SYMBOL INDEX (1371 symbols across 121 files) FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/__main__.py function _parse_args (line 7) | def _parse_args(): function main (line 45) | def main(): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/conditional.py class ConditionalGenerator (line 4) | class ConditionalGenerator(object): method __init__ (line 5) | def __init__(self, data, output_info, log_frequency): method random_choice_prob_index (line 66) | def random_choice_prob_index(self, idx): method sample (line 71) | def sample(self, batch): method sample_zero (line 87) | def sample_zero(self, batch): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/data.py function read_csv (line 7) | def read_csv(csv_filename, meta_filename=None, header=True, discrete=None): function read_tsv (line 32) | def read_tsv(data_filename, meta_filename): function write_tsv (line 78) | def write_tsv(data, meta, output_filename): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/demo.py function load_demo (line 6) | def load_demo(): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/models.py class Discriminator (line 5) | class Discriminator(Module): method calc_gradient_penalty (line 6) | def calc_gradient_penalty( method __init__ (line 33) | def __init__(self, input_dim, dis_dims, pack=10): method forward (line 46) | def forward(self, input): class Residual (line 51) | class Residual(Module): method __init__ (line 52) | def __init__(self, i, o): method forward (line 58) | def forward(self, input): class Generator (line 65) | class Generator(Module): method __init__ (line 66) | def __init__(self, embedding_dim, gen_dims, data_dim): method forward (line 76) | def forward(self, input): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/sampler.py class Sampler (line 4) | class Sampler(object): method __init__ (line 7) | def __init__(self, data, output_info): method sample (line 37) | def sample(self, n, col, opt): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/synthesizer.py class EarlyStopping (line 11) | class EarlyStopping: method __init__ (line 14) | def __init__(self, patience=7, verbose=False, delta=0): method __call__ (line 32) | def __call__(self, val_loss): class CTGANSynthesizer (line 48) | class CTGANSynthesizer(object): method __init__ (line 72) | def __init__( method _apply_activate (line 90) | def _apply_activate(self, data): method _cond_loss (line 107) | def _cond_loss(self, data, c, m): method fit (line 141) | def fit(self, train_data, discrete_columns=tuple(), epochs=300, log_fr... method sample (line 278) | def sample(self, n): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/ctgan/transformer.py class DataTransformer (line 9) | class DataTransformer(object): method __init__ (line 23) | def __init__(self, n_clusters=10, epsilon=0.005): method _fit_continuous (line 28) | def _fit_continuous(self, column, data): method _fit_discrete (line 47) | def _fit_discrete(self, column, data): method fit (line 59) | def fit(self, data, discrete_columns=tuple()): method _transform_continuous (line 81) | def _transform_continuous(self, column_meta, data): method _transform_discrete (line 109) | def _transform_discrete(self, column_meta, data): method transform (line 113) | def transform(self, data): method _inverse_transform_continuous (line 127) | def _inverse_transform_continuous(self, meta, data, sigma): method _inverse_transform_discrete (line 150) | def _inverse_transform_discrete(self, meta, data): method inverse_transform (line 154) | def inverse_transform(self, data, sigmas): FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/encoders.py function get_single_encoder (line 19) | def get_single_encoder(encoder_name: str, cat_cols: list): class DoubleValidationEncoderNumerical (line 68) | class DoubleValidationEncoderNumerical: method __init__ (line 73) | def __init__(self, cols, encoders_names_tuple=()): method fit_transform (line 89) | def fit_transform(self, X: pd.DataFrame, y: np.array) -> pd.DataFrame: method transform (line 133) | def transform(self, X: pd.DataFrame) -> pd.DataFrame: class MultipleEncoder (line 165) | class MultipleEncoder: method __init__ (line 170) | def __init__(self, cols: List[str], encoders_names_tuple=()): method fit_transform (line 186) | def fit_transform(self, X: pd.DataFrame, y: np.array) -> pd.DataFrame: method transform (line 212) | def transform(self, X) -> pd.DataFrame: class FrequencyEncoder (line 236) | class FrequencyEncoder: method __init__ (line 237) | def __init__(self, cols): method fit (line 241) | def fit(self, X: pd.DataFrame, y=None) -> pd.DataFrame: method transform (line 248) | def transform(self, X: pd.DataFrame) -> pd.DataFrame: method fit_transform (line 269) | def fit_transform(self, X: pd.DataFrame, y=None) -> pd.DataFrame: FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/model.py class Model (line 11) | class Model: method __init__ (line 12) | def __init__( method fit (line 41) | def fit(self, X: pd.DataFrame, y: np.array) -> tuple: method predict (line 105) | def predict(self, X: pd.DataFrame, return_shape=True) -> np.array: FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/run_experiment.py function execute_experiment (line 13) | def execute_experiment(dataset_name, encoders_list, validation_type, sam... FILE: DEEP LEARNING/Autoencoders GANS/GAN-for-tabular-data/utils.py function save_dict_to_file (line 12) | def save_dict_to_file(dic: dict, path: str, save_raw=False) -> None: function save_exp_to_file (line 31) | def save_exp_to_file(dic: dict, path: str) -> None: function cat_cols_info (line 52) | def cat_cols_info( function adversarial_test (line 81) | def adversarial_test(left_df, right_df, cat_cols): function extend_gan_train (line 125) | def extend_gan_train(x_train, y_train, x_test, cat_cols, gen_x_times=1.2... function extend_from_original (line 186) | def extend_from_original(x_train, y_train, x_test, cat_cols, gen_x_times... FILE: DEEP LEARNING/Autoencoders GANS/pytorch/CGAN/ConditionalGAN.py function to_onehot (line 58) | def to_onehot(x, num_classes=10): function get_sample_image (line 76) | def get_sample_image(G, n_noise=100): class Discriminator (line 94) | class Discriminator(nn.Module): method __init__ (line 99) | def __init__(self, input_size=784, condition_size=10, num_classes=1): method forward (line 110) | def forward(self, x, c): class Generator (line 120) | class Generator(nn.Module): method __init__ (line 125) | def __init__(self, input_size=100, condition_size=10, num_classes=784): method forward (line 143) | def forward(self, x, c): function save_checkpoint (line 281) | def save_checkpoint(state, file_name="checkpoint.pth.tar"): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/DCGAN/dcgan.py function weights_init (line 113) | def weights_init(m): class Generator (line 127) | class Generator(nn.Module): method __init__ (line 128) | def __init__(self, ngpu): method forward (line 154) | def forward(self, input): class Discriminator (line 176) | class Discriminator(nn.Module): method __init__ (line 177) | def __init__(self, ngpu): method forward (line 201) | def forward(self, input): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/ProgressiveGAN/progan_modules.py class EqualLR (line 8) | class EqualLR: method __init__ (line 9) | def __init__(self, name): method compute_weight (line 12) | def compute_weight(self, module): method apply (line 19) | def apply(module, name): method __call__ (line 29) | def __call__(self, module, input): function equal_lr (line 34) | def equal_lr(module, name="weight"): class PixelNorm (line 40) | class PixelNorm(nn.Module): method __init__ (line 41) | def __init__(self): method forward (line 44) | def forward(self, input): class EqualConv2d (line 48) | class EqualConv2d(nn.Module): method __init__ (line 49) | def __init__(self, *args, **kwargs): method forward (line 57) | def forward(self, input): class EqualConvTranspose2d (line 61) | class EqualConvTranspose2d(nn.Module): method __init__ (line 63) | def __init__(self, *args, **kwargs): method forward (line 71) | def forward(self, input): class EqualLinear (line 75) | class EqualLinear(nn.Module): method __init__ (line 76) | def __init__(self, in_dim, out_dim): method forward (line 85) | def forward(self, input): class ConvBlock (line 89) | class ConvBlock(nn.Module): method __init__ (line 90) | def __init__( method forward (line 123) | def forward(self, input): function upscale (line 128) | def upscale(feat): class Generator (line 132) | class Generator(nn.Module): method __init__ (line 133) | def __init__(self, input_code_dim=128, in_channel=128, pixel_norm=True... method progress (line 174) | def progress(self, feat, module): method output (line 179) | def output(self, feat1, feat2, module1, module2, alpha): method forward (line 189) | def forward(self, input, step=0, alpha=-1): class Discriminator (line 224) | class Discriminator(nn.Module): method __init__ (line 225) | def __init__(self, feat_dim=128): method forward (line 256) | def forward(self, input, step=0, alpha=-1): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/ProgressiveGAN/train.py function accumulate (line 15) | def accumulate(model1, model2, decay=0.999): function imagefolder_loader (line 23) | def imagefolder_loader(path): function sample_data (line 34) | def sample_data(dataloader, image_size=4): function train (line 50) | def train(generator, discriminator, init_step, loader, total_iter=600000): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/Semi-supervised GAN/Datasets.py function MnistLabel (line 7) | def MnistLabel(class_num): function MnistUnlabel (line 34) | def MnistUnlabel(): function MnistTest (line 44) | def MnistTest(): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/Semi-supervised GAN/ImprovedGAN.py class ImprovedGAN (line 21) | class ImprovedGAN(object): method __init__ (line 22) | def __init__(self, G, D, labeled, unlabeled, test, args): method trainD (line 48) | def trainD(self, x_label, y, x_unlabel): method trainG (line 91) | def trainG(self, x_unlabel): method train (line 105) | def train(self): method predict (line 205) | def predict(self, x): method eval (line 210) | def eval(self): method draw (line 223) | def draw(self, batch_size): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/Semi-supervised GAN/Nets.py class Discriminator (line 10) | class Discriminator(nn.Module): method __init__ (line 11) | def __init__(self, input_dim=28 ** 2, output_dim=10): method forward (line 25) | def forward(self, x, feature=False, cuda=False): class Generator (line 45) | class Generator(nn.Module): method __init__ (line 46) | def __init__(self, z_dim, output_dim=28 ** 2): method forward (line 59) | def forward(self, batch_size, cuda=False): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/Semi-supervised GAN/functional.py function log_sum_exp (line 8) | def log_sum_exp(x, axis=1): function reset_normal_param (line 13) | def reset_normal_param(L, stdv, weight_scale=1.0): class LinearWeightNorm (line 18) | class LinearWeightNorm(torch.nn.Module): method __init__ (line 19) | def __init__( method forward (line 43) | def forward(self, x): method __repr__ (line 51) | def __repr__(self): FILE: DEEP LEARNING/Autoencoders GANS/pytorch/VAE/VAR mnist.py class Encoder (line 58) | class Encoder(nn.Module): method __init__ (line 62) | def __init__(self, input_dim, hidden_dim, z_dim): method forward (line 75) | def forward(self, x): class Decoder (line 87) | class Decoder(nn.Module): method __init__ (line 91) | def __init__(self, z_dim, hidden_dim, output_dim): method forward (line 103) | def forward(self, x): class VAE (line 118) | class VAE(nn.Module): method __init__ (line 122) | def __init__(self, enc, dec): method forward (line 128) | def forward(self, x): function train (line 162) | def train(): function test (line 198) | def test(): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/avito_deepIQA/deepIQA/fr_model.py class FRModel (line 8) | class FRModel(chainer.Chain): method __init__ (line 9) | def __init__(self, top="patchwise"): method extract_features (line 35) | def extract_features(self, x, train=True): method forward (line 62) | def forward(self, x_data, x_ref_data, y_data, train=True, n_patches_pe... method patchwise_loss (line 103) | def patchwise_loss(self, h, a, t): method weighted_loss (line 114) | def weighted_loss(self, h, a, t): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/avito_deepIQA/deepIQA/nr_model.py class Model (line 11) | class Model(chainer.Chain): method __init__ (line 12) | def __init__(self, top="patchwise"): method forward (line 32) | def forward(self, x_data, y_data, train=True, n_patches=32): method patchwise_loss (line 84) | def patchwise_loss(self, h, a, t): method weighted_loss (line 95) | def weighted_loss(self, h, a, t): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/neural-image-assessment/utils/check_dataset.py function parse_data (line 54) | def parse_data(filename): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/neural-image-assessment/utils/data_loader.py function parse_data (line 51) | def parse_data(filename, scores): function parse_data_without_augmentation (line 71) | def parse_data_without_augmentation(filename, scores): function train_generator (line 89) | def train_generator(batchsize, shuffle=True): function val_generator (line 131) | def val_generator(batchsize): function features_generator (line 169) | def features_generator(record_path, faeture_size, batchsize, shuffle=True): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/neural-image-assessment/utils/nasnet.py function NASNet (line 81) | def NASNet( function NASNetLarge (line 393) | def NASNetLarge( function NASNetMobile (line 475) | def NASNetMobile( function NASNetCIFAR (line 557) | def NASNetCIFAR( function _separable_conv_block (line 639) | def _separable_conv_block( function _adjust_block (line 694) | def _adjust_block(p, ip, filters, weight_decay=5e-5, id=None): function _normal_A (line 785) | def _normal_A(ip, p, filters, weight_decay=5e-5, id=None): function _reduction_A (line 874) | def _reduction_A(ip, p, filters, weight_decay=5e-5, id=None): function _add_auxiliary_head (line 983) | def _add_auxiliary_head(x, classes, weight_decay): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/neural-image-assessment/utils/score_utils.py function mean_score (line 4) | def mean_score(scores): function std_score (line 11) | def std_score(scores): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/image feat. extraction/nn_image_features.py function get_names_paths (line 41) | def get_names_paths(images_dir): function img_generator (line 49) | def img_generator( function get_model_and_data (line 77) | def get_model_and_data(mode, model_name): function extract_features (line 146) | def extract_features(model_name="vgg16", batch_size=64): function create_features_df (line 178) | def create_features_df(model_name="vgg16", mode="train"): FILE: DEEP LEARNING/Kaggle Avito Demand Prediction Challenge/text embeddings.py function embeding_reading (line 9) | def embeding_reading(path): function text2features (line 22) | def text2features(embeddings_index, text): function df_to_embed_features (line 37) | def df_to_embed_features(df, column, embeddings_index): function load_emb (line 53) | def load_emb(embedding_path, tokenizer, max_features, default=False, emb... FILE: DEEP LEARNING/NLP/Kaggle Quora Insincere Questions Classification/3rd-place.py class AttentionWeightedAverage (line 57) | class AttentionWeightedAverage(Layer): method __init__ (line 63) | def __init__(self, return_attention=False, **kwargs): method build (line 69) | def build(self, input_shape): method call (line 81) | def call(self, x, mask=None): method get_output_shape_for (line 102) | def get_output_shape_for(self, input_shape): method compute_output_shape (line 105) | def compute_output_shape(self, input_shape): method compute_mask (line 111) | def compute_mask(self, input, input_mask=None): function words (line 128) | def words(text): function P (line 132) | def P(word): function correction (line 139) | def correction(word): function candidates (line 144) | def candidates(word): function known (line 149) | def known(words): function edits1 (line 154) | def edits1(word): function edits2 (line 165) | def edits2(word): function singlify (line 170) | def singlify(word): function load_glove (line 180) | def load_glove(word_dict, lemma_dict): function load_fasttext (line 243) | def load_fasttext(word_dict, lemma_dict): function load_para (line 308) | def load_para(word_dict, lemma_dict): function build_model (line 375) | def build_model(embedding_matrix, nb_words, embedding_size=300): FILE: DEEP LEARNING/NLP/Kaggle Quora Insincere Questions Classification/fix misspellings.py function include_spell_mistake (line 95) | def include_spell_mistake(word, similar_word, score): FILE: DEEP LEARNING/NLP/LSTM RNN/Next Chars pytorch/project-tv-script-generation/helper.py function load_data (line 9) | def load_data(path): function preprocess_and_save_data (line 20) | def preprocess_and_save_data(dataset_path, token_lookup, create_lookup_t... function load_preprocess (line 45) | def load_preprocess(): function save_model (line 52) | def save_model(filename, decoder): function load_model (line 57) | def load_model(filename): FILE: DEEP LEARNING/NLP/LSTM RNN/Next Chars pytorch/project-tv-script-generation/problem_unittests.py class _TestNN (line 6) | class _TestNN(torch.nn.Module): method __init__ (line 7) | def __init__(self, input_size, output_size): method forward (line 12) | def forward(self, nn_input, hidden): function _print_success_message (line 19) | def _print_success_message(): class AssertTest (line 23) | class AssertTest(object): method __init__ (line 24) | def __init__(self, params): method test (line 29) | def test(self, assert_condition, assert_message): function test_create_lookup_tables (line 37) | def test_create_lookup_tables(create_lookup_tables): function test_tokenize (line 121) | def test_tokenize(token_lookup): function test_rnn (line 173) | def test_rnn(RNN, train_on_gpu): function test_forward_back_prop (line 236) | def test_forward_back_prop(RNN, forward_back_prop, train_on_gpu): FILE: DEEP LEARNING/NLP/WSDM - Fake News Classification/Berd generate embeddings/0_bert_encode_en_train.py function gen_encodings (line 26) | def gen_encodings(df, column): FILE: DEEP LEARNING/NLP/WSDM - Fake News Classification/Berd generate embeddings/1_bert_encode_en_test.py function gen_encodings (line 26) | def gen_encodings(df, column): FILE: DEEP LEARNING/NLP/WSDM - Fake News Classification/Berd generate embeddings/2_bert_encode_ch_train.py function gen_encodings (line 26) | def gen_encodings(df, column): FILE: DEEP LEARNING/NLP/WSDM - Fake News Classification/Berd generate embeddings/3_bert_encode_ch_test.py function gen_encodings (line 26) | def gen_encodings(df, column): FILE: DEEP LEARNING/NLP/WSDM - Fake News Classification/Berd generate embeddings/4_gen_encoded_dfs.py function label_encode_target (line 9) | def label_encode_target(df, _inplace=True): FILE: DEEP LEARNING/Object detection/YOLO Object Localization Keras/yad2k/utils/utils.py function compose (line 6) | def compose(*funcs): FILE: DEEP LEARNING/Object detection/YOLO Object Localization Keras/yolo_utils.py function read_classes (line 11) | def read_classes(classes_path): function read_anchors (line 18) | def read_anchors(anchors_path): function generate_colors (line 26) | def generate_colors(class_names): function scale_boxes (line 38) | def scale_boxes(boxes, image_shape): function preprocess_image (line 48) | def preprocess_image(img_path, model_image_size): function draw_boxes (line 58) | def draw_boxes(image, out_scores, out_boxes, out_classes, class_names, c... FILE: DEEP LEARNING/Object detection/keras retinanet/train.py function makedirs (line 53) | def makedirs(path): function get_session (line 64) | def get_session(): function model_with_weights (line 72) | def model_with_weights(model, weights, skip_mismatch): function create_models (line 85) | def create_models( function create_callbacks (line 153) | def create_callbacks( function create_generators (line 238) | def create_generators(args, preprocess_image): function check_args (line 348) | def check_args(parsed_args): function parse_args (line 389) | def parse_args(args): function main (line 576) | def main(args=None): FILE: DEEP LEARNING/Pytorch from scratch/MLP/fc_model.py class Network (line 6) | class Network(nn.Module): method __init__ (line 7) | def __init__(self, input_size, output_size, hidden_layers, drop_p=0.5): method forward (line 29) | def forward(self, x): function validation (line 40) | def validation(model, testloader, criterion): function train (line 61) | def train( FILE: DEEP LEARNING/Pytorch from scratch/MLP/helper.py function test_network (line 7) | def test_network(net, trainloader): function imshow (line 31) | def imshow(image, ax=None, title=None, normalize=True): function view_recon (line 55) | def view_recon(img, recon): function view_classify (line 68) | def view_classify(img, ps, version="MNIST"): FILE: DEEP LEARNING/Pytorch from scratch/TODO/GAN/cycle-gan/helpers.py function checkpoint (line 23) | def checkpoint( function merge_images (line 38) | def merge_images(sources, targets, batch_size=16): function to_data (line 56) | def to_data(x): function save_samples (line 65) | def save_samples( FILE: DEEP LEARNING/Pytorch from scratch/TODO/GAN/project-face-generation/problem_unittests.py function _print_success_message (line 6) | def _print_success_message(): class AssertTest (line 10) | class AssertTest(object): method __init__ (line 11) | def __init__(self, params): method test (line 16) | def test(self, assert_condition, assert_message): function test_discriminator (line 24) | def test_discriminator(Discriminator): function test_generator (line 53) | def test_generator(Generator): FILE: DEEP LEARNING/Pytorch from scratch/word2vec-embeddings/utils.py function preprocess (line 5) | def preprocess(text): function create_lookup_tables (line 30) | def create_lookup_tables(words): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/data_loader.py function read_txt (line 9) | def read_txt(txt): class SaltDataset (line 16) | class SaltDataset(Dataset): method __init__ (line 17) | def __init__(self, transform, mode, image_size, fold_index, aug_list): method set_mode (line 35) | def set_mode(self, mode, fold_index): method __getitem__ (line 61) | def __getitem__(self, index): method __len__ (line 131) | def __len__(self): function get_foldloader (line 135) | def get_foldloader(image_size, batch_size, fold_index, aug_list=None, mo... FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/data_process/transform.py function do_resize2 (line 9) | def do_resize2(image, mask, H, W): function compute_center_pad (line 20) | def compute_center_pad(H, W, factor=32): function do_center_pad_to_factor (line 38) | def do_center_pad_to_factor(image, factor=32): function do_center_pad_to_factor_edgeYreflectX (line 46) | def do_center_pad_to_factor_edgeYreflectX(image, factor=32): function do_center_pad_to_factor2 (line 55) | def do_center_pad_to_factor2(image, mask, factor=32): function do_horizontal_flip (line 62) | def do_horizontal_flip(image): function do_horizontal_flip2 (line 68) | def do_horizontal_flip2(image, mask): function compute_random_pad (line 77) | def compute_random_pad(H, W, limit=(-4, 4), factor=32): function do_random_pad_to_factor2 (line 95) | def do_random_pad_to_factor2(image, mask, limit=(-4, 4), factor=32): function do_random_pad_to_factor2_edgeYreflectX (line 105) | def do_random_pad_to_factor2_edgeYreflectX(image, mask, limit=(-4, 4), f... function do_invert_intensity (line 121) | def do_invert_intensity(image): function do_brightness_shift (line 127) | def do_brightness_shift(image, alpha=0.125): function do_brightness_multiply (line 133) | def do_brightness_multiply(image, alpha=1): function do_gamma (line 140) | def do_gamma(image, gamma=1.0): function do_flip_transpose2 (line 147) | def do_flip_transpose2(image, mask, type=0): function do_shift_scale_crop (line 198) | def do_shift_scale_crop(image, mask, x0=0, y0=0, x1=1, y1=1): function do_random_shift_scale_crop_pad2 (line 211) | def do_random_shift_scale_crop_pad2(image, mask, limit=0.10): function do_shift_scale_rotate2 (line 228) | def do_shift_scale_rotate2(image, mask, dx=0, dy=0, scale=1, angle=0): function do_elastic_transform2 (line 269) | def do_elastic_transform2(image, mask, grid=32, distort=0.2): function do_horizontal_shear2 (line 330) | def do_horizontal_shear2(image, mask, dx=0): function resize_and_pad (line 366) | def resize_and_pad(image, resize_size, factor): function resize_and_pad_edgeYreflectX (line 372) | def resize_and_pad_edgeYreflectX(image, resize_size, factor): function resize_and_random_pad (line 378) | def resize_and_random_pad(image, mask, resize_size, factor, limit=(-13, ... function resize_and_random_pad_edgeYreflectX (line 385) | def resize_and_random_pad_edgeYreflectX(image, mask, resize_size, factor): function center_corp (line 395) | def center_corp(image, image_size, crop_size): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/evaluate.py function do_kaggle_metric (line 9) | def do_kaggle_metric(predict, truth, threshold=0.5): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/loss/bce_losses.py class DiceLoss (line 9) | class DiceLoss(nn.Module): method __init__ (line 10) | def __init__(self, smooth=0, eps=1e-7): method forward (line 15) | def forward(self, output, target): function mixed_dice_bce_loss (line 21) | def mixed_dice_bce_loss( function multiclass_dice_loss (line 43) | def multiclass_dice_loss(output, target, smooth=0, activation="softmax"): function where (line 73) | def where(cond, x_1, x_2): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/loss/cyclic_lr.py class CosineAnnealingLR_with_Restart (line 6) | class CosineAnnealingLR_with_Restart(_LRScheduler): method __init__ (line 37) | def __init__( method get_lr (line 62) | def get_lr(self): method step (line 74) | def step(self, epoch=None): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/loss/lovasz_losses.py function lovasz_grad (line 19) | def lovasz_grad(gt_sorted): function iou_binary (line 34) | def iou_binary(preds, labels, EMPTY=1.0, ignore=None, per_image=True): function iou (line 54) | def iou(preds, labels, C, EMPTY=1.0, ignore=None, per_image=False): function lovasz_hinge (line 81) | def lovasz_hinge(logits, labels, per_image=True, ignore=None): function lovasz_hinge_flat (line 101) | def lovasz_hinge_flat(logits, labels): function flatten_binary_scores (line 124) | def flatten_binary_scores(scores, labels, ignore=None): class StableBCELoss (line 139) | class StableBCELoss(torch.nn.modules.Module): method __init__ (line 140) | def __init__(self): method forward (line 143) | def forward(self, input, target): function binary_xloss (line 149) | def binary_xloss(logits, labels, ignore=None): function lovasz_softmax (line 164) | def lovasz_softmax(probas, labels, only_present=False, per_image=False, ... function lovasz_softmax_flat (line 188) | def lovasz_softmax_flat(probas, labels, only_present=False): function flatten_probas (line 209) | def flatten_probas(probas, labels, ignore=None): function xloss (line 224) | def xloss(logits, labels, ignore=None): function mean (line 234) | def mean(l, ignore_nan=False, empty=0): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/main.py class SingleModelSolver (line 20) | class SingleModelSolver(object): method __init__ (line 21) | def __init__(self, config): method build_model (line 67) | def build_model(self): method print_network (line 89) | def print_network(self, model, name): method load_pretrained_model (line 97) | def load_pretrained_model(self, fold_index, mode=None, Cycle=None): method update_lr (line 165) | def update_lr(self, g_lr): method to_var (line 169) | def to_var(self, x, volatile=False): method criterion (line 174) | def criterion(self, logits, label): method train_fold (line 180) | def train_fold(self, fold_index, aug_list): method val_TTA (line 405) | def val_TTA(self, fold_index, val_loader, is_load=False, mode=None, Cy... method get_infer_TTA (line 508) | def get_infer_TTA(self, fold_index, thres): method infer_fold_TTA (line 556) | def infer_fold_TTA(self, fold_index, mode="max_map", Cycle=None): method infer_fold_all_Cycle (line 582) | def infer_fold_all_Cycle(self, fold_index, mode="max_map"): function main (line 587) | def main(config, aug_list): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/model/ibnnet.py class IBN (line 19) | class IBN(nn.Module): method __init__ (line 20) | def __init__(self, planes): method forward (line 28) | def forward(self, x): class Bottleneck (line 36) | class Bottleneck(nn.Module): method __init__ (line 43) | def __init__( method forward (line 84) | def forward(self, x): class ResNeXt (line 107) | class ResNeXt(nn.Module): method __init__ (line 113) | def __init__(self, baseWidth, cardinality, layers, num_classes): method _make_layer (line 153) | def _make_layer(self, block, planes, blocks, stride=1): method forward (line 206) | def forward(self, x): function resnext50_ibn_a (line 222) | def resnext50_ibn_a(baseWidth, cardinality): function resnext101_ibn_a (line 233) | def resnext101_ibn_a(baseWidth, cardinality, pretrained=True): function resnext152_ibn_a (line 248) | def resnext152_ibn_a(baseWidth, cardinality): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/model/model.py function Stconv3x3 (line 9) | def Stconv3x3(in_, out, bias=True): function Stconv7x7 (line 13) | def Stconv7x7(in_, out, bias=True): function Stconv5x5 (line 17) | def Stconv5x5(in_, out, bias=True): function Stconv1x1 (line 21) | def Stconv1x1(in_, out, bias=True): class StConvRelu (line 25) | class StConvRelu(nn.Module): method __init__ (line 26) | def __init__(self, in_, out, kernel_size, norm_type=None): method forward (line 50) | def forward(self, x): class ImprovedIBNaDecoderBlock (line 61) | class ImprovedIBNaDecoderBlock(nn.Module): method __init__ (line 62) | def __init__(self, in_channels, n_filters): method forward (line 77) | def forward(self, x): class SELayer (line 90) | class SELayer(nn.Module): method __init__ (line 91) | def __init__(self, channel, reduction=16): method forward (line 101) | def forward(self, x): class SCSEBlock (line 108) | class SCSEBlock(nn.Module): method __init__ (line 109) | def __init__(self, channel, reduction=16): method forward (line 125) | def forward(self, x): class Decoder (line 138) | class Decoder(nn.Module): method __init__ (line 139) | def __init__(self, in_channels, channels, out_channels): method forward (line 153) | def forward(self, x, e=None): class model34_DeepSupervion (line 165) | class model34_DeepSupervion(nn.Module): method __init__ (line 166) | def __init__(self, num_classes=1, mask_class=2): method forward (line 214) | def forward(self, x): class model50A_DeepSupervion (line 254) | class model50A_DeepSupervion(nn.Module): method __init__ (line 255) | def __init__(self, num_classes=1): method forward (line 305) | def forward(self, x): class model101A_DeepSupervion (line 345) | class model101A_DeepSupervion(nn.Module): method __init__ (line 346) | def __init__(self, num_classes=1): method forward (line 397) | def forward(self, x): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/model/senet.py class SEModule (line 92) | class SEModule(nn.Module): method __init__ (line 93) | def __init__(self, channels, reduction): method forward (line 101) | def forward(self, x): class Bottleneck (line 111) | class Bottleneck(nn.Module): method forward (line 116) | def forward(self, x): class SEBottleneck (line 139) | class SEBottleneck(Bottleneck): method __init__ (line 146) | def __init__(self, inplanes, planes, groups, reduction, stride=1, down... class SEResNetBottleneck (line 168) | class SEResNetBottleneck(Bottleneck): method __init__ (line 177) | def __init__(self, inplanes, planes, groups, reduction, stride=1, down... class SEResNeXtBottleneck (line 195) | class SEResNeXtBottleneck(Bottleneck): method __init__ (line 202) | def __init__( class SENet (line 235) | class SENet(nn.Module): method __init__ (line 236) | def __init__( method _make_layer (line 364) | def _make_layer( method features (line 399) | def features(self, x): method logits (line 407) | def logits(self, x): method forward (line 415) | def forward(self, x): function initialize_pretrained_model (line 421) | def initialize_pretrained_model(model, num_classes, settings): function senet154 (line 435) | def senet154(num_classes=1000, pretrained="imagenet"): function se_resnet50 (line 450) | def se_resnet50(num_classes=1000, pretrained="imagenet"): function se_resnet101 (line 469) | def se_resnet101(num_classes=1000, pretrained="imagenet"): function se_resnet152 (line 488) | def se_resnet152(num_classes=1000, pretrained="imagenet"): function se_resnext50_32x4d (line 507) | def se_resnext50_32x4d(num_classes=1000, pretrained="imagenet"): function se_resnext101_32x4d (line 526) | def se_resnext101_32x4d(num_classes=1000, pretrained="imagenet"): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v1/utils.py function do_length_encode (line 21) | def do_length_encode(x): function do_length_decode (line 43) | def do_length_decode(rle, H, W, fill_value=255): function decode_csv (line 58) | def decode_csv(csv_name): function save_id_fea (line 73) | def save_id_fea(predict_dict, save_dir): function state_dict_remove_moudle (line 79) | def state_dict_remove_moudle(moudle_state_dict, model): function write_and_plot (line 94) | def write_and_plot(name, aver_num, logits, max_y=1.0, color="blue"): function decompose (line 109) | def decompose(labeled): function encode_rle (line 124) | def encode_rle(predictions): function create_submission (line 128) | def create_submission(predictions): function run_length_encoding (line 140) | def run_length_encoding(x): function run_length_decoding (line 158) | def run_length_decoding(mask_rle, shape): function sigmoid (line 179) | def sigmoid(x): function softmax (line 183) | def softmax(X, theta=1.0, axis=None): function from_pil (line 229) | def from_pil(*images): function to_pil (line 237) | def to_pil(*images): function binary_from_rle (line 245) | def binary_from_rle(rle): function get_crop_pad_sequence (line 249) | def get_crop_pad_sequence(vertical, horizontal): function get_list_of_image_predictions (line 257) | def get_list_of_image_predictions(batch_predictions): function set_seed (line 264) | def set_seed(seed): class ImgAug (line 272) | class ImgAug: method __init__ (line 273) | def __init__(self, augmenters): method _pre_call_hook (line 279) | def _pre_call_hook(self): method transform (line 284) | def transform(self, *images): method __call__ (line 291) | def __call__(self, *args): function get_seed (line 296) | def get_seed(): function reseed (line 301) | def reseed(augmenter, deterministic=True): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/augmentation.py function _perspective_transform_augment_images (line 9) | def _perspective_transform_augment_images(self, images, random_state, pa... function resize_pad_seq (line 84) | def resize_pad_seq(resize_target_size, pad_method, pad_size): function pad_to_fit_net (line 96) | def pad_to_fit_net(divisor, pad_mode, rest_of_augs=iaa.Noop()): class PadFixed (line 100) | class PadFixed(iaa.Augmenter): method __init__ (line 103) | def __init__( method _augment_images (line 115) | def _augment_images(self, images, random_state, parents, hooks): method _augment_keypoints (line 122) | def _augment_keypoints(self, keypoints_on_images, random_state, parent... method _pad (line 126) | def _pad(self, img): method get_parameters (line 147) | def get_parameters(self): method _is_expanded_grey_format (line 150) | def _is_expanded_grey_format(self, img): function test_time_augmentation_transform (line 157) | def test_time_augmentation_transform(image, tta_parameters): function test_time_augmentation_inverse_transform (line 169) | def test_time_augmentation_inverse_transform(image, tta_parameters): function per_channel_flipud (line 179) | def per_channel_flipud(x): function per_channel_fliplr (line 186) | def per_channel_fliplr(x): function per_channel_rotation (line 193) | def per_channel_rotation(x, angle): function rotate (line 197) | def rotate(image, angle, axes=(0, 1)): class RandomCropFixedSize (line 204) | class RandomCropFixedSize(iaa.Augmenter): method __init__ (line 205) | def __init__(self, px=None, name=None, deterministic=False, random_sta... method _augment_images (line 218) | def _augment_images(self, images, random_state, parents, hooks): method _augment_keypoints (line 228) | def _augment_keypoints(self, keypoints_on_images, random_state, parent... method _random_crop (line 232) | def _random_crop(self, seed, image): method get_parameters (line 259) | def get_parameters(self): class InferencePad (line 263) | class InferencePad(iaa.Augmenter): method __init__ (line 264) | def __init__( method _augment_keypoints (line 278) | def _augment_keypoints(self, keypoints_on_images, random_state, parent... method _augment_images (line 281) | def _augment_images(self, images, random_state, parents, hooks): method _pad_image (line 289) | def _pad_image(self, image): method _get_pad_sequence (line 297) | def _get_pad_sequence(self, height, width): method _get_pad (line 302) | def _get_pad(self, dim): method get_parameters (line 308) | def get_parameters(self): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/callbacks.py class Callback (line 35) | class Callback: method __init__ (line 36) | def __init__(self): method set_params (line 47) | def set_params(self, transformer, validation_datagen, *args, **kwargs): method on_train_begin (line 55) | def on_train_begin(self, *args, **kwargs): method on_train_end (line 59) | def on_train_end(self, *args, **kwargs): method on_epoch_begin (line 62) | def on_epoch_begin(self, *args, **kwargs): method on_epoch_end (line 65) | def on_epoch_end(self, *args, **kwargs): method training_break (line 68) | def training_break(self, *args, **kwargs): method on_batch_begin (line 71) | def on_batch_begin(self, *args, **kwargs): method on_batch_end (line 74) | def on_batch_end(self, *args, **kwargs): method get_validation_loss (line 77) | def get_validation_loss(self): class CallbackList (line 85) | class CallbackList: method __init__ (line 86) | def __init__(self, callbacks=None): method __len__ (line 94) | def __len__(self): method set_params (line 97) | def set_params(self, *args, **kwargs): method on_train_begin (line 101) | def on_train_begin(self, *args, **kwargs): method on_train_end (line 105) | def on_train_end(self, *args, **kwargs): method on_epoch_begin (line 109) | def on_epoch_begin(self, *args, **kwargs): method on_epoch_end (line 113) | def on_epoch_end(self, *args, **kwargs): method training_break (line 117) | def training_break(self, *args, **kwargs): method on_batch_begin (line 123) | def on_batch_begin(self, *args, **kwargs): method on_batch_end (line 127) | def on_batch_end(self, *args, **kwargs): class TrainingMonitor (line 132) | class TrainingMonitor(Callback): method __init__ (line 133) | def __init__(self, epoch_every=None, batch_every=None): method on_train_begin (line 145) | def on_train_begin(self, *args, **kwargs): method on_epoch_end (line 150) | def on_epoch_end(self, *args, **kwargs): method on_batch_end (line 162) | def on_batch_end(self, metrics, *args, **kwargs): class ExponentialLRScheduler (line 180) | class ExponentialLRScheduler(Callback): method __init__ (line 181) | def __init__(self, gamma, epoch_every=1, batch_every=None): method set_params (line 193) | def set_params(self, transformer, validation_datagen, *args, **kwargs): method on_train_begin (line 200) | def on_train_begin(self, *args, **kwargs): method on_epoch_end (line 209) | def on_epoch_end(self, *args, **kwargs): method on_batch_end (line 220) | def on_batch_end(self, *args, **kwargs): class ExperimentTiming (line 233) | class ExperimentTiming(Callback): method __init__ (line 234) | def __init__(self, epoch_every=None, batch_every=None): method on_train_begin (line 249) | def on_train_begin(self, *args, **kwargs): method on_train_end (line 254) | def on_train_end(self, *args, **kwargs): method on_epoch_begin (line 257) | def on_epoch_begin(self, *args, **kwargs): method on_batch_begin (line 272) | def on_batch_begin(self, *args, **kwargs): class NeptuneMonitor (line 292) | class NeptuneMonitor(Callback): method __init__ (line 293) | def __init__(self, image_nr, image_resize, model_name): method on_train_begin (line 301) | def on_train_begin(self, *args, **kwargs): method on_batch_end (line 306) | def on_batch_end(self, metrics, *args, **kwargs): method on_epoch_end (line 324) | def on_epoch_end(self, *args, **kwargs): method _send_numeric_channels (line 328) | def _send_numeric_channels(self, *args, **kwargs): class ValidationMonitor (line 350) | class ValidationMonitor(Callback): method __init__ (line 351) | def __init__(self, data_dir, loader_mode, epoch_every=None, batch_ever... method set_params (line 369) | def set_params( method get_validation_loss (line 382) | def get_validation_loss(self): method on_epoch_end (line 385) | def on_epoch_end(self, *args, **kwargs): method _get_validation_loss (line 399) | def _get_validation_loss(self): method _transform (line 421) | def _transform(self): method _generate_prediction (line 483) | def _generate_prediction(self, outputs): class ModelCheckpoint (line 495) | class ModelCheckpoint(Callback): method __init__ (line 496) | def __init__(self, filepath, metric_name="sum", epoch_every=1, minimiz... method on_train_begin (line 508) | def on_train_begin(self, *args, **kwargs): method on_epoch_end (line 513) | def on_epoch_end(self, *args, **kwargs): class EarlyStopping (line 539) | class EarlyStopping(Callback): method __init__ (line 540) | def __init__(self, metric_name="sum", patience=1000, minimize=True): method training_break (line 549) | def training_break(self, *args, **kwargs): method on_epoch_end (line 552) | def on_epoch_end(self, *args, **kwargs): function postprocessing_pipeline_simplified (line 576) | def postprocessing_pipeline_simplified(cache_dirpath, loader_mode): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/loaders.py class ImageReader (line 20) | class ImageReader(BaseTransformer): method __init__ (line 21) | def __init__(self, train_mode, x_columns, y_columns, target_format="pn... method transform (line 27) | def transform(self, meta): method load_images (line 39) | def load_images(self, filepaths, filetype, grayscale=False): method load_image (line 54) | def load_image(self, img_filepath, grayscale): method read_json (line 62) | def read_json(self, path): class XYSplit (line 69) | class XYSplit(BaseTransformer): method __init__ (line 70) | def __init__(self, train_mode, x_columns, y_columns): method transform (line 78) | def transform(self, meta): class ImageSegmentationBaseDataset (line 88) | class ImageSegmentationBaseDataset(Dataset): method __init__ (line 89) | def __init__( method __len__ (line 121) | def __len__(self): method __getitem__ (line 127) | def __getitem__(self, index): method load_from_memory (line 161) | def load_from_memory(self, data_source, index, **kwargs): method load_from_disk (line 164) | def load_from_disk(self, data_source, index, *, filetype, grayscale=Fa... method load_image (line 174) | def load_image(self, img_filepath, grayscale): method read_json (line 182) | def read_json(self, path): method load_target (line 188) | def load_target(self, data_source, index, load_func): class ImageSegmentationJsonDataset (line 192) | class ImageSegmentationJsonDataset(ImageSegmentationBaseDataset): method load_target (line 193) | def load_target(self, data_source, index, load_func): class ImageSegmentationPngDataset (line 198) | class ImageSegmentationPngDataset(ImageSegmentationBaseDataset): method load_target (line 199) | def load_target(self, data_source, index, load_func): class ImageSegmentationTTADataset (line 206) | class ImageSegmentationTTADataset(ImageSegmentationBaseDataset): method __init__ (line 207) | def __init__(self, tta_params, tta_transform, *args, **kwargs): method __getitem__ (line 212) | def __getitem__(self, index): class ImageSegmentationLoaderBasic (line 237) | class ImageSegmentationLoaderBasic(BaseTransformer): method __init__ (line 238) | def __init__(self, train_mode, loader_params, dataset_params, augmenta... method transform (line 255) | def transform(self, X, y, X_valid=None, y_valid=None, **kwargs): method get_datagen (line 274) | def get_datagen(self, X, y, train_mode, loader_params): method load (line 302) | def load(self, filepath): method save (line 307) | def save(self, filepath): class ImageSegmentationLoaderBasicTTA (line 312) | class ImageSegmentationLoaderBasicTTA(ImageSegmentationLoaderBasic): method __init__ (line 313) | def __init__(self, loader_params, dataset_params, augmentation_params): method transform (line 328) | def transform(self, X, tta_params, **kwargs): method get_datagen (line 337) | def get_datagen(self, X, tta_params, loader_params): class ImageSegmentationLoaderResizePad (line 356) | class ImageSegmentationLoaderResizePad(ImageSegmentationLoaderBasic): method __init__ (line 357) | def __init__(self, train_mode, loader_params, dataset_params, augmenta... class ImageSegmentationLoaderPadTTA (line 394) | class ImageSegmentationLoaderPadTTA(ImageSegmentationLoaderBasicTTA): method __init__ (line 395) | def __init__(self, loader_params, dataset_params, augmentation_params): class ImageSegmentationLoaderResize (line 420) | class ImageSegmentationLoaderResize(ImageSegmentationLoaderBasic): method __init__ (line 421) | def __init__(self, train_mode, loader_params, dataset_params, augmenta... class ImageSegmentationLoaderResizeTTA (line 459) | class ImageSegmentationLoaderResizeTTA(ImageSegmentationLoaderBasicTTA): method __init__ (line 460) | def __init__(self, loader_params, dataset_params, augmentation_params): class MetaTestTimeAugmentationGenerator (line 486) | class MetaTestTimeAugmentationGenerator(BaseTransformer): method __init__ (line 487) | def __init__(self, **kwargs): method transform (line 490) | def transform(self, X, **kwargs): method _get_tta_data (line 500) | def _get_tta_data(self, i, row): class TestTimeAugmentationGenerator (line 539) | class TestTimeAugmentationGenerator(BaseTransformer): method __init__ (line 540) | def __init__(self, **kwargs): method transform (line 543) | def transform(self, X, **kwargs): method _get_tta_data (line 553) | def _get_tta_data(self, i, row): class TestTimeAugmentationAggregator (line 592) | class TestTimeAugmentationAggregator(BaseTransformer): method __init__ (line 593) | def __init__(self, tta_inverse_transform, method, nthreads): method agg_method (line 599) | def agg_method(self): method transform (line 603) | def transform(self, images, tta_params, img_ids, **kwargs): function aggregate_augmentations (line 619) | def aggregate_augmentations( function to_array (line 633) | def to_array(x): function to_tensor (line 640) | def to_tensor(x): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/metrics.py function iou (line 8) | def iou(gt, pred): function compute_ious (line 21) | def compute_ious(gt, predictions): function compute_precision_at (line 37) | def compute_precision_at(ious, threshold): function compute_eval_metric (line 46) | def compute_eval_metric(gt, predictions): function intersection_over_union (line 53) | def intersection_over_union(y_true, y_pred): function intersection_over_union_thresholds (line 62) | def intersection_over_union_thresholds(y_true, y_pred): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/models.py class PyTorchUNet (line 89) | class PyTorchUNet(Model): method __init__ (line 90) | def __init__(self, architecture_config, training_config, callbacks_con... method fit (line 104) | def fit(self, datagen, validation_datagen=None, meta_valid=None): method _fit_loop (line 133) | def _fit_loop(self, data): method transform (line 169) | def transform(self, datagen, validation_datagen=None, *args, **kwargs): method _transform (line 183) | def _transform(self, datagen, validation_datagen=None, **kwargs): method set_model (line 216) | def set_model(self): method set_loss (line 225) | def set_loss(self): method load (line 253) | def load(self, filepath): function designed_loss (line 270) | def designed_loss(output, target): function mean (line 276) | def mean(l, ignore_nan=False, empty=0): function lovasz_grad (line 297) | def lovasz_grad(gt_sorted): function lovasz_hinge (line 312) | def lovasz_hinge(logits, labels, per_image=True, ignore=None): function lovasz_hinge_flat (line 332) | def lovasz_hinge_flat(logits, labels): function flatten_binary_scores (line 354) | def flatten_binary_scores(scores, labels, ignore=None): function weight_regularization (line 369) | def weight_regularization(model, regularize, weight_decay_conv2d): function callbacks_unet (line 382) | def callbacks_unet(callbacks_config): class DiceLoss (line 404) | class DiceLoss(nn.Module): method __init__ (line 405) | def __init__(self, smooth=0, eps=1e-7): method forward (line 410) | def forward(self, output, target): function mixed_dice_bce_loss (line 416) | def mixed_dice_bce_loss( function mixed_dice_cross_entropy_loss (line 437) | def mixed_dice_cross_entropy_loss( function multiclass_dice_loss (line 464) | def multiclass_dice_loss(output, target, smooth=0, activation="softmax"): function where (line 494) | def where(cond, x_1, x_2): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/pipelines.py function preprocessing_train (line 11) | def preprocessing_train(config, model_name="unet", suffix=""): function preprocessing_inference (line 82) | def preprocessing_inference(config, model_name="unet", suffix=""): function preprocessing_inference_tta (line 129) | def preprocessing_inference_tta(config, model_name="unet", suffix=""): function aggregator (line 197) | def aggregator(name, model, tta_generator, experiment_directory, config): function mask_postprocessing (line 214) | def mask_postprocessing(config, suffix=""): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/pnasnet.py class MaxPool (line 33) | class MaxPool(nn.Module): method __init__ (line 34) | def __init__(self, kernel_size, stride=1, padding=1, zero_pad=False): method forward (line 39) | def forward(self, x): class SeparableConv2d (line 48) | class SeparableConv2d(nn.Module): method __init__ (line 49) | def __init__( method forward (line 66) | def forward(self, x): class BranchSeparables (line 72) | class BranchSeparables(nn.Module): method __init__ (line 73) | def __init__( method forward (line 101) | def forward(self, x): class ReluConvBn (line 115) | class ReluConvBn(nn.Module): method __init__ (line 116) | def __init__(self, in_channels, out_channels, kernel_size, stride=1): method forward (line 128) | def forward(self, x): class FactorizedReduction (line 135) | class FactorizedReduction(nn.Module): method __init__ (line 136) | def __init__(self, in_channels, out_channels): method forward (line 168) | def forward(self, x): class CellBase (line 182) | class CellBase(nn.Module): method cell_forward (line 183) | def cell_forward(self, x_left, x_right): class CellStem0 (line 214) | class CellStem0(CellBase): method __init__ (line 215) | def __init__( method forward (line 265) | def forward(self, x_left): class Cell (line 271) | class Cell(CellBase): method __init__ (line 272) | def __init__( method forward (line 351) | def forward(self, x_left, x_right): class PNASNet5Large (line 358) | class PNASNet5Large(nn.Module): method __init__ (line 359) | def __init__(self, num_classes=1001): method features (line 467) | def features(self, x): method logits (line 485) | def logits(self, features): method forward (line 493) | def forward(self, input): function pnasnet5large (line 499) | def pnasnet5large(num_classes=1001, pretrained="imagenet"): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/postprocessing.py function resize_image (line 8) | def resize_image(image, target_size): function crop_image (line 26) | def crop_image(image, target_size): function binarize (line 48) | def binarize(image, threshold): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/preprocessing.py function img_cumsum (line 4) | def img_cumsum(img): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/resnext.py function conv3x3 (line 13) | def conv3x3(in_planes, out_planes, stride=1): class BasicBlock (line 20) | class BasicBlock(nn.Module): method __init__ (line 23) | def __init__(self, inplanes, planes, stride=1, downsample=None, num_gr... method forward (line 33) | def forward(self, x): class Bottleneck (line 52) | class Bottleneck(nn.Module): method __init__ (line 55) | def __init__(self, inplanes, planes, stride=1, downsample=None, num_gr... method forward (line 75) | def forward(self, x): class ResNeXt (line 98) | class ResNeXt(nn.Module): method __init__ (line 99) | def __init__(self, block, layers, num_classes=1000, num_group=32): method _make_layer (line 121) | def _make_layer(self, block, planes, blocks, num_group, stride=1): method forward (line 145) | def forward(self, x): function resnext18 (line 163) | def resnext18(**kwargs): function resnext34 (line 170) | def resnext34(**kwargs): function resnext50 (line 177) | def resnext50(**kwargs): function resnext101 (line 184) | def resnext101(**kwargs): function resnext152 (line 191) | def resnext152(**kwargs): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/unet_models.py function conv3x3 (line 34) | def conv3x3(in_, out): class ConvRelu (line 38) | class ConvRelu(nn.Module): method __init__ (line 39) | def __init__(self, in_, out): method forward (line 44) | def forward(self, x): method __init__ (line 894) | def __init__(self, in_, out): method forward (line 899) | def forward(self, x): class NoOperation (line 50) | class NoOperation(nn.Module): method forward (line 51) | def forward(self, x): class DecoderBlock_old (line 55) | class DecoderBlock_old(nn.Module): method __init__ (line 56) | def __init__(self, in_channels, middle_channels, out_channels): method forward (line 73) | def forward(self, x): class DecoderBlock (line 77) | class DecoderBlock(nn.Module): method __init__ (line 78) | def __init__(self, in_channels, middle_channels, out_channels): method forward (line 87) | def forward(self, x, e=None): class UNet11 (line 101) | class UNet11(nn.Module): method __init__ (line 102) | def __init__(self, num_classes=1, num_filters=32, pretrained=False): method forward (line 144) | def forward(self, x): function unet11 (line 164) | def unet11(pretrained=False, **kwargs): class DecoderBlockV2 (line 180) | class DecoderBlockV2(nn.Module): method __init__ (line 181) | def __init__(self, in_channels, middle_channels, out_channels, is_deco... method forward (line 206) | def forward(self, x): class DecoderCenter (line 210) | class DecoderCenter(nn.Module): method __init__ (line 211) | def __init__(self, in_channels, middle_channels, out_channels, is_deco... method forward (line 237) | def forward(self, x): class AlbuNet (line 241) | class AlbuNet(nn.Module): method __init__ (line 249) | def __init__( method forward (line 305) | def forward(self, x): class UNetVGG16 (line 325) | class UNetVGG16(nn.Module): method __init__ (line 326) | def __init__( method forward (line 398) | def forward(self, x): class UNetResNet (line 417) | class UNetResNet(nn.Module): method __init__ (line 418) | def __init__( method forward (line 492) | def forward(self, x): class UNetResNet_wo_pool (line 511) | class UNetResNet_wo_pool(nn.Module): method __init__ (line 512) | def __init__( method forward (line 575) | def forward(self, x): class UNetResNext_wo_pool (line 589) | class UNetResNext_wo_pool(nn.Module): method __init__ (line 590) | def __init__( method forward (line 644) | def forward(self, x): class UNetResNetAttentionv2 (line 666) | class UNetResNetAttentionv2(nn.Module): method __init__ (line 667) | def __init__( method forward (line 723) | def forward(self, x): class UNetResNetAttention (line 755) | class UNetResNetAttention(nn.Module): method __init__ (line 756) | def __init__( method forward (line 824) | def forward(self, x): class EncoderBlock (line 855) | class EncoderBlock(nn.Module): method __init__ (line 856) | def __init__(self, block, out_channels): method forward (line 863) | def forward(self, x): class ChannelAttentionGate (line 871) | class ChannelAttentionGate(nn.Module): method __init__ (line 872) | def __init__(self, channel, reduction=16): method forward (line 882) | def forward(self, x): function conv3x3 (line 889) | def conv3x3(in_, out): class ConvRelu (line 893) | class ConvRelu(nn.Module): method __init__ (line 39) | def __init__(self, in_, out): method forward (line 44) | def forward(self, x): method __init__ (line 894) | def __init__(self, in_, out): method forward (line 899) | def forward(self, x): class ConvBn2d (line 905) | class ConvBn2d(nn.Module): method __init__ (line 906) | def __init__( method forward (line 925) | def forward(self, x): class SpatialAttentionGate (line 931) | class SpatialAttentionGate(nn.Module): method __init__ (line 932) | def __init__(self, channel, reduction=16): method forward (line 937) | def forward(self, x): class UNetResNext_wo_pool_hyper (line 946) | class UNetResNext_wo_pool_hyper(nn.Module): method __init__ (line 947) | def __init__( method forward (line 1007) | def forward(self, x): class UNetResNext50 (line 1039) | class UNetResNext50(nn.Module): method __init__ (line 1040) | def __init__( method forward (line 1102) | def forward(self, x): class UNetResNext (line 1130) | class UNetResNext(nn.Module): method __init__ (line 1131) | def __init__( method forward (line 1205) | def forward(self, x): class UNetPNASNet (line 1224) | class UNetPNASNet(nn.Module): method __init__ (line 1225) | def __init__( method forward (line 1270) | def forward(self, x): class TernausNetV2 (line 1284) | class TernausNetV2(nn.Module): method __init__ (line 1289) | def __init__( method forward (line 1343) | def forward(self, x): class DecoderBlockTernaus (line 1361) | class DecoderBlockTernaus(nn.Module): method __init__ (line 1366) | def __init__(self, in_channels, middle_channels, out_channels, is_deco... method forward (line 1385) | def forward(self, x): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/common_blocks/utils.py function read_yaml (line 33) | def read_yaml(fallback_file=NEPTUNE_CONFIG_PATH): function init_logger (line 39) | def init_logger(): function get_logger (line 58) | def get_logger(): function create_submission (line 62) | def create_submission(meta, predictions): function encode_rle (line 72) | def encode_rle(predictions): function read_masks (line 76) | def read_masks(masks_filepaths): function read_images (line 87) | def read_images(filepaths): function run_length_encoding (line 95) | def run_length_encoding(x): function run_length_decoding (line 110) | def run_length_decoding(mask_rle, shape): function generate_metadata (line 129) | def generate_metadata(train_images_dir, test_images_dir, depths_filepath): function sigmoid (line 159) | def sigmoid(x): function softmax (line 163) | def softmax(X, theta=1.0, axis=None): function from_pil (line 207) | def from_pil(*images): function to_pil (line 215) | def to_pil(*images): function make_apply_transformer (line 223) | def make_apply_transformer(func, output_name="output", apply_on=None): function rle_from_binary (line 273) | def rle_from_binary(prediction): function binary_from_rle (line 278) | def binary_from_rle(rle): function get_segmentations (line 282) | def get_segmentations(labeled): function get_crop_pad_sequence (line 293) | def get_crop_pad_sequence(vertical, horizontal): function get_list_of_image_predictions (line 301) | def get_list_of_image_predictions(batch_predictions): function set_seed (line 308) | def set_seed(seed): class ImgAug (line 316) | class ImgAug: method __init__ (line 317) | def __init__(self, augmenters): method _pre_call_hook (line 323) | def _pre_call_hook(self): method transform (line 328) | def transform(self, *images): method __call__ (line 335) | def __call__(self, *args): function get_seed (line 340) | def get_seed(): function reseed (line 345) | def reseed(augmenter, deterministic=True): class KFoldBySortedValue (line 356) | class KFoldBySortedValue(BaseCrossValidator): method __init__ (line 357) | def __init__(self, n_splits=3, shuffle=False, random_state=None): method _iter_test_indices (line 362) | def _iter_test_indices(self, X, y=None, groups=None): method get_n_splits (line 373) | def get_n_splits(self, X=None, y=None, groups=None): function plot_list (line 377) | def plot_list(images=[], labels=[]): function clean_object_from_memory (line 393) | def clean_object_from_memory(obj): class FineTuneStep (line 400) | class FineTuneStep(Step): method __init__ (line 401) | def __init__( method _cached_fit_transform (line 433) | def _cached_fit_transform(self, step_inputs): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/bn.py function _pair (line 12) | def _pair(x): class ABN (line 18) | class ABN(nn.Sequential): method __init__ (line 24) | def __init__(self, num_features, activation=nn.ReLU(inplace=True), **k... class InPlaceABN (line 43) | class InPlaceABN(nn.Module): method __init__ (line 46) | def __init__( method reset_parameters (line 89) | def reset_parameters(self): method forward (line 96) | def forward(self, x): method __repr__ (line 110) | def __repr__(self): class InPlaceABNWrapper (line 122) | class InPlaceABNWrapper(nn.Module): method __init__ (line 125) | def __init__(self, *args, **kwargs): method forward (line 129) | def forward(self, x): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/functions.py function _check (line 13) | def _check(fn, *args, **kwargs): function _broadcast_shape (line 19) | def _broadcast_shape(x): function _reduce (line 29) | def _reduce(x): function _count_samples (line 37) | def _count_samples(x): function _act_forward (line 45) | def _act_forward(ctx, x): function _act_backward (line 54) | def _act_backward(ctx, x, dx): function _check_contiguous (line 65) | def _check_contiguous(*args): class InPlaceABN (line 70) | class InPlaceABN(autograd.Function): method forward (line 72) | def forward( method backward (line 130) | def backward(ctx, dz): class InPlaceABNSync (line 192) | class InPlaceABNSync(autograd.Function): method forward (line 194) | def forward( method backward (line 279) | def backward(ctx, dz): method _parse_extra (line 361) | def _parse_extra(ctx, extra): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/misc.py class GlobalAvgPool2d (line 4) | class GlobalAvgPool2d(nn.Module): method __init__ (line 5) | def __init__(self): method forward (line 9) | def forward(self, inputs): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/residual.py class IdentityResidualBlock (line 8) | class IdentityResidualBlock(nn.Module): method __init__ (line 9) | def __init__( method forward (line 127) | def forward(self, x): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/src/common.h function __device__ (line 14) | __device__ Pair() {} function __device__ (line 15) | __device__ Pair(T _v1, T _v2) : v1(_v1), v2(_v2) {} function __device__ (line 16) | __device__ Pair(T v) : v1(v), v2(v) {} function __device__ (line 17) | __device__ Pair(int v) : v1(v), v2(v) {} function getMSB (line 38) | int getMSB(int val) { return 31 - __clz(val); } function getNumThreads (line 40) | static int getNumThreads(int nElem) { FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/src/inplace_abn.cpp function mean_var (line 7) | std::vector mean_var(at::Tensor x) { function forward (line 15) | at::Tensor forward(at::Tensor x, at::Tensor mean, at::Tensor var, at::Te... function edz_eydz (line 24) | std::vector edz_eydz(at::Tensor z, at::Tensor dz, at::Tensor... function backward (line 33) | std::vector backward(at::Tensor z, at::Tensor dz, at::Tensor... function leaky_relu_forward (line 42) | void leaky_relu_forward(at::Tensor z, float slope) { function leaky_relu_backward (line 46) | void leaky_relu_backward(at::Tensor z, at::Tensor dz, float slope) { function elu_forward (line 54) | void elu_forward(at::Tensor z) { function elu_backward (line 58) | void elu_backward(at::Tensor z, at::Tensor dz) { function PYBIND11_MODULE (line 66) | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/src/inplace_abn_cpu.cpp function reduce_sum (line 7) | at::Tensor reduce_sum(at::Tensor x) { function broadcast_to (line 16) | at::Tensor broadcast_to(at::Tensor v, at::Tensor x) { function count (line 28) | int64_t count(at::Tensor x) { function invert_affine (line 36) | at::Tensor invert_affine(at::Tensor z, at::Tensor weight, at::Tensor bia... function mean_var_cpu (line 44) | std::vector mean_var_cpu(at::Tensor x) { function forward_cpu (line 53) | at::Tensor forward_cpu(at::Tensor x, at::Tensor mean, at::Tensor var, at... function edz_eydz_cpu (line 65) | std::vector edz_eydz_cpu(at::Tensor z, at::Tensor dz, at::Te... function backward_cpu (line 74) | std::vector backward_cpu(at::Tensor z, at::Tensor dz, at::Te... function leaky_relu_backward_cpu (line 92) | void leaky_relu_backward_cpu(at::Tensor z, at::Tensor dz, float slope) { function elu_backward_cpu (line 107) | void elu_backward_cpu(at::Tensor z, at::Tensor dz) { FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/v2/modules/wider_resnet.py class WiderResNet (line 7) | class WiderResNet(nn.Module): method __init__ (line 8) | def __init__(self, structure, norm_act=ABN, classes=0): method forward (line 78) | def forward(self, img): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/vanilla unet/utils/cyclelr_callback.py class CyclicLR (line 4) | class CyclicLR(Callback): method __init__ (line 63) | def __init__( method _reset (line 99) | def _reset(self, new_base_lr=None, new_max_lr=None, new_step_size=None): method clr (line 111) | def clr(self): method on_train_begin (line 123) | def on_train_begin(self, logs={}): method on_batch_end (line 131) | def on_batch_end(self, epoch, logs=None): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/vanilla unet/utils/lovasz_losses_tf.py function lovasz_grad (line 14) | def lovasz_grad(gt_sorted): function lovasz_hinge (line 30) | def lovasz_hinge(logits, labels, per_image=True, ignore=None): function lovasz_hinge_flat (line 53) | def lovasz_hinge_flat(logits, labels): function flatten_binary_scores (line 86) | def flatten_binary_scores(scores, labels, ignore=None): function lovasz_softmax (line 104) | def lovasz_softmax( function lovasz_softmax_flat (line 133) | def lovasz_softmax_flat(probas, labels, classes="all"): function flatten_probas (line 169) | def flatten_probas(probas, labels, ignore=None, order="BHWC"): FILE: DEEP LEARNING/segmentation/Kaggle TGS Salt Identification Challenge/vanilla unet/utils/zf_unet_224_model.py function preprocess_input (line 30) | def preprocess_input(x): function dice_coef (line 36) | def dice_coef(y_true, y_pred): function jacard_coef (line 43) | def jacard_coef(y_true, y_pred): function jacard_coef_loss (line 52) | def jacard_coef_loss(y_true, y_pred): function dice_coef_loss (line 56) | def dice_coef_loss(y_true, y_pred): function double_conv_layer (line 60) | def double_conv_layer(x, size, dropout=0.0, batch_norm=True): function ZF_UNET_224 (line 78) | def ZF_UNET_224(dropout_val=0.2, weights=None): FILE: DEEP LEARNING/segmentation/Segmentation pipeline/get dataset.py function html_url_parser (line 14) | def html_url_parser(url, save_dir, show=False, wait=False): FILE: DEEP LEARNING/segmentation/Segmentation pipeline/segmentation pipeline.py class DataGeneratorFolder (line 41) | class DataGeneratorFolder(Sequence): method __init__ (line 42) | def __init__(self, root_dir=r'../data/val_test', image_folder='img/', ... method __len__ (line 56) | def __len__(self): method on_epoch_end (line 62) | def on_epoch_end(self): method read_image_mask (line 67) | def read_image_mask(self, image_name, mask_name): method __getitem__ (line 70) | def __getitem__(self, index): function aug_with_crop (line 121) | def aug_with_crop(image_size=256, crop_prob=1): function plot_training_history (line 195) | def plot_training_history(history): function iou_metric (line 224) | def iou_metric(y_true_in, y_pred_in): function plot_mask_gt_image (line 251) | def plot_mask_gt_image(mask, groud_truth, img): function iou_metric_batch (line 259) | def iou_metric_batch(y_true_in, y_pred_in): function draw_get_best_threshold (line 305) | def draw_get_best_threshold(ious, thresholds): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/classification_pytorch_dummy.py function get_annot (line 57) | def get_annot(annot_path): function prev_get_transform (line 72) | def