SYMBOL INDEX (1092 symbols across 78 files)

FILE: LEO/config.py
  function get_data_config (line 121) | def get_data_config():
  function get_inner_model_config (line 131) | def get_inner_model_config():
  function get_outer_model_config (line 149) | def get_outer_model_config():
  function load_ifsl_config (line 165) | def load_ifsl_config(config):

FILE: LEO/data.py
  class StrEnum (line 39) | class StrEnum(enum.Enum):
    method __str__ (line 42) | def __str__(self):
    method __repr__ (line 45) | def __repr__(self):
  class MetaDataset (line 49) | class MetaDataset(StrEnum):
  class EmbeddingCrop (line 55) | class EmbeddingCrop(StrEnum):
  class MetaSplit (line 61) | class MetaSplit(StrEnum):
  class DataProvider (line 68) | class DataProvider(object):
    method __init__ (line 71) | def __init__(self, dataset_split, config, verbose=False, feat_dim=640,...
    method _check_config (line 80) | def _check_config(self):
    method _load_data (line 91) | def _load_data(self):
    method _load (line 112) | def _load(self, opened_file):
    method _index_data (line 119) | def _index_data(self, raw_data):
    method _check_data_index (line 143) | def _check_data_index(self, raw_data):
    method _get_full_pickle_path (line 161) | def _get_full_pickle_path(self, split_name):
    method get_instance (line 179) | def get_instance(self, num_classes, tr_size, val_size):
    method get_batch (line 269) | def get_batch(self, batch_size, num_classes, tr_size, val_size,
    method _check_labels (line 320) | def _check_labels(self, num_classes, tr_size, val_size,

FILE: LEO/ifsl_configs/baseline_config.py
  class Config (line 1) | class Config():
    method __init__ (line 2) | def __init__(self):
  function mini_5_resnet_baseline (line 6) | def mini_5_resnet_baseline():
  function mini_1_resnet_baseline (line 18) | def mini_1_resnet_baseline():
  function mini_5_wrn_baseline (line 30) | def mini_5_wrn_baseline():
  function mini_1_wrn_baseline (line 42) | def mini_1_wrn_baseline():
  function tiered_5_resnet_baseline (line 54) | def tiered_5_resnet_baseline():
  function tiered_1_resnet_baseline (line 67) | def tiered_1_resnet_baseline():
  function tiered_5_wrn_baseline (line 79) | def tiered_5_wrn_baseline():
  function tiered_1_wrn_baseline (line 92) | def tiered_1_wrn_baseline():

FILE: LEO/ifsl_configs/ifsl_config.py
  class Config (line 1) | class Config():
    method __init__ (line 2) | def __init__(self):
  function mini_5_resnet_ifsl (line 5) | def mini_5_resnet_ifsl():
  function mini_1_resnet_ifsl (line 31) | def mini_1_resnet_ifsl():
  function mini_5_wrn_ifsl (line 56) | def mini_5_wrn_ifsl():
  function mini_1_wrn_ifsl (line 82) | def mini_1_wrn_ifsl():
  function tiered_5_resnet_ifsl (line 108) | def tiered_5_resnet_ifsl():
  function tiered_1_resnet_ifsl (line 134) | def tiered_1_resnet_ifsl():
  function tiered_5_wrn_ifsl (line 160) | def tiered_5_wrn_ifsl():
  function tiered_1_wrn_ifsl (line 185) | def tiered_1_wrn_ifsl():

FILE: LEO/model.py
  function get_orthogonality_regularizer (line 36) | def get_orthogonality_regularizer(orthogonality_penalty_weight):
  function run_leo (line 60) | def run_leo(model, inputs, is_meta_training):
  class FeatureProcessor (line 74) | class FeatureProcessor():
    method __init__ (line 75) | def __init__(self, n_splits, pretrain_mean_filename, feat_dim, is_cosi...
    method get_d_features (line 96) | def get_d_features(self, logit):
    method preprocess (line 101) | def preprocess(self, data, center=None, method="none"):
    method get_split_features (line 111) | def get_split_features(self, data, center, method="none"):
    method get_features (line 126) | def get_features(self, data):
  class IFSL (line 158) | class IFSL(snt.AbstractModule):
    method __init__ (line 159) | def __init__(self, config=None, use_64bits_dtype=True, n_splits=4, is_...
    method build_input_data (line 191) | def build_input_data(self, support_features, query_features, data):
    method get_debug_data (line 203) | def get_debug_data(self, data):
    method fuse_features (line 210) | def fuse_features(self, x1, x2):
    method _build (line 219) | def _build(self, data, is_meta_training=True, debug=False, break_down=...
    method calculate_dacc (line 291) | def calculate_dacc(self, data, model_prediction, break_down=False):
    method grads_and_vars (line 330) | def grads_and_vars(self, metatrain_loss):
  class LEO (line 339) | class LEO(snt.AbstractModule):
    method __init__ (line 342) | def __init__(self, config=None, use_64bits_dtype=True, name="leo", dec...
    method _build (line 364) | def _build(self, data, is_meta_training=True):
    method leo_inner_loop (line 419) | def leo_inner_loop(self, data, latents):
    method finetuning_inner_loop (line 442) | def finetuning_inner_loop(self, data, leo_loss, classifier_weights):
    method forward_encoder (line 460) | def forward_encoder(self, data):
    method forward_decoder (line 468) | def forward_decoder(self, data, latents):
    method encoder (line 483) | def encoder(self, inputs):
    method relation_network (line 498) | def relation_network(self, inputs):
    method decoder (line 524) | def decoder(self, inputs):
    method average_codes_per_class (line 541) | def average_codes_per_class(self, codes):
    method possibly_sample (line 547) | def possibly_sample(self, distribution_params, stddev_offset=0.):
    method kl_divergence (line 562) | def kl_divergence(self, samples, normal_distribution):
    method predict (line 569) | def predict(self, inputs, weights):
    method calculate_inner_loss (line 580) | def calculate_inner_loss(self, inputs, true_outputs, classifier_weights):
    method save_problem_instance_stats (line 589) | def save_problem_instance_stats(self, instance):
    method dropout_rate (line 607) | def dropout_rate(self):
    method loss_fn (line 610) | def loss_fn(self, model_outputs, original_classes):
    method grads_and_vars (line 617) | def grads_and_vars(self, metatrain_loss):
    method _l2_regularization (line 656) | def _l2_regularization(self):
    method _decoder_orthogonality_reg (line 662) | def _decoder_orthogonality_reg(self):

FILE: LEO/model_test.py
  function get_test_config (line 36) | def get_test_config():
  function mockify_everything (line 53) | def mockify_everything(test_function=None,
  function _random_problem_instance (line 126) | def _random_problem_instance(num_classes=7,
  class LEOTest (line 149) | class LEOTest(tf.test.TestCase, parameterized.TestCase):
    method setUp (line 151) | def setUp(self):
    method test_instantiate_leo (line 160) | def test_instantiate_leo(self):
    method test_inner_loop_adaptation (line 168) | def test_inner_loop_adaptation(self):
    method test_map_input (line 193) | def test_map_input(self):
    method test_setting_is_meta_training (line 230) | def test_setting_is_meta_training(self):
    method test_finetuning_improves_loss (line 237) | def test_finetuning_improves_loss(self):
    method test_gradients_dont_flow_through_input (line 254) | def test_gradients_dont_flow_through_input(self):
    method test_inferring_embedding_dim (line 262) | def test_inferring_embedding_dim(self):
    method test_variable_creation (line 267) | def test_variable_creation(self):
    method test_graph_construction (line 284) | def test_graph_construction(self):
    method test_possibly_sample (line 287) | def test_possibly_sample(self):
    method test_different_shapes (line 308) | def test_different_shapes(self):
    method test_encoder_penalty (line 315) | def test_encoder_penalty(self):
    method test_construct_float32_leo_graph (line 332) | def test_construct_float32_leo_graph(self):

FILE: LEO/runner.py
  function _clip_gradients (line 49) | def _clip_gradients(gradients, gradient_threshold, gradient_norm_thresho...
  function _construct_validation_summaries (line 63) | def _construct_validation_summaries(metavalid_loss, metavalid_accuracy):
  function _construct_training_summaries (line 69) | def _construct_training_summaries(metatrain_loss, metatrain_accuracy,
  function _construct_examples_batch (line 80) | def _construct_examples_batch(batch_size, split, num_classes,
  function _construct_loss_and_accuracy (line 92) | def _construct_loss_and_accuracy(inner_model, inputs, is_meta_training):
  function construct_debug_graph (line 107) | def construct_debug_graph(outer_model_config):
  function construct_graph (line 126) | def construct_graph(outer_model_config):
  function run_debug_loop (line 208) | def run_debug_loop(checkpoint_path):
  function write_output_message (line 227) | def write_output_message(message, file_name=None):
  function run_training_loop (line 235) | def run_training_loop(checkpoint_path):
  function main (line 309) | def main(argv):

