury.io/py/DoubleML) [](https://anaconda.org/conda-forge/doubleml) [](https://codecov.io/gh/DoubleML/doubleml-for-py) [](https://www.codacy.com/gh/DoubleML/doubleml-for-py/dashboard?utm_source=github.com&amp;utm_medium=referral&amp;utm_content=DoubleML/doubleml-for-py&amp;utm_campaign=Badge_Grade) [](https://www.python.org/) The Python package **DoubleML** provides an implementation of the double / debiased machine learning framework of [Chernozhukov et al. (2018)](https://doi.org/10.1111/ectj.12097). It is built on top of [scikit-learn](https://scikit-learn.org) (Pedregosa et al., 2011).

Doyosae/Deep_Complex_UNet

['speech enhancement']

['Phase-aware Speech Enhancement with Deep Complex U-Net']

model_module.py model_train.py model_test.py model_data.py complex_layers/activations.py complex_layers/__init__.py model_loss.py complex_layers/layer.py complex_layers/normalization.py complex_layers/STFT.py complex_layers/dcunet.py datagenerator weighted_SDR_loss modified_SDR_loss convariance_encoder_module decoder_module convariance_decoder_module mask_processing tranposed_STFT transpoed_ISTFT encoder_module complex_BatchNormalization2d inference get_file_list learning_rate_scheduler loop_test data_generator model_flow loop_train zReLU modReLU CReLU complex_softmax complex_tanh CLeaky_ReLU complex_flatten Naive_DCUnet16 DCUnet16 Naive_DCUnet20 DCUnet20 complex_NaiveBatchNormalization sqrt_init complex_batchnorm sanitizedInitGet sanitizedInitSer complex_BatchNorm2d complex_standardization complex_BatchNormalization2d STFT_network ISTFT_network sqrt sum sum SDR_loss reshape transpose transpose squeeze tanh multiply square divide sqrt concat CLeaky_ReLU CLeaky_ReLU concatenate CLeaky_ReLU complex_BatchNormalization2d CLeaky_ReLU complex_BatchNormalization2d concatenate join natsorted append array walk int read format reshape write extend tqdm enumerate array range predict len datagenerator trainable_variables list gradient apply_gradients zip weighted_SDR_loss modified_SDR_loss model str list format print len Adam tqdm loop_test numpy save_weights listdir range enumerate loop_train leaky_relu multiply bool cast float32 complex relu Variable multiply zeros abs tanh complex abs ones sqrt sqrt ndim reshape concatenate reshape complex_standardization ndim concatenate

# Introuduction Impelmentation Phase-aware Speech Enhnacement Deep Complex UNet This is convolution neural networks model for Speech Enhancement Papers URL 1. [Phase-aware Speech Enhancement Deep Complex UNet - openreview](https://openreview.net/pdf?id=SkeRTsAcYm) 2. [Phase-aware Speech Enhancement Deep Complex UNet - arxiv](https://arxiv.org/abs/1903.03107) ## Architecture  ## Issue! Don't use DCUnet16 and DCUnet20 via ComplexBatchNormalization

DreamInvoker/GAIN

['relation extraction']

['Double Graph Based Reasoning for Document-level Relation Extraction']

code/utils.py code/test.py code/data.py code/models/GAIN.py code/models/GAIN_nomention.py code/train.py code/config.py get_opt BERTDGLREDataset DGLREDataloader DGLREDataset test train Accuracy print_params get_cuda logging RelGraphConvLayer GAIN_BERT BiLSTM GAIN_GloVe Bert RelEdgeLayer Attention RelGraphConvLayer GAIN_BERT BiLSTM GAIN_GloVe RelEdgeLayer Attention add_argument ArgumentParser logging argmax max values open list append range get dump format asarray float auc enumerate print sort numpy len checkpoint_dir GAIN_BERT BCE logging model grid zero_grad coslr clip_grad_value_ unsqueeze BERTDGLREDataset numpy save BCEWithLogitsLoss argmax clip list get_cuda train_set map dev_set_save ylabel add ylim title savefig DGLREDataloader load_state_dict legend sum range dev_set get epoch format plot log_step train_set_save test eval lr mkdir CosineAnnealingLR xlim pretrain_model fig_result_dir enumerate load clear time join backward print AdamW xlabel GAIN_GloVe relation_nums parameters print_params model_name DGLREDataset step is_available print print