prev_get_transform(phase): function get_transform (line 91) | def get_transform(phase): function criterion (line 110) | def criterion(logit, truth, weight=None): class SteelDataset (line 129) | class SteelDataset(Dataset): method __init__ (line 130) | def __init__(self, annot, image_folder, phase): method __getitem__ (line 136) | def __getitem__(self, index): method __len__ (line 144) | def __len__(self): function get_dataloader (line 148) | def get_dataloader(annot, image_folder, phase, batch_size=16, num_worker... function get_model (line 152) | def get_model(model_name): class Trainer (line 162) | class Trainer: method __init__ (line 163) | def __init__(self, model_name="resnet34", pretrained=False, epochs=1): method forward (line 187) | def forward(self, inputs, targets): method iterate (line 195) | def iterate(self, epoch, phase): method save_model (line 230) | def save_model(self, epoch): method summary (line 241) | def summary(self): method start (line 253) | def start(self): class Flatten (line 284) | class Flatten(nn.Module): method forward (line 285) | def forward(self, x): method forward (line 350) | def forward(self, x): method forward (line 640) | def forward(self, x): method forward (line 705) | def forward(self, x): class ChannelGate (line 287) | class ChannelGate(nn.Module): method __init__ (line 288) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 301) | def forward(self, in_tensor): method __init__ (line 354) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 364) | def forward(self, x): method __init__ (line 643) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 656) | def forward(self, in_tensor): method __init__ (line 709) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 719) | def forward(self, x): class SpatialGate (line 305) | class SpatialGate(nn.Module): method __init__ (line 306) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 318) | def forward(self, in_tensor): method __init__ (line 400) | def __init__(self): method forward (line 405) | def forward(self, x): method __init__ (line 661) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 673) | def forward(self, in_tensor): method __init__ (line 755) | def __init__(self): method forward (line 760) | def forward(self, x): class BAM (line 320) | class BAM(nn.Module): method __init__ (line 321) | def __init__(self, gate_channel): method forward (line 325) | def forward(self,in_tensor): method __init__ (line 676) | def __init__(self, gate_channel): method forward (line 680) | def forward(self,in_tensor): class BasicConv (line 333) | class BasicConv(nn.Module): method __init__ (line 334) | def __init__(self, in_planes, out_planes, kernel_size, stride=1, paddi... method forward (line 341) | def forward(self, x): method __init__ (line 689) | def __init__(self, in_planes, out_planes, kernel_size, stride=1, paddi... method forward (line 696) | def forward(self, x): class Flatten (line 349) | class Flatten(nn.Module): method forward (line 285) | def forward(self, x): method forward (line 350) | def forward(self, x): method forward (line 640) | def forward(self, x): method forward (line 705) | def forward(self, x): class ChannelGate (line 353) | class ChannelGate(nn.Module): method __init__ (line 288) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 301) | def forward(self, in_tensor): method __init__ (line 354) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 364) | def forward(self, x): method __init__ (line 643) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 656) | def forward(self, in_tensor): method __init__ (line 709) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 719) | def forward(self, x): function logsumexp_2d (line 389) | def logsumexp_2d(tensor): class ChannelPool (line 395) | class ChannelPool(nn.Module): method forward (line 396) | def forward(self, x): method forward (line 751) | def forward(self, x): class SpatialGate (line 399) | class SpatialGate(nn.Module): method __init__ (line 306) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 318) | def forward(self, in_tensor): method __init__ (line 400) | def __init__(self): method forward (line 405) | def forward(self, x): method __init__ (line 661) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 673) | def forward(self, in_tensor): method __init__ (line 755) | def __init__(self): method forward (line 760) | def forward(self, x): class CBAM (line 411) | class CBAM(nn.Module): method __init__ (line 412) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 418) | def forward(self, x): method __init__ (line 767) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 773) | def forward(self, x): function conv3x3 (line 431) | def conv3x3(in_planes, out_planes, stride=1): class BasicBlock (line 436) | class BasicBlock(nn.Module): method __init__ (line 439) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 454) | def forward(self, x): method __init__ (line 794) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 809) | def forward(self, x): class Bottleneck (line 475) | class Bottleneck(nn.Module): method __init__ (line 478) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 496) | def forward(self, x): method __init__ (line 833) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 851) | def forward(self, x): class ResNet (line 521) | class ResNet(nn.Module): method __init__ (line 522) | def __init__(self, block, layers, network_type, num_classes, att_type... method _make_layer (line 564) | def _make_layer(self, block, planes, blocks, stride=1, att_type=None): method forward (line 581) | def forward(self, x): method __init__ (line 877) | def __init__(self, block, layers, network_type, num_classes, att_type... method _make_layer (line 919) | def _make_layer(self, block, planes, blocks, stride=1, att_type=None): method forward (line 936) | def forward(self, x): function ResidualNet (line 610) | def ResidualNet(network_type, depth, num_classes, att_type): class Flatten (line 639) | class Flatten(nn.Module): method forward (line 285) | def forward(self, x): method forward (line 350) | def forward(self, x): method forward (line 640) | def forward(self, x): method forward (line 705) | def forward(self, x): class ChannelGate (line 642) | class ChannelGate(nn.Module): method __init__ (line 288) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 301) | def forward(self, in_tensor): method __init__ (line 354) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 364) | def forward(self, x): method __init__ (line 643) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 656) | def forward(self, in_tensor): method __init__ (line 709) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 719) | def forward(self, x): class SpatialGate (line 660) | class SpatialGate(nn.Module): method __init__ (line 306) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 318) | def forward(self, in_tensor): method __init__ (line 400) | def __init__(self): method forward (line 405) | def forward(self, x): method __init__ (line 661) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 673) | def forward(self, in_tensor): method __init__ (line 755) | def __init__(self): method forward (line 760) | def forward(self, x): class BAM (line 675) | class BAM(nn.Module): method __init__ (line 321) | def __init__(self, gate_channel): method forward (line 325) | def forward(self,in_tensor): method __init__ (line 676) | def __init__(self, gate_channel): method forward (line 680) | def forward(self,in_tensor): class BasicConv (line 688) | class BasicConv(nn.Module): method __init__ (line 334) | def __init__(self, in_planes, out_planes, kernel_size, stride=1, paddi... method forward (line 341) | def forward(self, x): method __init__ (line 689) | def __init__(self, in_planes, out_planes, kernel_size, stride=1, paddi... method forward (line 696) | def forward(self, x): class Flatten (line 704) | class Flatten(nn.Module): method forward (line 285) | def forward(self, x): method forward (line 350) | def forward(self, x): method forward (line 640) | def forward(self, x): method forward (line 705) | def forward(self, x): class ChannelGate (line 708) | class ChannelGate(nn.Module): method __init__ (line 288) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 301) | def forward(self, in_tensor): method __init__ (line 354) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 364) | def forward(self, x): method __init__ (line 643) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 656) | def forward(self, in_tensor): method __init__ (line 709) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 719) | def forward(self, x): function logsumexp_2d (line 744) | def logsumexp_2d(tensor): class ChannelPool (line 750) | class ChannelPool(nn.Module): method forward (line 396) | def forward(self, x): method forward (line 751) | def forward(self, x): class SpatialGate (line 754) | class SpatialGate(nn.Module): method __init__ (line 306) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 318) | def forward(self, in_tensor): method __init__ (line 400) | def __init__(self): method forward (line 405) | def forward(self, x): method __init__ (line 661) | def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num... method forward (line 673) | def forward(self, in_tensor): method __init__ (line 755) | def __init__(self): method forward (line 760) | def forward(self, x): class CBAM (line 766) | class CBAM(nn.Module): method __init__ (line 412) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 418) | def forward(self, x): method __init__ (line 767) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg... method forward (line 773) | def forward(self, x): function conv3x3 (line 786) | def conv3x3(in_planes, out_planes, stride=1): class BasicBlock (line 791) | class BasicBlock(nn.Module): method __init__ (line 439) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 454) | def forward(self, x): method __init__ (line 794) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 809) | def forward(self, x): class Bottleneck (line 830) | class Bottleneck(nn.Module): method __init__ (line 478) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 496) | def forward(self, x): method __init__ (line 833) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 851) | def forward(self, x): class ResNet (line 876) | class ResNet(nn.Module): method __init__ (line 522) | def __init__(self, block, layers, network_type, num_classes, att_type... method _make_layer (line 564) | def _make_layer(self, block, planes, blocks, stride=1, att_type=None): method forward (line 581) | def forward(self, x): method __init__ (line 877) | def __init__(self, block, layers, network_type, num_classes, att_type... method _make_layer (line 919) | def _make_layer(self, block, planes, blocks, stride=1, att_type=None): method forward (line 936) | def forward(self, x): function ResidualNet (line 965) | def ResidualNet(network_type, depth, num_classes, att_type): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/bam.py class Flatten (line 7) | class Flatten(nn.Module): method forward (line 8) | def forward(self, x): class ChannelGate (line 12) | class ChannelGate(nn.Module): method __init__ (line 13) | def __init__(self, gate_channel, reduction_ratio=16, num_layers=1): method forward (line 33) | def forward(self, in_tensor): class SpatialGate (line 38) | class SpatialGate(nn.Module): method __init__ (line 39) | def __init__( method forward (line 72) | def forward(self, in_tensor): class BAM (line 76) | class BAM(nn.Module): method __init__ (line 77) | def __init__(self, gate_channel): method forward (line 82) | def forward(self, in_tensor): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/cbam.py class BasicConv (line 7) | class BasicConv(nn.Module): method __init__ (line 8) | def __init__( method forward (line 40) | def forward(self, x): class Flatten (line 49) | class Flatten(nn.Module): method forward (line 50) | def forward(self, x): class ChannelGate (line 54) | class ChannelGate(nn.Module): method __init__ (line 55) | def __init__(self, gate_channels, reduction_ratio=16, pool_types=["avg... method forward (line 66) | def forward(self, x): function logsumexp_2d (line 98) | def logsumexp_2d(tensor): class ChannelPool (line 105) | class ChannelPool(nn.Module): method forward (line 106) | def forward(self, x): class SpatialGate (line 112) | class SpatialGate(nn.Module): method __init__ (line 113) | def __init__(self): method forward (line 121) | def forward(self, x): class CBAM (line 128) | class CBAM(nn.Module): method __init__ (line 129) | def __init__( method forward (line 142) | def forward(self, x): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/dataloader.py class SteelDataset (line 45) | class SteelDataset(Dataset): method __init__ (line 46) | def __init__(self, df, data_folder, mean, std, phase, df_full, data_fo... method __getitem__ (line 58) | def __getitem__(self, idx): method __len__ (line 76) | def __len__(self): function get_transforms (line 80) | def get_transforms(phase, mean, std): function provider_trai_test_split (line 111) | def provider_trai_test_split( function provider_cv (line 141) | def provider_cv( function provider_cv___ (line 198) | def provider_cv___( FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/logger.py function flatten (line 19) | def flatten(l): function getargnames (line 29) | def getargnames(func): function getcallargs_ordered (line 38) | def getcallargs_ordered(func, *args, **kwargs): function describe_call (line 46) | def describe_call(func, *args, **kwargs): function log_to (line 55) | def log_to(logger_func): function timeit (line 81) | def timeit(method): function debug (line 100) | def debug(fn): function myfunc (line 115) | def myfunc(a, b, c, *args, **kwargs): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/losses.py