FILE: LEO/utils.py
  function unpack_data (line 31) | def unpack_data(problem_instance):
  function copy_checkpoint (line 38) | def copy_checkpoint(checkpoint_path, global_step, accuracy):
  function _save_files_in_tmp_directory (line 69) | def _save_files_in_tmp_directory(tmp_checkpoint_path, checkpoint_files,
  function _is_previous_accuracy_better (line 95) | def _is_previous_accuracy_better(best_checkpoint_path, accuracy):
  function evaluate_and_average (line 106) | def evaluate_and_average(session, tensor, num_estimates):
  function evaluate_and_average_acc_dacc (line 111) | def evaluate_and_average_acc_dacc(session, acc, dacc, num_estimates):

FILE: MAML_MN_FT/backbone.py
  function init_layer (line 18) | def init_layer(L):
  class NNClassifier (line 28) | class NNClassifier():
    method __init__ (line 29) | def __init__(self, n_way):
    method normalize (line 32) | def normalize(self, x):
    method preprocess (line 37) | def preprocess(self, data):
    method dist (line 46) | def dist(self, x1, x2):
    method kl_divergence (line 49) | def kl_divergence(self, k1, k2):
    method fit (line 55) | def fit(self, support, support_labels, support_weights=None):
    method predict (line 73) | def predict(self, query):
    method predict_alt (line 85) | def predict_alt(self, query, measure="euclidean", norm_scores=False, t...
  class MultiNNBiClassifier (line 114) | class MultiNNBiClassifier():
    method __init__ (line 115) | def __init__(self, n_way, n_classifiers, measure="linear", fusion="lin...
    method fit (line 124) | def fit(self, support_x, support_d, support_labels, support_weights=No...
    method fuse_proba (line 133) | def fuse_proba(self, p1, p2):
    method predict (line 145) | def predict(self, query_x, query_d, weights=None, counterfactual=False):
  class MultiNNClassifier (line 167) | class MultiNNClassifier():
    method __init__ (line 168) | def __init__(self, n_way, n_classifiers, measure="euclidean", temp=1.0):
    method fit (line 178) | def fit(self, support, support_labels, support_weights=None):
    method predict (line 189) | def predict(self, query, weights=None):
  class BidrectionalLSTM (line 209) | class BidrectionalLSTM(nn.Module):
    method __init__ (line 210) | def __init__(self, size: int, layers: int):
    method forward (line 229) | def forward(self, inputs):
  class AttentionLSTM (line 242) | class AttentionLSTM(nn.Module):
    method __init__ (line 243) | def __init__(self, size: int, unrolling_steps: int):
    method forward (line 258) | def forward(self, support, queries):
  class MultiLinearClassifier (line 288) | class MultiLinearClassifier(nn.Module):
    method __init__ (line 289) | def __init__(self, n_clf, feat_dim, n_way, sum_log=True, permute=False...
    method create_clf (line 303) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 309) | def forward(self, X):
  class MultiBiLinearClassifier (line 330) | class MultiBiLinearClassifier(nn.Module):
    method __init__ (line 331) | def __init__(self, n_clf, x_feat_dim, d_feat_dim, n_way, sum_log=True,...
    method fuse_logits (line 343) | def fuse_logits(self, p1, p2):
    method create_clf (line 355) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 361) | def forward(self, X, D, counterfactual=False):
  class ResNetKernelClusterAgent (line 384) | class ResNetKernelClusterAgent():
    method __init__ (line 385) | def __init__(self, pretrain, n_clusters, pca_dim, cluster_method="kmea...
    method fit (line 391) | def fit(self):
  class ResNetParamClusterModel (line 407) | class ResNetParamClusterModel():
    method __init__ (line 408) | def __init__(self, pretrain, n_clusters, cluster_method="kmeans"):
    method cluster (line 413) | def cluster(self, features, n_clusters):
    method get_weight_features (line 418) | def get_weight_features(self, weights):
    method fit (line 424) | def fit(self):
    method conv_forward (line 432) | def conv_forward(self, inputs, labels, n_clusters, original_conv):
    method forward (line 452) | def forward(self, imgs):
    method forward2 (line 491) | def forward2(self, imgs):
    method forward3 (line 517) | def forward3(self, imgs):
  class BasisTransformer (line 524) | class BasisTransformer():
    method __init__ (line 525) | def __init__(self, pretrain, recluster=False, cluster_method="kmeans",...
    method fit (line 532) | def fit(self, n_clusters, feat_dim, pca_dim=50):
    method transform (line 573) | def transform(self, X):
  class KernelTransformer (line 582) | class KernelTransformer():
    method __init__ (line 583) | def __init__(self, feat_dim, kernel):
    method fit (line 587) | def fit(self, features):
    method transform (line 594) | def transform(self, X):
    method kernel_f (line 602) | def kernel_f(self, x1, x2):
  class ChannelwiseClassifier (line 613) | class ChannelwiseClassifier(nn.Module):
    method __init__ (line 614) | def __init__(self, feat_dim, n_way, weight, bias=False):
    method reset_parameters (line 626) | def reset_parameters(self):
    method forward (line 635) | def forward(self, X):
  class UnbiasedClassifier (line 660) | class UnbiasedClassifier(nn.Module):
    method __init__ (line 673) | def __init__(self, n_way, x_feature_dim, z_feature_dim, d_feature_dim=0,
    method create_clf (line 694) | def create_clf(self, feat_dim):
    method create_logit_fusion_fn (line 701) | def create_logit_fusion_fn(self):
    method get_fused_feature (line 710) | def get_fused_feature(self, feature_array):
    method forward (line 714) | def forward(self, X, Z, D=None):
  class XDBiClassifier (line 733) | class XDBiClassifier(nn.Module):
    method __init__ (line 734) | def __init__(self, n_way, x_feature_dim, d_feature_dim, architecture="...
    method create_clf (line 751) | def create_clf(self, feat_dim):
    method create_logit_fusion_fn (line 758) | def create_logit_fusion_fn(self):
    method cat_for_logit_fusion (line 769) | def cat_for_logit_fusion(self, A, B):
    method forward (line 774) | def forward(self, X, D):
  class XDClassifier (line 791) | class XDClassifier(nn.Module):
    method __init__ (line 804) | def __init__(self, n_way, x_feature_dim, d_feature_dim, architecture="...
    method get_feature_dim (line 826) | def get_feature_dim(self):
    method create_clf (line 840) | def create_clf(self, feat_dim):
    method get_fused_feature (line 847) | def get_fused_feature(self, feature_array):
    method forward (line 855) | def forward(self, X, D):
  class ProductGate (line 870) | class ProductGate(nn.Module):
    method __init__ (line 871) | def __init__(self):
    method forward (line 874) | def forward(self, x):
  class HarmonicGate (line 882) | class HarmonicGate(nn.Module):
    method __init__ (line 883) | def __init__(self):
    method forward (line 886) | def forward(self, x):
  class SumGate (line 895) | class SumGate(nn.Module):
    method __init__ (line 896) | def __init__(self):
    method forward (line 899) | def forward(self, x):
  class distLinear (line 906) | class distLinear(nn.Module):
    method __init__ (line 907) | def __init__(self, indim, outdim, class_wise_learnable_norm=True):
    method forward (line 919) | def forward(self, x):
  class Flatten (line 930) | class Flatten(nn.Module):
    method __init__ (line 931) | def __init__(self):
    method forward (line 934) | def forward(self, x):
  class Linear_fw (line 938) | class Linear_fw(nn.Linear): #used in MAML to forward input with fast weight
    method __init__ (line 939) | def __init__(self, in_features, out_features):
    method forward (line 944) | def forward(self, x):
  class Conv2d_fw (line 951) | class Conv2d_fw(nn.Conv2d): #used in MAML to forward input with fast weight
    method __init__ (line 952) | def __init__(self, in_channels, out_channels, kernel_size, stride=1,pa...
    method forward (line 958) | def forward(self, x):
  class BatchNorm2d_fw (line 972) | class BatchNorm2d_fw(nn.BatchNorm2d): #used in MAML to forward input wit...
    method __init__ (line 973) | def __init__(self, num_features):
    method forward (line 978) | def forward(self, x):
  class ConvBlock (line 989) | class ConvBlock(nn.Module):
    method __init__ (line 991) | def __init__(self, indim, outdim, pool = True, padding = 1):
    method forward (line 1014) | def forward(self,x):
  class SimpleBlock (line 1019) | class SimpleBlock(nn.Module):
    method __init__ (line 1021) | def __init__(self, indim, outdim, half_res):
    method forward (line 1060) | def forward(self, x):
  class BottleneckBlock (line 1074) | class BottleneckBlock(nn.Module):
    method __init__ (line 1076) | def __init__(self, indim, outdim, half_res):
    method forward (line 1117) | def forward(self, x):
  class ConvNet (line 1134) | class ConvNet(nn.Module):
    method __init__ (line 1135) | def __init__(self, depth, flatten = True):
    method forward (line 1150) | def forward(self,x):
  class ConvNetNopool (line 1154) | class ConvNetNopool(nn.Module): #Relation net use a 4 layer conv with po...
    method __init__ (line 1155) | def __init__(self, depth):
    method forward (line 1167) | def forward(self,x):
  class ConvNetS (line 1171) | class ConvNetS(nn.Module): #For omniglot, only 1 input channel, output d...
    method __init__ (line 1172) | def __init__(self, depth, flatten = True):
    method forward (line 1187) | def forward(self,x):
  class ConvNetSNopool (line 1192) | class ConvNetSNopool(nn.Module): #Relation net use a 4 layer conv with p...
    method __init__ (line 1193) | def __init__(self, depth):
    method forward (line 1205) | def forward(self,x):
  class ResNet (line 1210) | class ResNet(nn.Module):
    method __init__ (line 1212) | def __init__(self,block,list_of_num_layers, list_of_out_dims, flatten ...
    method forward (line 1254) | def forward(self,x):
  function Conv4 (line 1258) | def Conv4():
  function Conv6 (line 1261) | def Conv6():
  function Conv4NP (line 1264) | def Conv4NP():
  function Conv6NP (line 1267) | def Conv6NP():
  function Conv4S (line 1270) | def Conv4S():
  function Conv4SNP (line 1273) | def Conv4SNP():
  function ResNet10 (line 1276) | def ResNet10( flatten = True):
  function ResNet18 (line 1279) | def ResNet18( flatten = True):
  function ResNet34 (line 1282) | def ResNet34( flatten = True):
  function ResNet50 (line 1285) | def ResNet50( flatten = True):
  function ResNet101 (line 1288) | def ResNet101( flatten = True):