# Double Graph Based Reasoning for Document-level Relation Extraction Source code for EMNLP 2020 paper: [Double Graph Based Reasoning for Document-level Relation Extraction](https://arxiv.org/abs/2009.13752) > Document-level relation extraction aims to extract relations among entities within a document. Different from sentence-level relation extraction, it requires reasoning over multiple sentences across a document. In this paper, we propose Graph Aggregation-and-Inference Network (GAIN) featuring double graphs. GAIN first constructs a heterogeneous mention-level graph (hMG) to model complex interaction among different mentions across the document. It also constructs an entity-level graph (EG), based on which we propose a novel path reasoning mechanism to infer relations between entities. Experiments on the public dataset, DocRED, show GAIN achieves a significant performance improvement (2.85 on F1) over the previous state-of-the-art. + Architecture  + Overall Results  ## 0. Package Description ``` GAIN/

DreamtaleCore/Refool

['backdoor attack']

['Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks']

train.py utils/__init__.py scripts/compute_each_class_attack_rate.py models/alexnet.py data.py models/wrn.py eval.py models/resnext.py test.py trainer.py models/vgg.py models/__init__.py models/densenet.py models/preresnet.py strategy.py scripts/insert_reflection.py models/resnet.py models/lenet5.py ImageLabelFilelist is_image_file make_dataset ImageFolder ImageFilelist default_loader get_all_data_loaders get_data_file_list default_flist_reader get_reflection_name blend_image extract_list_from_dict main collect_reflection_image ClassifierTrainer AlexNet alexnet densenet Transition DenseNet Bottleneck BasicBlock lenet5 LeNet5 preresnet PreResNet Bottleneck conv3x3 BasicBlock ResNet Bottleneck conv3x3 resnet BasicBlock ResNeXtBottleneck resnext CifarResNeXt vgg19 VGG vgg16_bn vgg19_bn vgg11_bn make_layers vgg11 vgg13 vgg13_bn vgg16 wrn BasicBlock NetworkBlock WideResNet get_dnn_model gather_reflection_images get_image_label_id gen_main_func blend_images get_pretrained_model_list write_loss get_config prepare_sub_folder get_local_time check_dir tensor2img is_image_file join sorted append walk ImageLabelFilelist DataLoader Compose get_data_file_list get_reflection_name print tqdm set_description append listdir range len append items list uint8 mean resize imread clip imwrite infect_ratio gpu_id Train list get_reflection_name ones choices append imread range format n_images n_iterations target_class copy choice blend_image float UpdateWeight PrepareData join int items print makedirs system index median collect_reflection_image Test len AlexNet LeNet5 CifarResNeXt Conv2d make_layers VGG make_layers VGG make_layers VGG make_layers VGG make_layers VGG make_layers VGG make_layers VGG make_layers VGG WideResNet join parse text index getroot iter append open random resize abs max multiply shape uniform pad ceil sum range gen_kernel INTER_CUBIC power GaussianBlur float int uint8 min float32 dstack join get_image_label_id any append split join list partial imwrite blend_images print gather_reflection_images tqdm mean set_description ssim_func imread listdir range makedirs makedirs getattr add_scalar print join format makedirs data uint8 asarray ToPILImage add_ zip to_pil sort

# Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks     Our paper is accepted by **ECCV 2020**. We investigate the use of a natural phenomenon, *i.e.*, reflection, as the backdoor pattern, and propose the reflection backdoor (*Refool*) attack to install stealthy and effective backdoor into DNN models. <div align=center> <img src="figures/teaser.png" alt="Teaser" width="500" align="bottom" /> </div> **Picture:** *Our reflection backdoors (rightmost column) are crafted based on the natural reflection phenomenon, thus need not to mislabel the poisoned samples on purpose (A - D, mislabels are in red texts), nor rely on obvious patterns (A - C, E), unpleasant blending (D), or suspicious stripes (F). Therefore, our reflection backdoor attacks are stealthier.*