class FocalLoss (line 10) | class FocalLoss(nn.Module): method __init__ (line 11) | def __init__(self, alpha=1, gamma=2, logits=False, reduce=True): method forward (line 18) | def forward(self, inputs, targets): class JaccardLoss (line 32) | class JaccardLoss(nn.Module): method __init__ (line 35) | def __init__(self, eps=1e-7, activation="sigmoid"): method forward (line 40) | def forward(self, y_pr, y_gt): class DiceLoss (line 46) | class DiceLoss(nn.Module): method __init__ (line 49) | def __init__(self, eps=1e-7, activation="sigmoid"): method forward (line 54) | def forward(self, y_pr, y_gt): class BCEJaccardLoss (line 65) | class BCEJaccardLoss(JaccardLoss): method __init__ (line 68) | def __init__(self, eps=1e-7, activation="sigmoid"): method forward (line 72) | def forward(self, y_pr, y_gt): class BCEDiceLoss (line 78) | class BCEDiceLoss(DiceLoss): method __init__ (line 81) | def __init__(self, eps=1e-7, activation="sigmoid"): method forward (line 85) | def forward(self, y_pr, y_gt): function iou (line 91) | def iou(pr, gt, eps=1e-7, threshold=None, activation="sigmoid"): function f_score (line 126) | def f_score(pr, gt, beta=1, eps=1e-7, threshold=None, activation="sigmoi... FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/lovasz_losses.py function _lovasz_grad (line 19) | def _lovasz_grad(gt_sorted): function _lovasz_hinge (line 33) | def _lovasz_hinge(logits, labels, per_image=True, ignore=None): function _lovasz_hinge_flat (line 53) | def _lovasz_hinge_flat(logits, labels): function _flatten_binary_scores (line 73) | def _flatten_binary_scores(scores, labels, ignore=None): function _lovasz_softmax (line 90) | def _lovasz_softmax( function _lovasz_softmax_flat (line 118) | def _lovasz_softmax_flat(probas, labels, classes="present", anti=False): function _flatten_probas (line 152) | def _flatten_probas(probas, labels, ignore=None, anti=False): function isnan (line 171) | def isnan(x): function mean (line 175) | def mean(values, ignore_nan=False, empty=0): class BinaryLovaszLoss (line 195) | class BinaryLovaszLoss(_Loss): method __init__ (line 196) | def __init__(self, per_image=False, ignore=None): method forward (line 201) | def forward(self, logits, target): class LovaszLoss (line 207) | class LovaszLoss(_Loss): method __init__ (line 208) | def __init__(self, per_image=False, ignore=None, anti=False, classes="... method forward (line 215) | def forward(self, logits, target): class LovaszLossSymmetric (line 230) | class LovaszLossSymmetric(_Loss): method __init__ (line 231) | def __init__(self, per_image=True, ignore=None, classes="all"): method forward (line 237) | def forward(self, logits, target): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/metric.py class Meter (line 18) | class Meter: method __init__ (line 21) | def __init__(self, phase, epoch): method update (line 28) | def update(self, targets, outputs): method get_metrics (line 40) | def get_metrics(self): function predict (line 49) | def predict(X, threshold): function metric_old (line 56) | def metric_old(probability, truth, threshold=0.5): function dice_channel_torch (line 92) | def dice_channel_torch(probability, truth, threshold): function dice_single_channel (line 119) | def dice_single_channel(probability, truth, threshold, batch_size, eps=1... function epoch_log (line 126) | def epoch_log(phase, epoch, epoch_loss, meter, start): function compute_ious (line 133) | def compute_ious(pred, label, classes, ignore_index=255, only_present=Tr... function compute_iou_batch (line 150) | def compute_iou_batch(outputs, labels, classes=None): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/new_metrics.py function pixel_accuracy (line 30) | def pixel_accuracy(outputs: torch.Tensor, targets: torch.Tensor, ignore_... class PixelAccuracyCallback (line 48) | class PixelAccuracyCallback(MetricCallback): method __init__ (line 52) | def __init__( class ConfusionMatrixCallback (line 74) | class ConfusionMatrixCallback(Callback): method __init__ (line 80) | def __init__( method on_loader_start (line 104) | def on_loader_start(self, state): method on_batch_end (line 108) | def on_batch_end(self, state: RunnerState): method on_loader_end (line 123) | def on_loader_end(self, state): class MacroF1Callback (line 214) | class MacroF1Callback(Callback): method __init__ (line 219) | def __init__( method on_batch_end (line 243) | def on_batch_end(self, state: RunnerState): method on_loader_start (line 267) | def on_loader_start(self, state): method on_loader_end (line 271) | def on_loader_end(self, state): function binary_dice_iou_score (line 280) | def binary_dice_iou_score( function multiclass_dice_iou_score (line 324) | def multiclass_dice_iou_score( function multilabel_dice_iou_score (line 354) | def multilabel_dice_iou_score( class IoUMetricsCallback (line 383) | class IoUMetricsCallback(Callback): method __init__ (line 389) | def __init__( method on_loader_start (line 457) | def on_loader_start(self, state): method on_batch_end (line 461) | def on_batch_end(self, state: RunnerState): method on_loader_end (line 477) | def on_loader_end(self, state): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/optimizers.py class RAdam (line 6) | class RAdam(Optimizer): method __init__ (line 7) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weig... method __setstate__ (line 12) | def __setstate__(self, state): method step (line 15) | def step(self, closure=None): class PlainRAdam (line 93) | class PlainRAdam(Optimizer): method __init__ (line 94) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weig... method __setstate__ (line 99) | def __setstate__(self, state): method step (line 102) | def step(self, closure=None): class AdamW (line 169) | class AdamW(Optimizer): method __init__ (line 170) | def __init__( method __setstate__ (line 178) | def __setstate__(self, state): method step (line 181) | def step(self, closure=None): method __init__ (line 276) | def __init__( method __setstate__ (line 298) | def __setstate__(self, state): method step (line 303) | def step(self, closure=None): class AdamW (line 253) | class AdamW(Optimizer): method __init__ (line 170) | def __init__( method __setstate__ (line 178) | def __setstate__(self, state): method step (line 181) | def step(self, closure=None): method __init__ (line 276) | def __init__( method __setstate__ (line 298) | def __setstate__(self, state): method step (line 303) | def step(self, closure=None): class Novograd (line 368) | class Novograd(Optimizer): method __init__ (line 386) | def __init__( method __setstate__ (line 415) | def __setstate__(self, state): method step (line 420) | def step(self, closure=None): class Lookahead (line 494) | class Lookahead(Optimizer): method __init__ (line 495) | def __init__(self, base_optimizer, alpha=0.5, k=6): method step (line 513) | def step(self, closure=None): function LookaheadAdam (line 530) | def LookaheadAdam(params, alpha=0.5, k=6, *args, **kwargs): class Ralamb (line 540) | class Ralamb(Optimizer): method __init__ (line 541) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weig... method __setstate__ (line 546) | def __setstate__(self, state): method step (line 549) | def step(self, closure=None): function Over9000 (line 648) | def Over9000(params, alpha=0.5, k=6, *args, **kwargs): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/training_helper.py class Trainer_cv (line 31) | class Trainer_cv(object): method __init__ (line 34) | def __init__( method forward (line 90) | def forward(self, images, targets): method iterate (line 97) | def iterate(self, epoch, phase): method start (line 131) | def start(self): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/common_blocks/utils.py function mask2rle (line 16) | def mask2rle(img): function make_mask (line 28) | def make_mask(row_id, df): function plot (line 47) | def plot(scores, name, fold=0, safe_pic=True): function set_seed (line 62) | def set_seed(seed=42): function load_model_unet (line 70) | def load_model_unet(_model_weights, is_inference=False): function load_model_fpn (line 111) | def load_model_fpn(_model_weights, is_inference=False): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/inference.py class Model (line 147) | class Model: method __init__ (line 148) | def __init__(self, models): method __call__ (line 151) | def __call__(self, x): function create_transforms (line 186) | def create_transforms(additional): class ConvBn2d (line 1032) | class ConvBn2d(nn.Module): method __init__ (line 1033) | def __init__(self, in_channel, out_channel, kernel_size=3, padding=1, ... method forward (line 1045) | def forward(self, x): class BasicBlock (line 1058) | class BasicBlock(nn.Module): method __init__ (line 1059) | def __init__(self, in_channel, channel, out_channel, stride=1, is_shor... method forward (line 1075) | def forward(self, x): class ResNet34 (line 1087) | class ResNet34(nn.Module): method __init__ (line 1088) | def __init__(self, num_class=1000): method forward (line 1124) | def forward(self, x): class Resnet34_classification (line 1137) | class Resnet34_classification(nn.Module): method __init__ (line 1138) | def __init__(self, num_class=4): method forward (line 1146) | def forward(self, x): class TestDataset (line 1170) | class TestDataset(Dataset): method __init__ (line 1173) | def __init__(self, root, df, mean, std): method __getitem__ (line 1180) | def __getitem__(self, idx): method __len__ (line 1187) | def __len__(self): function sharpen (line 1191) | def sharpen(p, t=0.5): function get_classification_preds (line 1201) | def get_classification_preds(net, test_loader): FILE: DEEP LEARNING/segmentation/Severstal-Steel-Defect-Detection-master/model_resnet.py function conv3x3 (line 10) | def conv3x3(in_planes, out_planes, stride=1): class BasicBlock (line 17) | class BasicBlock(nn.Module): method __init__ (line 20) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 35) | def forward(self, x): class Bottleneck (line 57) | class Bottleneck(nn.Module): method __init__ (line 60) | def __init__(self, inplanes, planes, stride=1, downsample=None, use_cb... method forward (line 79) | def forward(self, x): class ResNet (line 105) | class ResNet(nn.Module): method __init__ (line 106) | def __init__(self, block, layers, network_type, num_classes, att_type=... method _make_layer (line 158) | def _make_layer(self, block, planes, blocks, stride=1, att_type=None): method forward (line 184) | def forward(self, x): function ResidualNet (line 214) | def ResidualNet(network_type, depth, num_classes, att_type): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/augs.py function to_tensor (line 4) | def to_tensor(x, **kwargs): function get_training_augmentation (line 19) | def get_training_augmentation( function get_training_augmentation0 (line 44) | def get_training_augmentation0(image_size: tuple = (320, 640)): function get_training_augmentation1 (line 66) | def get_training_augmentation1(image_size: tuple = (320, 640)): function get_training_augmentation2 (line 87) | def get_training_augmentation2(image_size: tuple = (320, 640)): function get_validation_augmentation (line 110) | def get_validation_augmentation(image_size: tuple = (320, 640)): function get_preprocessing (line 123) | def get_preprocessing(preprocessing_fn): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/callbacks.py function calculate_confusion_matrix_from_arrays (line 9) | def calculate_confusion_matrix_from_arrays( function get_confusion_matrix (line 35) | def get_confusion_matrix(y_pred_logits: torch.Tensor, y_true: torch.Tens... function calculate_tp_fp_fn (line 44) | def calculate_tp_fp_fn(confusion_matrix): function calculate_dice (line 65) | def calculate_dice(tp_fp_fn_dict): class MulticlassDiceMetricCallback (line 83) | class MulticlassDiceMetricCallback(Callback): method __init__ (line 84) | def __init__( method _reset_stats (line 102) | def _reset_stats(self): method on_batch_end (line 105) | def on_batch_end(self, state: RunnerState): method on_loader_end (line 116) | def on_loader_end(self, state: RunnerState): class CustomSegmentationInferCallback (line 138) | class CustomSegmentationInferCallback(Callback): method __init__ (line 139) | def __init__(self, return_valid: bool = False): method on_batch_end (line 145) | def on_batch_end(self, state: RunnerState): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/dataset.py function get_img (line 21) | def get_img(x: str = "img_name", folder: str = "train_images"): function rle_decode (line 38) | def rle_decode(mask_rle: str = "", shape: tuple = (1400, 2100)): function make_mask (line 60) | def make_mask( function mask2rle (line 86) | def mask2rle(img): class CloudDataset (line 103) | class CloudDataset(Dataset): method __init__ (line 104) | def __init__( method save_processed_ (line 171) | def save_processed_(self): method __getitem__ (line 187) | def __getitem__(self, idx): method __len__ (line 208) | def __len__(self): class CloudDatasetClassification (line 212) | class CloudDatasetClassification(Dataset): method __init__ (line 213) | def __init__( method save_processed_ (line 281) | def save_processed_(self): method __getitem__ (line 291) | def __getitem__(self, idx): method __len__ (line 310) | def __len__(self): function prepare_loaders (line 314) | def prepare_loaders( FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/inference_blend.py class Model (line 36) | class Model: method __init__ (line 37) | def __init__(self, models): method __call__ (line 40) | def __call__(self, x): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/losses/losses.py class FocalLoss (line 9) | class FocalLoss(_Loss): method __init__ (line 10) | def __init__(self, alpha=0.5, gamma=2, ignore_index=None): method forward (line 24) | def forward(self, label_input, label_target): class