FILE: MAML_MN_FT/data/additional_transforms.py
  class ImageJitter (line 15) | class ImageJitter(object):
    method __init__ (line 16) | def __init__(self, transformdict):
    method __call__ (line 20) | def __call__(self, img):

FILE: MAML_MN_FT/data/datamgr.py
  class TransformLoader (line 12) | class TransformLoader:
    method __init__ (line 13) | def __init__(self, image_size,
    method parse_transform (line 20) | def parse_transform(self, transform_type):
    method get_composed_transform (line 36) | def get_composed_transform(self, aug=False):
  class DataManager (line 47) | class DataManager:
    method get_data_loader (line 49) | def get_data_loader(self, data_file, aug):
  class SimpleDataManager (line 53) | class SimpleDataManager(DataManager):
    method __init__ (line 54) | def __init__(self, image_size, batch_size):
    method get_data_loader (line 59) | def get_data_loader(self, data_file, aug, num_workers=12, tiered_mini=...
  class SetDataManager (line 70) | class SetDataManager(DataManager):
    method __init__ (line 71) | def __init__(self, image_size, n_way, n_support, n_query, n_eposide=100):
    method get_data_loader (line 80) | def get_data_loader(self, data_file, aug, debug=False):  # parameters ...

FILE: MAML_MN_FT/data/dataset.py
  class SimpleDataset (line 14) | class SimpleDataset:
    method __init__ (line 15) | def __init__(self, data_file, transform, target_transform=identity):
    method __getitem__ (line 21) | def __getitem__(self, i):
    method __len__ (line 28) | def __len__(self):
  class SimpleTieredDataset (line 32) | class SimpleTieredDataset:
    method __init__ (line 33) | def __init__(self, setname, transform):
    method __len__ (line 66) | def __len__(self):
    method __getitem__ (line 69) | def __getitem__(self, i):
  class SetDataset (line 75) | class SetDataset:
    method __init__ (line 76) | def __init__(self, data_file, batch_size, transform):
    method __getitem__ (line 98) | def __getitem__(self, i):
    method __len__ (line 101) | def __len__(self):
  class SubDataset (line 105) | class SubDataset:
    method __init__ (line 106) | def __init__(self, sub_meta, cl, transform=transforms.ToTensor(), targ...
    method __getitem__ (line 112) | def __getitem__(self, i):
    method __len__ (line 120) | def __len__(self):
  class EpisodicBatchSampler (line 124) | class EpisodicBatchSampler(object):
    method __init__ (line 125) | def __init__(self, n_classes, n_way, n_episodes):
    method __len__ (line 130) | def __len__(self):
    method __iter__ (line 133) | def __iter__(self):

FILE: MAML_MN_FT/data/feature_loader.py
  class SimpleHDF5Dataset (line 5) | class SimpleHDF5Dataset:
    method __init__ (line 6) | def __init__(self, file_handle = None):
    method __getitem__ (line 18) | def __getitem__(self, i):
    method __len__ (line 21) | def __len__(self):
  function init_loader (line 24) | def init_loader(filename, get_path=False, path_file=None):

FILE: MAML_MN_FT/io_utils.py
  function parse_args (line 20) | def parse_args(script):
  function get_assigned_file (line 53) | def get_assigned_file(checkpoint_dir,num):
  function get_resume_file (line 57) | def get_resume_file(checkpoint_dir):
  function get_best_file (line 68) | def get_best_file(checkpoint_dir):
  function print_accuracy (line 75) | def print_accuracy(acc):
  function print_with_carriage_return (line 79) | def print_with_carriage_return(line):
  function end_carriage_return_print (line 83) | def end_carriage_return_print():
  function append_to_file (line 86) | def append_to_file(file, line):
  function get_result_file (line 91) | def get_result_file(test_name, method_name):
  function calc_recall_precision (line 101) | def calc_recall_precision(y, pred):

FILE: MAML_MN_FT/main.py
  function func_not_found (line 4) | def func_not_found():  # just in case we dont have the function
  function main (line 8) | def main():

FILE: MAML_MN_FT/methods/DMAML.py
  class DMAML (line 14) | class DMAML(MetaTemplate):
    method __init__ (line 15) | def __init__(self, model_func, n_way, n_support, pretrain, n_splits, i...
    method get_feat_dim (line 65) | def get_feat_dim(self):
    method fuse_features (line 75) | def fuse_features(self, x1, x2):
    method normalize (line 83) | def normalize(self, x, dim=1):
    method fuse_proba (line 88) | def fuse_proba(self, p1, p2):
    method set_forward (line 100) | def set_forward(self, x, is_feature=False):
    method set_forward_adaptation (line 148) | def set_forward_adaptation(self, x, is_feature=False):
    method set_forward_loss (line 151) | def set_forward_loss(self, x):
    method train_loop (line 158) | def train_loop(self, epoch, train_loader, optimizer):
    method test_loop (line 183) | def test_loop(self, test_loader, return_std=False, metric="acc"):

FILE: MAML_MN_FT/methods/DMatchingNet.py
  class DMatchingNet (line 12) | class DMatchingNet(MetaTemplate):
    method __init__ (line 13) | def __init__(self, model_func, n_way, n_support, pretrain, n_splits, i...
    method get_feat_dim (line 58) | def get_feat_dim(self):
    method fuse_features (line 68) | def fuse_features(self, x1, x2):
    method normalize (line 76) | def normalize(self, x, dim=1):
    method fuse_proba (line 81) | def fuse_proba(self, p1, p2):
    method set_forward (line 93) | def set_forward(self, x, is_feature=False):
    method predict (line 143) | def predict(self, support, query):
    method set_forward_loss (line 158) | def set_forward_loss(self, x):

FILE: MAML_MN_FT/methods/MethodTester.py
  class MethodTester (line 21) | class MethodTester():
    method __init__ (line 22) | def __init__(self, params):
    method get_backbone (line 29) | def get_backbone(self):
    method baseline_s2m2_initialize (line 44) | def baseline_s2m2_initialize(self, params, provide_original_image):
    method simpleshot_initialize (line 82) | def simpleshot_initialize(self, params, provide_original_image):
    method feat_initialize (line 92) | def feat_initialize(self, params, provide_original_image):
    method sib_initialize (line 102) | def sib_initialize(self, params, provide_original_image):
    method initialize (line 111) | def initialize(self, params, provide_original_image=False):
    method get_task (line 126) | def get_task(self, all_from_same_class=False, provide_original_image=F...
    method get_task_special (line 149) | def get_task_special(self, sampling="sim"):
    method set_experiment_config (line 241) | def set_experiment_config(self, config):
    method set_conditional_config (line 244) | def set_conditional_config(self, config_func):
    method set_experiment_method (line 247) | def set_experiment_method(self, method):
    method add_early_stop_criteria (line 250) | def add_early_stop_criteria(self, iter, acc):
    method _should_reinitialize (line 257) | def _should_reinitialize(self, config):
    method start_experiment (line 260) | def start_experiment(self, method, config, test_name="No name", condit...
    method _get_config_list (line 341) | def _get_config_list(self, config_dict):
    method _increment_config_counter (line 356) | def _increment_config_counter(self, config_dict, config_key_counter):
    method _generate_test_case_name (line 364) | def _generate_test_case_name(self, config):
    method _run_model (line 371) | def _run_model(self, model, show_current_accuracy=True, provide_origin...
    method _feature_evaluation (line 403) | def _feature_evaluation(self, model, provide_original_image, sampling=...
    method cosine_similarity (line 432) | def cosine_similarity(self, a, b):
    method _evaluate_hardness (line 435) | def _evaluate_hardness(self, z_all):
    method normalize (line 468) | def normalize(self, x):
    method _evaluate_hardness_logodd (line 473) | def _evaluate_hardness_logodd(self, z_all):
    method _should_early_stop (line 510) | def _should_early_stop(self, iter, acc):
    method meta_train (line 517) | def meta_train(self, config, method, descriptor_str, debug=True, use_t...
    method meta_test (line 610) | def meta_test(self, config, method, descriptor_str, debug=True, requir...