DreamtaleCore/USI3D

['intrinsic image decomposition']

['Unsupervised Learning for Intrinsic Image Decomposition from a Single Image']

scripts/gen_train_test_file.py train.py scripts/organize_cgintrisic_data.py data.py networks.py test.py trainer.py scripts/batch_rename.py utils.py ImageLabelFilelist is_image_file rgb_to_irg make_dataset rgb_to_chromaticity MPIFolder ImageFilelist default_loader ImageFolder CGIFolder srgb_to_rgb default_flist_reader is_image_file UnsupIntrinsicTrainer get_data_loader_list slerp Timer eformat get_model_list get_slerp_interp __write_images get_local_time get_scheduler weights_init write_loss get_config write_one_row_html prepare_sub_folder write_2images write_html get_all_data_loaders get_intrinsic_data_loader get_data_loader_folder is_image_file proc_dataset check_dir convert_dataset join sum zeros_like power zeros_like sum zeros_like DataLoader format CGIFolder join get_data_loader_folder DataLoader Compose ImageFilelist ImageFolder Compose DataLoader split data save_image make_grid cat __write_images len print join format makedirs write basename write close write_one_row_html range open getattr add_scalar dot norm arccos sin randn vstack linspace empty array range sort StepLR makedirs percentile join replace float32 square mean expand_dims repeat imread clip binary_erosion join check_dir sort tqdm listdir

# Unsupervised Learning for Intrinsic Image Decomposition from a Single Image     Our paper is accepted by **CVPR2020**. <div align=center> <img src="figures/teaser.jpg" alt="Teaser" width="500" align="bottom" /> </div> **Picture:** *Our method learns intrinsic image decomposition in an unsupervised fashion where the ground truth reflectance and shading is not available in the training data.* <div align=center> <img src="./figures/main_image.jpg" alt="Main image" width="800" align="center" /> </div> **Picture:** *The proposed architecture.*

E666GT/TrafficPredictionNN

['traffic prediction']

['Graph WaveNet for Deep Spatial-Temporal Graph Modeling']

train.py test.py util.py engine.py train_test.py Test/print_test.py model.py Test/test_time.py trainer linear gcn nconv gwnet main main main load_pickle calculate_scaled_laplacian calculate_normalized_laplacian load_adj masked_mae metric masked_rmse asym_adj DataLoader masked_mape sym_adj load_dataset StandardScaler masked_mse data gwnet batch_size nodevec2 adjdata numpy device inverse_transform DataFrame max heatmap load_adj transpose squeeze metric savefig load_state_dict load_dataset append to adjtype range cat format randomadj dropout relu get_iterator mean eval softmax num_nodes checkpoint enumerate load print to_csv nodevec1 aptonly mm in_dim gcn_bool weight_decay save round nhid seq_length str addaptadj argmin state_dict trainer shuffle expid train time learning_rate epochs diags flatten coo_matrix sum array flatten coo_matrix diags diags tocoo flatten coo_matrix eye sum array calculate_normalized_laplacian csr_matrix reduce identity shape eigsh load_pickle load join DataLoader transform StandardScaler isnan float zeros_like where zeros_like where isnan float abs zeros_like where isnan float abs item

# Graph WaveNet for Deep Spatial-Temporal Graph Modeling ## Updating Log ### Variables sensor_ids, len=207, cont_sample="773869", a random 6-digit number\ adj_mx, shape=207,207 , if Identity, it is a eye(207)\ scaler, a variable maybe used in the later part to scale paras. It includes mean and std of the data sensor_id_to_ind, adjinit， used in gwnet as addaptadj ``` if gcn_bool and addaptadj:

EBjerrum/Deep-Chemometrics

['data augmentation']

['Data Augmentation of Spectral Data for Convolutional Neural Network (CNN) Based Deep Chemometrics']