MulticlassDiceLoss (line 46) | class MulticlassDiceLoss(_Loss): method __init__ (line 50) | def __init__( method forward (line 62) | def forward(self, y_pred: torch.Tensor, y_true: torch.Tensor) -> torch... class BCEMulticlassDiceLoss (line 110) | class BCEMulticlassDiceLoss(MulticlassDiceLoss): method __init__ (line 113) | def __init__(self, eps=1e-7, activation="sigmoid"): method forward (line 117) | def forward(self, y_pr, y_gt): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/losses/lovasz_losses.py function lovasz_grad (line 19) | def lovasz_grad(gt_sorted): function iou_binary (line 34) | def iou_binary(preds, labels, EMPTY=1.0, ignore=None, per_image=True): function iou (line 54) | def iou(preds, labels, C, EMPTY=1.0, ignore=None, per_image=False): function lovasz_hinge (line 81) | def lovasz_hinge(logits, labels, per_image=True, ignore=None): function lovasz_hinge_flat (line 101) | def lovasz_hinge_flat(logits, labels): function flatten_binary_scores (line 121) | def flatten_binary_scores(scores, labels, ignore=None): class StableBCELoss (line 136) | class StableBCELoss(torch.nn.modules.Module): method __init__ (line 137) | def __init__(self): method forward (line 140) | def forward(self, input, target): function binary_xloss (line 146) | def binary_xloss(logits, labels, ignore=None): function lovasz_softmax (line 161) | def lovasz_softmax(probas, labels, classes="present", per_image=False, i... function lovasz_softmax_flat (line 186) | def lovasz_softmax_flat(probas, labels, classes="present"): function flatten_probas (line 217) | def flatten_probas(probas, labels, ignore=None): function xloss (line 236) | def xloss(logits, labels, ignore=None): function isnan (line 244) | def isnan(x): function mean (line 248) | def mean(l, ignore_nan=False, empty=0): FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/optimizers.py class Ranger (line 11) | class Ranger(Optimizer): method __init__ (line 15) | def __init__( method __setstate__ (line 64) | def __setstate__(self, state): method step (line 68) | def step(self, closure=None): class RAdam (line 164) | class RAdam(Optimizer): method __init__ (line 165) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weig... method __setstate__ (line 170) | def __setstate__(self, state): method step (line 173) | def step(self, closure=None): class Lookahead (line 249) | class Lookahead(Optimizer): method __init__ (line 250) | def __init__(self, base_optimizer, alpha=0.5, k=6): method step (line 268) | def step(self, closure=None): class Ralamb (line 285) | class Ralamb(Optimizer): method __init__ (line 290) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weig... method __setstate__ (line 295) | def __setstate__(self, state): method step (line 298) | def step(self, closure=None): function get_optimizer (line 391) | def get_optimizer( FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/predict.py function sigmoid (line 9) | def sigmoid(x): function predict (line 13) | def predict( function predict_blend (line 65) | def predict_blend( FILE: DEEP LEARNING/segmentation/Understanding-Clouds-from-Satellite-Images-master/utils.py function sigmoid (line 8) | def sigmoid(x): function visualize (line 12) | def visualize(image, mask, original_image=None, original_mask=None, font... function visualize_with_raw (line 56) | def visualize_with_raw( function plot_with_augmentation (line 102) | def plot_with_augmentation(image, mask, augment): function post_process (line 123) | def post_process( function dice (line 151) | def dice(img1: np.array, img2: np.array) -> float: function get_optimal_postprocess (line 169) | def get_optimal_postprocess(loaders=None, runner=None, logdir: str = ""): FILE: classification/Kaggle Malware Prediction/kaggle.py function fast_auc (line 21) | def fast_auc(y_true, y_prob): function eval_auc (line 35) | def eval_auc(preds, dtrain): class Kaggle (line 40) | class Kaggle: method __init__ (line 49) | def __init__(self, data_path, metric="auc", mode=0): method get_metric (line 148) | def get_metric(metric): method read_train_data (line 155) | def read_train_data( method read_test_data (line 189) | def read_test_data(self, test_name="test.csv"): method reduce_mem_usage (line 195) | def reduce_mem_usage(data, verbose=True): method create_validation_split (line 248) | def create_validation_split(self, n_folds=5, stratified=False): method general_feature_engineering (line 280) | def general_feature_engineering(self, train_only=True): method _categorical_preprocess (line 634) | def _categorical_preprocess(self, df, cat_feature, how="ohe"): method fold_feature_engineering (line 655) | def fold_feature_engineering(self, train, test, total_test): method get_predictions (line 743) | def get_predictions(self, model_name, params, X_train, y_train, X_test... method plot_feature_importance (line 802) | def plot_feature_importance(self, feature_importance): method run_single_model_validation_test_pred (line 822) | def run_single_model_validation_test_pred( method run_single_model_validation (line 930) | def run_single_model_validation( method run_stacked_model_validation (line 990) | def run_stacked_model_validation( method find_optimal_params (line 1073) | def find_optimal_params(self, model_name="xgboost"): method get_single_model_test_prediction (line 1092) | def get_single_model_test_prediction( method get_stacked_model_test_prediction (line 1116) | def get_stacked_model_test_prediction( FILE: classification/Kaggle Malware Prediction/models.py function fast_auc (line 5) | def fast_auc(y_true, y_prob): function eval_auc (line 19) | def eval_auc(preds, dtrain): function predict_chunk (line 25) | def predict_chunk(model, test): function train_model (line 40) | def train_model( FILE: classification/Kaggle Malware Prediction/models_zoo.py class BesXGboost (line 14) | class BesXGboost: method __init__ (line 56) | def __init__( method fit (line 96) | def fit(self, X_train, y_train): method predict (line 140) | def predict(self, X_test): method feature_importance (line 145) | def feature_importance(self): method _optimize_single_param (line 162) | def _optimize_single_param(self): method find_best_params (line 166) | def find_best_params(kag): class BesLightGBM (line 360) | class BesLightGBM: method __init__ (line 376) | def __init__(self, params, metric="auc", maximize=True, verbose=True, ... method fit (line 383) | def fit(self, X_train, y_train): method predict (line 426) | def predict(self, X_test): method feature_importance (line 430) | def feature_importance(self): method find_best_params (line 436) | def find_best_params(kag): class BesCatBoost (line 440) | class BesCatBoost: method __init__ (line 456) | def __init__(self, params, metric="AUC", maximize=True, verbose=True, ... method fit (line 463) | def fit(self, X_train, y_train): method predict (line 475) | def predict(self, X_test): method feature_importance (line 479) | def feature_importance(self): method find_best_params (line 483) | def find_best_params(kag): FILE: classification/Kaggle Malware Prediction/target_encoding.py class TargetEncoding (line 6) | class TargetEncoding(object): method __init__ (line 12) | def __init__(self, C=10): method fit (line 15) | def fit(self, data_train, data_test, total_test, feature, target): FILE: classification/Kaggle Petfinder/8th-place-solution-code.py class PetFinderParser (line 90) | class PetFinderParser(object): method __init__ (line 91) | def __init__(self, debug=False): method open_metadata_file (line 99) | def open_metadata_file(self, filename): method open_sentiment_file (line 107) | def open_sentiment_file(self, filename): method open_image_file (line 115) | def open_image_file(self, filename): method parse_sentiment_file (line 122) | def parse_sentiment_file(self, file): method parse_metadata_file (line 158) | def parse_metadata_file(self, file): function confusion_matrix (line 205) | def confusion_matrix(rater_a, rater_b, min_rating=None, max_rating=None): function histogram (line 221) | def histogram(ratings, min_rating=None, max_rating=None): function quadratic_weighted_kappa (line 236) | def quadratic_weighted_kappa(y, y_pred): class OptimizedRounder (line 268) | class OptimizedRounder(object): method __init__ (line 269) | def __init__(self): method _kappa_loss (line 272) | def _kappa_loss(self, coef, X, y): method fit (line 289) | def fit(self, X, y): method predict (line 296) | def predict(self, X, coef): method coefficients (line 311) | def coefficients(self): function rmse (line 315) | def rmse(actual, predicted): function extract_additional_features (line 320) | def extract_additional_features(pet_id, mode="train"): function set_seed (line 352) | def set_seed(seed=0): function faith (line 362) | def faith(title): function reduce_mem_usage (line 368) | def reduce_mem_usage(df, verbose=True): function clean_name (line 408) | def clean_name(x): function relative_age (line 426) | def relative_age(cols): function VerifibalePhotoAmy (line 436) | def VerifibalePhotoAmy(number): function seo_value (line 444) | def seo_value(cols): function genuine_name (line 451) | def genuine_name(cols): function rankbyG (line 463) | def rankbyG(alldata, group): function get_new_columns (line 475) | def get_new_columns(name, aggs): function agg_features (line 479) | def agg_features(df, groupby, agg, prefix): function bounding_features (line 485) | def bounding_features( function open_breeds_info_file (line 579) | def open_breeds_info_file(filename): function parse_sentiment_file (line 585) | def parse_sentiment_file(file): function resize_to_square (line 603) | def resize_to_square(im, img_size): function load_image (line 620) | def load_image(path): function load_image2 (line 627) | def load_image2(path, image_size): function getSize (line 634) | def getSize(filename): function getDimensions (line 639) | def getDimensions(filename): function meta_nlp_feats (line 644) | def meta_nlp_feats(df, col): function load_tabular_data (line 680) | def load_tabular_data(): function load_image_data (line 711) | def load_image_data(): function load_metadata (line 737) | def load_metadata(): function load_sentiment_data (line 763) | def load_sentiment_data(): function build_model (line 788) | def build_model( function train_model (line 803) | def train_model(model, train, test, nn_params={"batch_size": 64, "img_si... function image_feature (line 872) | def image_feature(model, train, test, nn_params={"batch_size": 64, "img_... function basic_features (line 951) | def basic_features(train, test): function image_dim_features (line 1287) | def image_dim_features(train, test): function metadata_features (line 1349) | def metadata_features(train, test): function breed_maps (line 1508) | def breed_maps(train_proc, test_proc, labels_breed): function nlp_features (line 1555) | def nlp_features(X_temp): function run_lgbm (line 1595) | def run_lgbm(X_temp, test): function run_xgb (line 1738) | def run_xgb(X_temp, test): function add_noise (line 1876) | def add_noise(series, noise_level): function target_encode (line 1880) | def target_encode( FILE: deployment/docker flask fit predict/hello.py function hello (line 21) | def hello(): function squar_val (line 28) | def squar_val(username): function average (line 33) | def average(lst): function avg (line 38) | def avg(nums): function fit_predict_iris (line 45) | def fit_predict_iris(params): function show_image (line 56) | def show_image(): function add_message (line 62) | def add_message(): function bad_request (line 79) | def bad_request(): class MyForm (line 97) | class MyForm(FlaskForm): function submit (line 103) | def submit(): function allowed_file (line 131) | def allowed_file(filename): function upload_file (line 136) | def upload_file(): FILE: deployment/ds docker db template/docker/postgres/initdb.sql type d_date (line 3) | CREATE TABLE d_date type d_date_date_actual_idx (line 38) | CREATE INDEX d_date_date_actual_idx FILE: general studies/get feature importance.py function mutual_incoherence (line 39) | def mutual_incoherence(X_relevant, X_irelevant): FILE: time series regression/ARIMA/AR.py function plotds (line 11) | def plotds(xt, nlag=30, fig_size=(12, 10)): FILE: time series regression/ARIMA/ARIMA.py function plotds (line 13) | def plotds(xt, nlag=30, fig_size=(12, 10)): FILE: time series regression/ARIMA/ARMA.py function plotds (line 11) | def plotds(xt, nlag=30, fig_size=(12, 10)): FILE: time series regression/ARIMA/MA.py function plotds (line 10) | def plotds(xt, nlag=30, fig_size=(12, 10)): FILE: time series regression/anomaly detection/anomaly-detection-using-facebook-s-prophet.py function add_autoincrement (line 65) | def add_autoincrement(render_func): function render (line 88) | def render(chart, id="vega-chart"): function fit_predict_model (line 226) | def fit_predict_model(dataframe, interval_width=0.99, changepoint_range=... function detect_anomalies (line 255) | def detect_anomalies(forecast): function plot_anomalies (line 291) | def plot_anomalies(forecasted): FILE: time series regression/autocorelation, mov avg etc/doubleExponentialSmoothing.py function double_exp_smoothing (line 34) | def double_exp_smoothing(x, alpha, beta): function single_exp_smoothing (line 59) | def single_exp_smoothing(x, alpha): FILE: time series regression/autocorelation, mov avg etc/simpleExponentialSmoothing.py function single_exp_smoothing (line 18) | def single_exp_smoothing(x, alpha): FILE: time series regression/autocorelation, mov avg etc/tripleExponentialSmoothing.py function initialize_T (line 21) | def initialize_T(x, seasonLength): function initialize_seasonalilty (line 31) | def initialize_seasonalilty(x, seasonLength): function triple_exp_smoothing (line 55) | def triple_exp_smoothing(x, seasonLength, alpha, beta, gamma, h):