FILE: MAML_MN_FT/methods/NNEDSplitNew.py
  class NNEDSplitNew (line 9) | class NNEDSplitNew(MetaTemplate):
    method __init__ (line 10) | def __init__(self, model_func, n_way, n_support, n_query, pretrain, n_...
    method set_forward (line 46) | def set_forward(self, x, is_feature=True):
    method calc_pd (line 49) | def calc_pd(self, x, clf_idx):
    method calc_ed (line 61) | def calc_ed(self, x):
    method temp (line 72) | def temp(self, x):
    method normalize (line 81) | def normalize(self, x, dim=1):
    method fuse_proba (line 86) | def fuse_proba(self, p1, p2):
    method fuse_features (line 98) | def fuse_features(self, x1, x2):
    method nn_preprocess (line 106) | def nn_preprocess(self, data, center=None, preprocessing="l2n"):
    method get_split_features (line 117) | def get_split_features(self, x, preprocess=False, center=None, preproc...
    method set_forward_adaptation (line 136) | def set_forward_adaptation(self, x, image_paths=None, is_feature=True):
    method set_forward_loss (line 222) | def set_forward_loss(self, x):

FILE: MAML_MN_FT/methods/PretrainedModel.py
  class PretrainedModel (line 17) | class PretrainedModel():
    method __init__ (line 18) | def __init__(self, params):
    method baseline_s2m2_init (line 40) | def baseline_s2m2_init(self, params):
    method simpleshot_init (line 87) | def simpleshot_init(self, params):
    method feat_init (line 133) | def feat_init(self, params):
    method sib_init (line 152) | def sib_init(self, params):
    method get_features (line 177) | def get_features(self, x):
    method classify (line 191) | def classify(self, x, normalize_prob=True):
    method load_d_specific_classifiers (line 212) | def load_d_specific_classifiers(self, n_clf):
    method train_d_specific_classifiers (line 227) | def train_d_specific_classifiers(self, n_clf):
    method test_d_specific_classifiers (line 268) | def test_d_specific_classifiers(self, n_clf):
    method save_pretrain_dataset (line 295) | def save_pretrain_dataset(self, split):
    method get_pretrain_dataset (line 339) | def get_pretrain_dataset(self, split):
    method normalize (line 350) | def normalize(self, x):
    method _calc_pretrained_class_mean (line 355) | def _calc_pretrained_class_mean(self, normalize=False):
    method get_kmeans_pca_model (line 399) | def get_kmeans_pca_model(self, k=8, n_clusters=10, normalize=False):
    method get_pretrained_class_mean (line 451) | def get_pretrained_class_mean(self, normalize=False):

FILE: MAML_MN_FT/methods/VanillaMAML.py
  class VanillaMAML (line 13) | class VanillaMAML(MetaTemplate):
    method __init__ (line 14) | def __init__(self, model_func, n_way, n_support, approx=False, update_...
    method set_forward (line 25) | def set_forward(self, x, is_feature=False):
    method set_forward_adaptation (line 33) | def set_forward_adaptation(self, x, is_feature=False):
    method set_forward_loss (line 36) | def set_forward_loss(self, x):
    method train_loop (line 42) | def train_loop(self, epoch, train_loader, optimizer):
    method test_loop (line 67) | def test_loop(self, test_loader, return_std=False, metric="acc"):

FILE: MAML_MN_FT/methods/VanillaMatchingNet.py
  class VanillaMatchingNet (line 11) | class VanillaMatchingNet(MetaTemplate):
    method __init__ (line 12) | def __init__(self, model_func, n_way, n_support):
    method normalize (line 19) | def normalize(self, x, dim=1):
    method set_forward (line 24) | def set_forward(self, x, is_feature=False):
    method set_forward_loss (line 39) | def set_forward_loss(self, x):

FILE: MAML_MN_FT/methods/meta_template.py
  class MetaTemplate (line 12) | class MetaTemplate(nn.Module):
    method __init__ (line 13) | def __init__(self, model_func, n_way, n_support, change_way = True, im...
    method set_forward (line 27) | def set_forward(self,x,is_feature):
    method set_forward_loss (line 31) | def set_forward_loss(self, x):
    method forward (line 34) | def forward(self,x):
    method parse_feature (line 38) | def parse_feature(self,x,is_feature):
    method parse_images (line 55) | def parse_images(self, image_paths):
    method correct (line 75) | def correct(self, x, metric="acc"):
    method normalize (line 88) | def normalize(self, x):
    method _evaluate_hardness_logodd (line 93) | def _evaluate_hardness_logodd(self, z_all):
    method calc_correct (line 130) | def calc_correct(self, scores):
    method train_loop (line 137) | def train_loop(self, epoch, train_loader, optimizer ):
    method test_loop (line 158) | def test_loop(self, test_loader, record = None, metric="acc"):
    method set_forward_adaptation (line 182) | def set_forward_adaptation(self, x, is_feature = True): #further adapt...

FILE: MAML_MN_FT/methods/meta_toolkits.py
  class MatchingNetModule (line 9) | class MatchingNetModule(nn.Module):
    method __init__ (line 10) | def __init__(self, feat_dim):
    method forward (line 16) | def forward(self, support, query):
    method cuda (line 25) | def cuda(self):
  class FullyContextualEmbedding (line 32) | class FullyContextualEmbedding(nn.Module):
    method __init__ (line 33) | def __init__(self, feat_dim):
    method forward (line 40) | def forward(self, f, G):
    method cuda (line 54) | def cuda(self):
  class FeatureProcessor (line 61) | class FeatureProcessor():
    method __init__ (line 62) | def __init__(self, pretrain, n_splits, is_cosine_feature=False, d_feat...
    method get_split_features (line 88) | def get_split_features(self, x, preprocess=False, center=None, preproc...
    method nn_preprocess (line 106) | def nn_preprocess(self, data, center=None, preprocessing="l2n"):
    method calc_pd (line 117) | def calc_pd(self, x, clf_idx):
    method normalize (line 122) | def normalize(self, x, dim=1):
    method get_d_feature (line 127) | def get_d_feature(self, x):
    method get_features (line 144) | def get_features(self, support, query):
  class MAMLBlock (line 163) | class MAMLBlock(nn.Module):
    method __init__ (line 164) | def __init__(self, feat_dim, n_way, update_step, approx=True, lr=0.01):
    method forward (line 183) | def forward(self, x):
    method fit (line 188) | def fit(self, support, labels):
    method predict (line 215) | def predict(self, query):
  class LEOBlock (line 219) | class LEOBlock(nn.Module):
    method __init__ (line 220) | def __init__(self, feat_dim, latent_dim, n_way, drop_rate):
    method prepare_for_rel_net (line 243) | def prepare_for_rel_net(self, embeddings):
    method get_sampled_weights (line 250) | def get_sampled_weights(self, weight_stats):
    method finetune_z (line 259) | def finetune_z(self, support, initial_z, initial_w, labels):
    method finetune_w (line 277) | def finetune_w(self, support, initial_w, labels):
    method forward_relation_net (line 293) | def forward_relation_net(self, embeddings):
    method average_codes_per_class (line 301) | def average_codes_per_class(self, relation_net_outputs):
    method possibly_sample (line 307) | def possibly_sample(self, distribution_params, stddev_offset=0.):
    method forward_decoder (line 322) | def forward_decoder(self, latents):
    method calc_kl_penalty (line 332) | def calc_kl_penalty(self, latent_samples, latent_distributions):
    method calc_encoder_penalty (line 339) | def calc_encoder_penalty(self, z_f, z):
    method calc_l2_penalty (line 342) | def calc_l2_penalty(self):
    method calc_orthogonality_penalty (line 351) | def calc_orthogonality_penalty(self):
    method fit (line 361) | def fit(self, support, labels):
    method predict (line 377) | def predict(self, query, w):