ChemUtils.py EmscScaler dataaugment GlobalStandardScaler SavGolFilt float random range array

# Deep Chemometrics with Data Augmentation  The repository contains two examples of using convolutional neural networks to model spectroscopical data with data augmentation or EMSC as means to handle the variation in baseline. The first example show what kind of results are possible using data augmentation:  The other example use the same setup, but with EMSC used for baseline correction:  The approach and background + a comparison with PLS models are further described in https://arxiv.org/pdf/1710.01927 and the blog post https://www.wildcardconsulting.dk/useful-information/deep-chemometrics-deep-learning-for-spectroscopy/ Feel free to leave a comment on the blog post if you find it useful ;-) ## Reference Please cite: https://arxiv.org/abs/1710.01927

ECNU-ICA/ECNU-SenseMaker

['graph attention', 'data augmentation']

['ECNU-SenseMaker at SemEval-2020 Task 4: Leveraging Heterogeneous Knowledge Resources for Commonsense Validation and Explanation']

optimizer.py utils/getGraphUtils.py utils/text_to_uri.py loss.py utils/gpu_mem_track.py utils/commonsenseQAutils.py SemEval2020-Task4-Commonsense-Validation-and-Explanation-master/evaluation tools/taskB_scorer.py functions.py run_single_model.py run_ensemble_model.py SemEval2020-Task4-Commonsense-Validation-and-Explanation-master/starting_kit/scoring_program/evaluate.py model_modify.py SemEval2020-Task4-Commonsense-Validation-and-Explanation-master/evaluation tools/taskA_scorer.py utils/testUtils.py utils/semevalUtils.py SemEval2020-Task4-Commonsense-Validation-and-Explanation-master/test/sanity_check_task3.py utils/MyDataset.py utils/ensembleUtils.py SemEval2020-Task4-Commonsense-Validation-and-Explanation-master/evaluation tools/taskC_scorer.py models.py utils/GraphUtils.py utils/attentionUtils.py config.py gelu FocalLoss BertForSequenceClassification RobertaForMultipleChoiceWithLM2 AlbertForMultipleChoice RobertaForMultipleChoiceWithLM SOTA_goal_model AlbertModel GPT2ForMultipleChoice BertForMultipleChoice GCNNet RobertaForMultipleChoice RobertaForMaskedLM simple_split create_datasets_with_kbert simple_split_with_graph train_and_finetune load_checkpoint k_fold_cross_validation test create_datasets get_features save_checkpoint create_datasets_with_graph create_datasets_solo train AdamW AdamW_GCC Adam_GCC2 Adam_GCC Adam_GC AdamW_GCC2 build_parse main calculate_accuracy read_predictions read_gold main calculate_accuracy read_predictions read_gold _get_ngrams read_references _compute_bleu main read_predictions calculate_bleu _get_ngrams calculate_accuracy read_predictions_taskAB read_predictions_taskC _compute_bleu main read_references_taskC read_gold_taskAB calculate_bleu main read_references SelfAttention get_all_features_from_commonsenseQA get_graph_features_from_commonsenseQA_solo get_graph_features_from_commonsenseQA get_features_from_commonsenseQA_with_kbert get_features_from_commonsenseQA_solo get_features_from_commonsenseQA Stacking save_graph_pickle load_graph_pickle StackingNNet merge_graph_by_downgrade save_graph_pickle load_conceptnet_numberbatch get_words_from_id requests_start get_data_from_task_2 load_graph_pickle get_graph_from_sentences get_conceptnet_json encode_index get_id_from_words get_datas get_features_from_words MemTracker GraphUtils MyDataset MyDataLoader get_graph_features_from_task_1_solo get_features_from_task_1_with_kbert get_graph_features_from_task_2 get_features_from_task_1 get_features_from_task_2 get_features_from_task_2_with_kbert get_features_from_task_2_solo get_all_features_from_task_2 get_features_from_task_2_test test get_graph_features_from_task_2_solo get_graph_features_from_task_1 get_all_features_from_task_1 train get_features_from_task_1_solo add_knowledge_with_vm test2 replace_numbers simple_tokenize _standardized_text english_filter standardized_uri _standardized_concept_uri argmax model backward zero_grad