FILE: MAML_MN_FT/models/FeatWRN.py
  function conv3x3 (line 10) | def conv3x3(in_planes, out_planes, stride=1):
  function conv_init (line 13) | def conv_init(m):
  class wide_basic (line 22) | class wide_basic(nn.Module):
    method __init__ (line 23) | def __init__(self, in_planes, planes, dropout_rate, stride=1):
    method forward (line 37) | def forward(self, x):
  class Wide_ResNet (line 44) | class Wide_ResNet(nn.Module):
    method __init__ (line 45) | def __init__(self, depth, widen_factor, dropout_rate):
    method _wide_layer (line 62) | def _wide_layer(self, block, planes, num_blocks, dropout_rate, stride):
    method forward (line 72) | def forward(self, x):
  class FEATWRN (line 84) | class FEATWRN(nn.Module):
    method __init__ (line 85) | def __init__(self, num_classes=64):
    method forward (line 92) | def forward(self, x):
    method forward_feature (line 96) | def forward_feature(self, x):

FILE: MAML_MN_FT/models/SimpleShotResNet.py
  function conv3x3 (line 7) | def conv3x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 13) | def conv1x1(in_planes, out_planes, stride=1):
  class BasicBlock (line 18) | class BasicBlock(nn.Module):
    method __init__ (line 21) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 31) | def forward(self, x):
  class Bottleneck (line 50) | class Bottleneck(nn.Module):
    method __init__ (line 53) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 65) | def forward(self, x):
  class ResNet (line 88) | class ResNet(nn.Module):
    method __init__ (line 90) | def __init__(self, block, layers, num_classes=1000, zero_init_residual...
    method _make_layer (line 125) | def _make_layer(self, block, planes, blocks, stride=1):
    method forward (line 141) | def forward(self, x, feature=False):
  function resnet10 (line 165) | def resnet10(**kwargs):
  function resnet18 (line 172) | def resnet18(**kwargs):
  function resnet34 (line 179) | def resnet34(**kwargs):
  function resnet50 (line 186) | def resnet50(**kwargs):
  function resnet101 (line 193) | def resnet101(**kwargs):
  function resnet152 (line 200) | def resnet152(**kwargs):

FILE: MAML_MN_FT/models/SimpleShotWideResNet.py
  function conv3x3 (line 9) | def conv3x3(in_planes, out_planes, stride=1):
  function conv_init (line 13) | def conv_init(m):
  class wide_basic (line 23) | class wide_basic(nn.Module):
    method __init__ (line 24) | def __init__(self, in_planes, planes, dropout_rate, stride=1):
    method forward (line 38) | def forward(self, x):
  class Wide_ResNet (line 46) | class Wide_ResNet(nn.Module):
    method __init__ (line 47) | def __init__(self, depth, widen_factor, dropout_rate, num_classes, rem...
    method _wide_layer (line 75) | def _wide_layer(self, block, planes, num_blocks, dropout_rate, stride):
    method forward (line 85) | def forward(self, x, feature=False):
  function wideres (line 105) | def wideres(num_classes=64, remove_linear=False):

FILE: MAML_MN_FT/save_features.py
  function save_features (line 25) | def save_features(model, data_loader, outfile, params):
  function baseline_s2m2_init (line 57) | def baseline_s2m2_init(params):
  function remove_module_from_param_name (line 173) | def remove_module_from_param_name(params_name_str):
  function simple_shot_init (line 179) | def simple_shot_init(params, split):
  function feat_init (line 218) | def feat_init(params, split):
  function sib_init (line 246) | def sib_init(params, split):
  function cosine_init (line 273) | def cosine_init(params, split):
  function initialize_and_save (line 307) | def initialize_and_save(params, split):

FILE: MAML_MN_FT/tests/MetaTrain.py
  class MetaTrain (line 8) | class MetaTrain(MethodTester):
    method __init__ (line 9) | def __init__(self, params):
    method maml5_resnet (line 16) | def maml5_resnet(self):
    method maml5_wrn (line 37) | def maml5_wrn(self):
    method maml1_resnet (line 58) | def maml1_resnet(self):
    method maml1_wrn (line 79) | def maml1_wrn(self):
    method maml5_resnet_tiered (line 100) | def maml5_resnet_tiered(self):
    method maml5_wrn_tiered (line 121) | def maml5_wrn_tiered(self):
    method maml1_resnet_tiered (line 142) | def maml1_resnet_tiered(self):
    method maml1_wrn_tiered (line 163) | def maml1_wrn_tiered(self):
    method maml5_ifsl_resnet (line 186) | def maml5_ifsl_resnet(self):
    method maml5_ifsl_wrn (line 224) | def maml5_ifsl_wrn(self):
    method maml1_ifsl_resnet (line 262) | def maml1_ifsl_resnet(self):
    method maml1_ifsl_wrn (line 300) | def maml1_ifsl_wrn(self):
    method maml5_ifsl_resnet_tiered (line 338) | def maml5_ifsl_resnet_tiered(self):
    method maml5_ifsl_wrn_tiered (line 376) | def maml5_ifsl_wrn_tiered(self):
    method maml1_ifsl_resnet_tiered (line 414) | def maml1_ifsl_resnet_tiered(self):
    method maml1_ifsl_wrn_tiered (line 452) | def maml1_ifsl_wrn_tiered(self):
    method mn5_resnet (line 492) | def mn5_resnet(self):
    method mn5_wrn (line 507) | def mn5_wrn(self):
    method mn1_resnet (line 522) | def mn1_resnet(self):
    method mn1_wrn (line 537) | def mn1_wrn(self):
    method mn5_resnet_tiered (line 552) | def mn5_resnet_tiered(self):
    method mn5_wrn_tiered (line 567) | def mn5_wrn_tiered(self):
    method mn1_resnet_tiered (line 582) | def mn1_resnet_tiered(self):
    method mn1_wrn_tiered (line 597) | def mn1_wrn_tiered(self):
    method mn5_ifsl_resnet (line 613) | def mn5_ifsl_resnet(self):
    method mn5_ifsl_wrn (line 646) | def mn5_ifsl_wrn(self):
    method mn1_ifsl_resnet (line 679) | def mn1_ifsl_resnet(self):
    method mn1_ifsl_wrn (line 712) | def mn1_ifsl_wrn(self):
    method mn5_ifsl_resnet_tiered (line 745) | def mn5_ifsl_resnet_tiered(self):
    method mn5_ifsl_wrn_tiered (line 779) | def mn5_ifsl_wrn_tiered(self):
    method mn1_ifsl_resnet_tiered (line 813) | def mn1_ifsl_resnet_tiered(self):
    method mn1_ifsl_wrn_tiered (line 847) | def mn1_ifsl_wrn_tiered(self):

FILE: MAML_MN_FT/utils.py
  function one_hot (line 4) | def one_hot(y, num_class):
  function DBindex (line 7) | def DBindex(cl_data_file):
  function sparsity (line 25) | def sparsity(cl_data_file):

FILE: MTL/configs/baseline_config.py
  class Params (line 1) | class Params():
    method __init__ (line 2) | def __init__(self):
  function mini_5_resnet_baseline (line 6) | def mini_5_resnet_baseline():
  function mini_1_resnet_baseline (line 21) | def mini_1_resnet_baseline():
  function tiered_5_resnet_baseline (line 36) | def tiered_5_resnet_baseline():
  function tiered_1_resnet_baseline (line 51) | def tiered_1_resnet_baseline():
  function mini_5_wrn_baseline (line 66) | def mini_5_wrn_baseline():
  function mini_1_wrn_baseline (line 81) | def mini_1_wrn_baseline():
  function tiered_5_wrn_baseline (line 96) | def tiered_5_wrn_baseline():
  function tiered_1_wrn_baseline (line 111) | def tiered_1_wrn_baseline():

FILE: MTL/configs/ifsl_resnet_config.py
  class Params (line 1) | class Params():
    method __init__ (line 2) | def __init__(self):
  function mini_5_resnet_d (line 6) | def mini_5_resnet_d():
  function mini_1_resnet_d (line 44) | def mini_1_resnet_d():
  function tiered_5_resnet_d (line 82) | def tiered_5_resnet_d():
  function tiered_1_resnet_d (line 120) | def tiered_1_resnet_d():

FILE: MTL/configs/ifsl_wrn_config.py
  class Params (line 1) | class Params():
    method __init__ (line 2) | def __init__(self):
  function mini_5_wrn_d (line 6) | def mini_5_wrn_d():
  function mini_1_wrn_d (line 44) | def mini_1_wrn_d():
  function tiered_5_wrn_d (line 82) | def tiered_5_wrn_d():
  function tiered_1_wrn_d (line 120) | def tiered_1_wrn_d():