close tqdm mean softmax set_postfix append to step enumerate len print eval format append tensor randperm len randperm append tensor range len randperm tensor len randperm tensor len list format print DataLoader append gcd range cat len load join format print load_state_dict isfile join format print mkdir save state_dict from_pretrained create_datasets_with_kbert DistributedDataParallel save_checkpoint MyDataLoader str list hasattr Adam device_count getattr to range get_all_features_from_task_2 SummaryWriter format close test type enumerate join time add_scalar print load_checkpoint named_parameters parameters filter split train len list DataLoader train_and_finetune train_and_finetune MyDataLoader extend seed with_kemb hasattr add_argument with_lm device getattr ArgumentParser manual_seed parse_args keys with_kegat items join error exit len error exit pred_labels print calculate_accuracy gold_labels read_predictions read_gold tuple range Counter len _get_ngrams exp Counter zip float sum range len items join error exit append len references read_references predictions calculate_bleu error exit join error makedirs exit write read_predictions_taskAB read_predictions_taskC close read_references_taskC read_gold_taskAB listdir open list get_graph_features_from_commonsenseQA_solo DataLoader reduce_graph_noise print GraphUtils init merge_graph_by_downgrade reduce_graph_noise print load_mp_all_by_pickle GraphUtils merge_graph_by_downgrade list get_graph_features_from_commonsenseQA zip get_semantic_function get format quote json join format print add lower get_conceptnet_json tokenize range read_csv merge extend get replace clear clear items load_conceptnet_numberbatch get_id_from_words append refine_sent items defaultdict add pop str save_graph_pickle format print get_graph_from_sentences len get str T convert_ids_to_tokens len range append DataFrame encode_plus read_csv values merge reduce_graph_noise tensor merge_graph_by_downgrade DataFrame values str convert_tokens_to_ids load_mp_all_by_pickle append add_knowledge_with_vm range cls_token sep_token format GraphUtils merge T print read_csv pop convert_tokens_to_ids len append tokenize read_csv values merge get_graph_features_from_task_1_solo list DataLoader str T Data reduce_graph_noise print DataFrame ones len range GraphUtils encode_index init append tensor merge_graph_by_downgrade read_csv values merge get str convert_ids_to_tokens len merge append range read_csv values encode_plus get convert_ids_to_tokens len merge range lower append merge_graph_by_downgrade read_csv values encode_plus get str T convert_ids_to_tokens len range append DataFrame encode_plus read_csv values merge pad append zeros conceptnet_relation_to_nl range len reduce_graph_noise tensor merge_graph_by_downgrade DataFrame values str convert_tokens_to_ids load_mp_all_by_pickle append add_knowledge_with_vm range cls_token sep_token format GraphUtils merge T print read_csv list get_graph_features_from_task_2_solo DataLoader str T Data reduce_graph_noise print DataFrame ones len range GraphUtils encode_index init append tensor merge_graph_by_downgrade read_csv values merge list zip get_graph_features_from_task_1 get_semantic_function list get_graph_features_from_task_2 zip get_semantic_function print eval format _standardized_concept_uri search lower _standardized_text simple_tokenize token_filter replace

# ECNU-SenseMaker (SemEval-2020 Task 4) <img src="https://upload.wikimedia.org/wikipedia/commons/thumb/9/96/Pytorch_logo.png/800px-Pytorch_logo.png" width="10%">[](https://raw.githubusercontent.com/im0qianqian/CodeforcesEduHacking/master/LICENSE) Source code for "[ECNU-SenseMaker at SemEval-2020 Task 4: Leveraging Heterogeneous Knowledge Resources for Commonsense Validation and Explanation](https://www.aclweb.org/anthology/2020.semeval-1.48.pdf)" (SemEval 2020). ## Introduction - The overview of ECNU-SenseMaker.  - The KEmb module.  ## Leaderboard ### Subtask B (Top-5)