FILE: MTL/dataloader/dataset_loader.py
  class DatasetLoader (line 20) | class DatasetLoader(Dataset):
    method __init__ (line 22) | def __init__(self, setname, args, dataset="miniImagenet", train_aug=Fa...
    method __len__ (line 99) | def __len__(self):
    method __getitem__ (line 102) | def __getitem__(self, i):

FILE: MTL/dataloader/samplers.py
  class CategoriesSampler (line 15) | class CategoriesSampler():
    method __init__ (line 17) | def __init__(self, label, n_batch, n_cls, n_per):
    method __len__ (line 29) | def __len__(self):
    method __iter__ (line 31) | def __iter__(self):

FILE: MTL/models/IFSL.py
  class PretrainNet (line 14) | class PretrainNet():
    method __init__ (line 15) | def __init__(self, args):
    method load_classifier (line 36) | def load_classifier(self, n_splits, epoch=22, num_classes=64):
    method train_classifier (line 50) | def train_classifier(self, n_splits, num_classes=64):
    method normalize (line 94) | def normalize(self, x):
    method get_base_means (line 99) | def get_base_means(self, num_classes=64, is_cosine_feature=False):
    method save_base_means (line 107) | def save_base_means(self, num_classes=64, is_cosine_feature=False):
  class distLinear (line 137) | class distLinear(nn.Module):
    method __init__ (line 138) | def __init__(self, indim, outdim, class_wise_learnable_norm=True):
    method forward (line 150) | def forward(self, x):
  class MultiLinearClassifier (line 162) | class MultiLinearClassifier(nn.Module):
    method __init__ (line 163) | def __init__(self, n_clf, feat_dim, n_way, sum_log=True, permute=False...
    method create_clf (line 177) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 183) | def forward(self, X):
  class MultiBiLinearClassifier (line 204) | class MultiBiLinearClassifier(nn.Module):
    method __init__ (line 205) | def __init__(self, n_clf, x_feat_dim, d_feat_dim, n_way, sum_log=True,...
    method fuse_logits (line 217) | def fuse_logits(self, p1, p2):
    method create_clf (line 229) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 235) | def forward(self, X, D, counterfactual=False):
  class BaseLearner (line 258) | class BaseLearner(nn.Module):
    method __init__ (line 260) | def __init__(self, args, z_dim):
    method forward (line 271) | def forward(self, input_x, the_vars=None):
    method parameters (line 279) | def parameters(self):
    method initialize (line 282) | def initialize(self):
  class DeconfoundedLearner (line 287) | class DeconfoundedLearner():
    method __init__ (line 288) | def __init__(self, pretrain, classifier="bi", logit_fusion="product", ...
    method calc_pd (line 355) | def calc_pd(self, x, clf_idx):
    method get_pd_features (line 359) | def get_pd_features(self, x):
    method get_ed_features (line 369) | def get_ed_features(self, x):
    method get_split_features (line 379) | def get_split_features(self, x):
    method fuse_feature (line 389) | def fuse_feature(self, a, b, dim=2):
    method fuse_logits (line 397) | def fuse_logits(self, p1, p2):
    method backward_loss_and_step (line 409) | def backward_loss_and_step(self, loss, optimizer):
    method fit (line 421) | def fit(self, support, query, support_labels, support_embedding, query...
    method fit_no_split (line 427) | def fit_no_split(self, support, query, support_labels, support_embeddi...
    method no_split_update (line 451) | def no_split_update(self, x, d, label):
    method calc_no_split_logit (line 471) | def calc_no_split_logit(self, x, d, fast_weight_x=None, fast_weight_d=...
    method fit_multi_splits (line 481) | def fit_multi_splits(self, support, query, support_labels, support_emb...
    method predict (line 549) | def predict(self, support_labels, support_embedding, query_embedding):

FILE: MTL/models/IFSL_modules.py
  class Linear_fw (line 7) | class Linear_fw(nn.Linear): #used in MAML to forward input with fast weight
    method __init__ (line 8) | def __init__(self, in_features, out_features):
    method forward (line 13) | def forward(self, x):
  class IFSLBaseLearner (line 21) | class IFSLBaseLearner(nn.Module):
    method __init__ (line 22) | def __init__(self, feat_dim, n_way, update_step, approx=True, lr=0.01):
    method forward (line 33) | def forward(self, x):
    method fit (line 37) | def fit(self, support, labels):
    method predict (line 64) | def predict(self, query):
  class FeatureProcessor (line 68) | class FeatureProcessor():
    method __init__ (line 69) | def __init__(self, pretrain, n_splits, is_cosine_feature=False, d_feat...
    method get_split_features (line 95) | def get_split_features(self, x, preprocess=False, center=None, preproc...
    method nn_preprocess (line 120) | def nn_preprocess(self, data, center=None, preprocessing="l2n"):
    method calc_pd (line 131) | def calc_pd(self, x, clf_idx):
    method normalize (line 135) | def normalize(self, x, dim=1):
    method get_d_feature (line 140) | def get_d_feature(self, x, x_ori):
    method get_features (line 160) | def get_features(self, support, query, support_ori, query_ori):

FILE: MTL/models/IFSL_pretrain.py
  class Pretrain (line 11) | class Pretrain():
    method __init__ (line 12) | def __init__(self, dataset, method, model, init_model=True):
    method simpleshot_init (line 22) | def simpleshot_init(self):
    method classify (line 69) | def classify(self, x, normalize_prob=True):
    method get_features (line 89) | def get_features(self, x):
    method get_pretrained_class_mean (line 103) | def get_pretrained_class_mean(self, normalize=False):
    method normalize (line 117) | def normalize(self, x):
    method get_base_means (line 122) | def get_base_means(self, normalize=False):

FILE: MTL/models/ResNet10.py
  class ResNet (line 7) | class ResNet(nn.Module):
    method __init__ (line 8) | def __init__(self, block, conv, layers, num_classes=1000, zero_init_re...
    method _make_layer (line 46) | def _make_layer(self, block, planes, blocks, stride=1):
    method forward (line 63) | def forward(self, x, feature=False):
  function ResNet10MTL (line 95) | def ResNet10MTL(**kwargs):
  function ResNet10 (line 100) | def ResNet10(**kwargs):

FILE: MTL/models/WRN28.py
  class wide_basic (line 6) | class wide_basic(nn.Module):
    method __init__ (line 7) | def __init__(self, in_planes, planes, dropout_rate, stride=1):
    method forward (line 21) | def forward(self, x):
  class wide_basic_mtl (line 28) | class wide_basic_mtl(nn.Module):
    method __init__ (line 29) | def __init__(self, in_planes, planes, dropout_rate, stride=1):
    method forward (line 43) | def forward(self, x):
  function conv3x3 (line 50) | def conv3x3(in_planes, out_planes, stride=1):
  function conv3x3mtl (line 54) | def conv3x3mtl(in_planes, out_planes, stride=1):
  class Wide_ResNet (line 58) | class Wide_ResNet(nn.Module):
    method __init__ (line 59) | def __init__(self, block, conv3x3proto, depth, widen_factor, dropout_r...
    method _wide_layer (line 87) | def _wide_layer(self, block, planes, num_blocks, dropout_rate, stride):
    method forward (line 97) | def forward(self, x, feature=False):
  function WideRes28 (line 117) | def WideRes28(num_classes=64, remove_linear=False):
  function WideRes28Mtl (line 123) | def WideRes28Mtl(num_classes=64, remove_linear=False):

FILE: MTL/models/conv2d_mtl.py
  class _ConvNdMtl (line 19) | class _ConvNdMtl(Module):
    method __init__ (line 21) | def __init__(self, in_channels, out_channels, kernel_size, stride,
    method reset_parameters (line 55) | def reset_parameters(self):
    method extra_repr (line 66) | def extra_repr(self):
  class Conv2dMtl (line 81) | class Conv2dMtl(_ConvNdMtl):
    method __init__ (line 83) | def __init__(self, in_channels, out_channels, kernel_size, stride=1,
    method forward (line 93) | def forward(self, inp):