EEEGUI/PSPNet

['scene parsing', 'semantic segmentation']

['Semantic Understanding of Scenes through the ADE20K Dataset']

train.py network.py tools.py inference.py evaluate.py image_reader.py model.py video.py load main get_arguments read_images_from_disk ImageReader image_scaling read_labeled_image_list random_crop_and_pad_image_and_labels image_mirroring load main get_arguments save PSPNet50 PSPNet101 layer Network load_img decode_labels preprocess prepare_label read_labelcolours load main get_arguments save VideoSegmentation show_img add_argument ArgumentParser print restore format crop_to_bounding_box where set_shape add_n Saver flipped_eval gather argmax decode_image Session run open checkpoints global_variables PSPNet squeeze resize_bilinear placeholder add shape streaming_mean_iou cast expand_dims get_arguments format get_checkpoint_state preprocess flip_left_right ConfigProto trange local_variables_initializer load int join constant not_equal print reshape model_checkpoint_path read_file int32 global_variables_initializer split less stack boolean_mask reverse to_float resize_images to_int32 squeeze multiply stack random_uniform resize_nearest_neighbor expand_dims pad_to_bounding_box random_crop concat maximum shape cast set_shape append join split open image_scaling concat image_mirroring read_file cast random_crop_and_pad_image_and_labels decode_png decode_jpeg split print join makedirs decode_labels save_dir imsave img_path load_img makedirs shape loadmat constant one_hot reshape matmul read_labelcolours format print exit read_file isfile decode_png decode_jpeg pad_to_bounding_box concat cast expand_dims split set_random_seed save num_classes less_equal get_collection map scalar_mul range num_steps start_queue_runners sparse_softmax_cross_entropy_with_logits stack random_seed power time learning_rate PSPNet101 snapshot_dir request_stop Coordinator reduce_mean pow UPDATE_OPS prepare_label show imshow axis figure

# PSPNet_tensorflow ## Introduction This is an implementation of PSPNet in TensorFlow for semantic segmentation on the [cityscapes](https://www.cityscapes-dataset.com/) dataset. We first convert weight from [Original Code](https://github.com/hszhao/PSPNet) by using [caffe-tensorflow](https://github.com/ethereon/caffe-tensorflow) framework. ## Update: #### 2018/01/24: 1. `Support evaluation code for ade20k dataset` #### 2018/01/19: 1. `Support inference phase for ade20k dataset` using model of pspnet50 (convert weights from original author) 2. Using `tf.matmul` to decode label, so as to improve the speed of inference. #### 2017/11/06:

EKirschbaum/DISCo

['cell segmentation', 'instance segmentation', 'semantic segmentation']

['DISCo: Deep learning, Instance Segmentation, and Correlations for cell segmentation in calcium imaging']

get_affs.py train.py run.py convert_results.py dataloading.py get_segmentation.py model.py combine_results convert_single_result DISCoDataset GetAffs get_segmentation get_mask BackgroundLabelSuperpixelGenerator DISCoNet Train_DISCoNet list File close where unique zip append keys list File close where unique zip append keys ones shape enumerate

# DISCo: Deep learning, Instance Segmentation and Correlations for cell segmentation in calcium imaging This is a method to perform the cell segmentaiton step in caclium imaging analysis, which uses the temporal information from caclium imaging videos in form of correlations, and combines a deep learning model with an instance segmentation algorithm. ## Publication **"DISCo: Deep learning, Instance Segmentation, and Correlations for cell segmentation in calcium imaging"**, E. Kirschbaum, A. Bailoni, F. A. Hamprecht, *arXiv preprint arXiv:1908.07957*, 2019. [[pdf]](https://arxiv.org/pdf/1908.07957.pdf) ## Requirements: * [**Python 3.6 (or later)**](https://www.python.org/): we recommend installing it with [Anaconda](https://www.anaconda.com/download/) * [**PyTorch 1.0 (or later)**](http://pytorch.org/) * [**GASP**](https://github.com/abailoni/GASP) * [**inferno 0.3 (or later)**](https://github.com/inferno-pytorch/inferno)