FILE: MTL/models/mtl.py
  class BaseLearner (line 24) | class BaseLearner(nn.Module):
    method __init__ (line 26) | def __init__(self, args, z_dim):
    method forward (line 37) | def forward(self, input_x, the_vars=None):
    method parameters (line 45) | def parameters(self):
    method initialize (line 48) | def initialize(self):
  class MtlLearner (line 53) | class MtlLearner(nn.Module):
    method __init__ (line 55) | def __init__(self, args, mode='meta', num_cls=64):
    method load_pretrain_weight (line 85) | def load_pretrain_weight(self, model_dir):
    method encode (line 101) | def encode(self, x):
    method forward (line 105) | def forward(self, inp):
    method pretrain_forward (line 123) | def pretrain_forward(self, inp):
    method meta_forward (line 132) | def meta_forward(self, data_shot, label_shot, data_query, val=False):
    method predict (line 174) | def predict(self, embedding_shot, label_shot, embedding_query):
    method backward_loss_and_step (line 180) | def backward_loss_and_step(self, loss, optimizer=None):
    method preval_forward (line 187) | def preval_forward(self, data_shot, label_shot, data_query):

FILE: MTL/models/resnet_mtl.py
  function conv3x3 (line 15) | def conv3x3(in_planes, out_planes, stride=1):
  class BasicBlock (line 19) | class BasicBlock(nn.Module):
    method __init__ (line 22) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 32) | def forward(self, x):
  class Bottleneck (line 50) | class Bottleneck(nn.Module):
    method __init__ (line 53) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 66) | def forward(self, x):
  function conv3x3mtl (line 88) | def conv3x3mtl(in_planes, out_planes, stride=1):
  class BasicBlockMtl (line 93) | class BasicBlockMtl(nn.Module):
    method __init__ (line 96) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 106) | def forward(self, x):
  class BottleneckMtl (line 125) | class BottleneckMtl(nn.Module):
    method __init__ (line 128) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 141) | def forward(self, x):
  class ResNetMtl (line 164) | class ResNetMtl(nn.Module):
    method __init__ (line 165) | def __init__(self, layers=[4, 4, 4], mtl=True):
    method _make_layer (line 190) | def _make_layer(self, block, planes, blocks, stride=1):
    method forward (line 206) | def forward(self, x):

FILE: MTL/run_meta.py
  function run_exp (line 13) | def run_exp(num_batch=1000, shot=1, query=15, lr1=0.0001, lr2=0.001, bas...

FILE: MTL/run_pre.py
  function run_exp (line 13) | def run_exp(lr=0.1, gamma=0.2, step_size=30):

FILE: MTL/run_pre_clfs.py
  function run_exp (line 13) | def run_exp(lr=0.1, gamma=0.2, step_size=30, n_clfs=10):

FILE: MTL/run_test.py
  function run_exp (line 13) | def run_exp(num_batch=1000, shot=1, query=15, lr1=0.0001, lr2=0.001, bas...

FILE: MTL/trainer/meta.py
  class MetaTrainer (line 31) | class MetaTrainer(object):
    method __init__ (line 33) | def __init__(self, args):
    method write_output_message (line 123) | def write_output_message(self, message, file_name=None):
    method save_model (line 131) | def save_model(self, name):
    method train (line 138) | def train(self):
    method eval (line 301) | def eval(self):

FILE: MTL/trainer/pre.py
  class PreTrainer (line 23) | class PreTrainer(object):
    method __init__ (line 25) | def __init__(self, args):
    method save_model (line 70) | def save_model(self, name):
    method train (line 77) | def train(self):

FILE: MTL/utils/gpu_tools.py
  function set_gpu (line 14) | def set_gpu(cuda_device):

FILE: MTL/utils/hacc.py
  class Hacc (line 4) | class Hacc():
    method __init__ (line 5) | def __init__(self, splits=10, topk=10):
    method add_data (line 11) | def add_data(self, hardness, correct_prediction):
    method get_splits_hacc (line 15) | def get_splits_hacc(self):
    method get_topk_hacc (line 31) | def get_topk_hacc(self):
    method get_topk_hard_acc (line 42) | def get_topk_hard_acc(self):
    method get_acc_in_range (line 50) | def get_acc_in_range(self, start, end):
    method get_plot (line 60) | def get_plot(self, splits):

FILE: MTL/utils/misc.py
  function ensure_path (line 22) | def ensure_path(path):
  class Averager (line 32) | class Averager():
    method __init__ (line 34) | def __init__(self):
    method add (line 38) | def add(self, x):
    method item (line 42) | def item(self):
  function count_acc (line 45) | def count_acc(logits, label):
  function normalize (line 58) | def normalize(x):
  function count_dacc (line 63) | def count_dacc(pred_logits, support_labels, query_labels, support_imgs, ...
  function get_hardness_correct (line 93) | def get_hardness_correct(pred_logits, support_labels, query_labels, supp...
  class Timer (line 121) | class Timer():
    method __init__ (line 123) | def __init__(self):
    method measure (line 126) | def measure(self, p=1):
  function pprint (line 137) | def pprint(x):
  function format_time (line 141) | def format_time(seconds):
  function compute_confidence_interval (line 174) | def compute_confidence_interval(data):
  function progress_bar (line 197) | def progress_bar(current, total, msg=None):

FILE: SIB/PretrainedModel.py
  class PretrainedModel (line 18) | class PretrainedModel():
    method __init__ (line 19) | def __init__(self, params):
    method simpleshot_init (line 40) | def simpleshot_init(self, params):
    method sib_init (line 78) | def sib_init(self, params):
    method get_features (line 84) | def get_features(self, x):
    method classify (line 100) | def classify(self, x, normalize_prob=True):
    method load_d_specific_classifiers (line 122) | def load_d_specific_classifiers(self, n_clf):
    method load_classifier_weights (line 139) | def load_classifier_weights(self, n_clf, idx):
    method train_d_specific_classifiers (line 148) | def train_d_specific_classifiers(self, n_clf):
    method test_d_specific_classifiers (line 188) | def test_d_specific_classifiers(self, n_clf):
    method save_pretrain_dataset (line 215) | def save_pretrain_dataset(self, split):
    method get_pretrain_dataset (line 258) | def get_pretrain_dataset(self, split):
    method normalize (line 269) | def normalize(self, x):
    method _calc_pretrained_class_mean (line 274) | def _calc_pretrained_class_mean(self, normalize=False):
    method get_kmeans_pca_model (line 305) | def get_kmeans_pca_model(self, k=8, n_clusters=10, normalize=False):
    method get_pretrained_class_mean (line 357) | def get_pretrained_class_mean(self, normalize=False):

FILE: SIB/algorithm.py
  class Params (line 26) | class Params:
    method __init__ (line 27) | def __init__(self):
  class Algorithm (line 30) | class Algorithm:
    method __init__ (line 46) | def __init__(self, args, logger, netFeat, netSIB, optimizer, criterion...
    method load_ckpt (line 84) | def load_ckpt(self, ckptPth):
    method compute_grad_loss (line 104) | def compute_grad_loss(self, clsScore, QueryLabel):
    method cosine_similarity (line 127) | def cosine_similarity(self, a, b):
    method calc_diff_scores (line 130) | def calc_diff_scores(self, pretrain, support, query, support_labels, q...
    method normalize (line 145) | def normalize(self, x):
    method _evaluate_hardness_logodd (line 150) | def _evaluate_hardness_logodd(self, pretrain, support, query, support_...
    method validate (line 181) | def validate(self, valLoader, lr=None, mode='val'):
    method write_output_message (line 255) | def write_output_message(self, message):
    method train (line 260) | def train(self, trainLoader, valLoader, lr=None, coeffGrad=0.0) :

FILE: SIB/backbone.py
  function init_layer (line 18) | def init_layer(L):
  class NNClassifier (line 28) | class NNClassifier():
    method __init__ (line 29) | def __init__(self, n_way):
    method normalize (line 32) | def normalize(self, x):
    method preprocess (line 37) | def preprocess(self, data):
    method dist (line 46) | def dist(self, x1, x2):
    method kl_divergence (line 49) | def kl_divergence(self, k1, k2):
    method fit (line 55) | def fit(self, support, support_labels, support_weights=None):
    method predict (line 73) | def predict(self, query):
    method predict_alt (line 85) | def predict_alt(self, query, measure="euclidean", norm_scores=False, t...
  class MultiNNBiClassifier (line 114) | class MultiNNBiClassifier():
    method __init__ (line 115) | def __init__(self, n_way, n_classifiers, measure="linear", fusion="lin...
    method fit (line 124) | def fit(self, support_x, support_d, support_labels, support_weights=No...
    method fuse_proba (line 133) | def fuse_proba(self, p1, p2):
    method predict (line 145) | def predict(self, query_x, query_d, weights=None, counterfactual=False):
  class MultiNNClassifier (line 167) | class MultiNNClassifier():
    method __init__ (line 168) | def __init__(self, n_way, n_classifiers, measure="euclidean", temp=1.0):
    method fit (line 178) | def fit(self, support, support_labels, support_weights=None):
    method predict (line 189) | def predict(self, query, weights=None):
  class BidrectionalLSTM (line 209) | class BidrectionalLSTM(nn.Module):
    method __init__ (line 210) | def __init__(self, size: int, layers: int):
    method forward (line 229) | def forward(self, inputs):
  class AttentionLSTM (line 242) | class AttentionLSTM(nn.Module):
    method __init__ (line 243) | def __init__(self, size: int, unrolling_steps: int):
    method forward (line 258) | def forward(self, support, queries):
  class MultiLinearClassifier (line 288) | class MultiLinearClassifier(nn.Module):
    method __init__ (line 289) | def __init__(self, n_clf, feat_dim, n_way, sum_log=True, permute=False...
    method create_clf (line 303) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 309) | def forward(self, X):
  class MultiBiLinearClassifier (line 330) | class MultiBiLinearClassifier(nn.Module):
    method __init__ (line 331) | def __init__(self, n_clf, x_feat_dim, d_feat_dim, n_way, sum_log=True,...
    method fuse_logits (line 343) | def fuse_logits(self, p1, p2):
    method create_clf (line 355) | def create_clf(self, loss_type, in_dim, out_dim):
    method forward (line 361) | def forward(self, X, D, counterfactual=False):
  class ResNetKernelClusterAgent (line 384) | class ResNetKernelClusterAgent():
    method __init__ (line 385) | def __init__(self, pretrain, n_clusters, pca_dim, cluster_method="kmea...
    method fit (line 391) | def fit(self):
  class ResNetParamClusterModel (line 407) | class ResNetParamClusterModel():
    method __init__ (line 408) | def __init__(self, pretrain, n_clusters, cluster_method="kmeans"):
    method cluster (line 413) | def cluster(self, features, n_clusters):
    method get_weight_features (line 418) | def get_weight_features(self, weights):
    method fit (line 424) | def fit(self):
    method conv_forward (line 432) | def conv_forward(self, inputs, labels, n_clusters, original_conv):
    method forward (line 452) | def forward(self, imgs):
    method forward2 (line 491) | def forward2(self, imgs):
    method forward3 (line 517) | def forward3(self, imgs):
  class BasisTransformer (line 524) | class BasisTransformer():
    method __init__ (line 525) | def __init__(self, pretrain, recluster=False, cluster_method="kmeans",...
    method fit (line 532) | def fit(self, n_clusters, feat_dim, pca_dim=50):
    method transform (line 573) | def transform(self, X):
  class KernelTransformer (line 582) | class KernelTransformer():
    method __init__ (line 583) | def __init__(self, feat_dim, kernel):
    method fit (line 587) | def fit(self, features):
    method transform (line 594) | def transform(self, X):
    method kernel_f (line 602) | def kernel_f(self, x1, x2):
  class ChannelwiseClassifier (line 613) | class ChannelwiseClassifier(nn.Module):
    method __init__ (line 614) | def __init__(self, feat_dim, n_way, weight, bias=False):
    method reset_parameters (line 626) | def reset_parameters(self):
    method forward (line 635) | def forward(self, X):
  class UnbiasedClassifier (line 660) | class UnbiasedClassifier(nn.Module):
    method __init__ (line 673) | def __init__(self, n_way, x_feature_dim, z_feature_dim, d_feature_dim=0,
    method create_clf (line 694) | def create_clf(self, feat_dim):
    method create_logit_fusion_fn (line 701) | def create_logit_fusion_fn(self):
    method get_fused_feature (line 710) | def get_fused_feature(self, feature_array):
    method forward (line 714) | def forward(self, X, Z, D=None):
  class XDBiClassifier (line 733) | class XDBiClassifier(nn.Module):
    method __init__ (line 734) | def __init__(self, n_way, x_feature_dim, d_feature_dim, architecture="...
    method create_clf (line 751) | def create_clf(self, feat_dim):
    method create_logit_fusion_fn (line 758) | def create_logit_fusion_fn(self):
    method cat_for_logit_fusion (line 769) | def cat_for_logit_fusion(self, A, B):
    method forward (line 774) | def forward(self, X, D):
  class XDClassifier (line 791) | class XDClassifier(nn.Module):
    method __init__ (line 804) | def __init__(self, n_way, x_feature_dim, d_feature_dim, architecture="...
    method get_feature_dim (line 826) | def get_feature_dim(self):
    method create_clf (line 840) | def create_clf(self, feat_dim):
    method get_fused_feature (line 847) | def get_fused_feature(self, feature_array):
    method forward (line 855) | def forward(self, X, D):
  class ProductGate (line 870) | class ProductGate(nn.Module):
    method __init__ (line 871) | def __init__(self):
    method forward (line 874) | def forward(self, x):
  class HarmonicGate (line 882) | class HarmonicGate(nn.Module):
    method __init__ (line 883) | def __init__(self):
    method forward (line 886) | def forward(self, x):
  class SumGate (line 895) | class SumGate(nn.Module):
    method __init__ (line 896) | def __init__(self):
    method forward (line 899) | def forward(self, x):
  class distLinear (line 906) | class distLinear(nn.Module):
    method __init__ (line 907) | def __init__(self, indim, outdim, class_wise_learnable_norm=True):
    method forward (line 919) | def forward(self, x):
  class Flatten (line 930) | class Flatten(nn.Module):
    method __init__ (line 931) | def __init__(self):
    method forward (line 934) | def forward(self, x):
  class Linear_fw (line 938) | class Linear_fw(nn.Linear): #used in MAML to forward input with fast weight
    method __init__ (line 939) | def __init__(self, in_features, out_features):
    method forward (line 944) | def forward(self, x):
  class Conv2d_fw (line 951) | class Conv2d_fw(nn.Conv2d): #used in MAML to forward input with fast weight
    method __init__ (line 952) | def __init__(self, in_channels, out_channels, kernel_size, stride=1,pa...
    method forward (line 958) | def forward(self, x):
  class BatchNorm2d_fw (line 972) | class BatchNorm2d_fw(nn.BatchNorm2d): #used in MAML to forward input wit...
    method __init__ (line 973) | def __init__(self, num_features):
    method forward (line 978) | def forward(self, x):
  class ConvBlock (line 989) | class ConvBlock(nn.Module):
    method __init__ (line 991) | def __init__(self, indim, outdim, pool = True, padding = 1):
    method forward (line 1014) | def forward(self,x):
  class SimpleBlock (line 1019) | class SimpleBlock(nn.Module):
    method __init__ (line 1021) | def __init__(self, indim, outdim, half_res):
    method forward (line 1060) | def forward(self, x):
  class BottleneckBlock (line 1074) | class BottleneckBlock(nn.Module):
    method __init__ (line 1076) | def __init__(self, indim, outdim, half_res):
    method forward (line 1117) | def forward(self, x):
  class ConvNet (line 1134) | class ConvNet(nn.Module):
    method __init__ (line 1135) | def __init__(self, depth, flatten = True):
    method forward (line 1150) | def forward(self,x):
  class ConvNetNopool (line 1154) | class ConvNetNopool(nn.Module): #Relation net use a 4 layer conv with po...
    method __init__ (line 1155) | def __init__(self, depth):
    method forward (line 1167) | def forward(self,x):
  class ConvNetS (line 1171) | class ConvNetS(nn.Module): #For omniglot, only 1 input channel, output d...
    method __init__ (line 1172) | def __init__(self, depth, flatten = True):
    method forward (line 1187) | def forward(self,x):
  class ConvNetSNopool (line 1192) | class ConvNetSNopool(nn.Module): #Relation net use a 4 layer conv with p...
    method __init__ (line 1193) | def __init__(self, depth):
    method forward (line 1205) | def forward(self,x):
  class ResNet (line 1210) | class ResNet(nn.Module):
    method __init__ (line 1212) | def __init__(self,block,list_of_num_layers, list_of_out_dims, flatten ...
    method forward (line 1254) | def forward(self,x):
  function Conv4 (line 1258) | def Conv4():
  function Conv6 (line 1261) | def Conv6():
  function Conv4NP (line 1264) | def Conv4NP():
  function Conv6NP (line 1267) | def Conv6NP():
  function Conv4S (line 1270) | def Conv4S():
  function Conv4SNP (line 1273) | def Conv4SNP():
  function ResNet10 (line 1276) | def ResNet10( flatten = True):
  function ResNet18 (line 1279) | def ResNet18( flatten = True):
  function ResNet34 (line 1282) | def ResNet34( flatten = True):
  function ResNet50 (line 1285) | def ResNet50( flatten = True):
  function ResNet101 (line 1288) | def ResNet101( flatten = True):

FILE: SIB/data/additional_transforms.py
  class ImageJitter (line 15) | class ImageJitter(object):
    method __init__ (line 16) | def __init__(self, transformdict):
    method __call__ (line 20) | def __call__(self, img):

FILE: SIB/data/datamgr.py
  class TransformLoader (line 12) | class TransformLoader:
    method __init__ (line 13) | def __init__(self, image_size,
    method parse_transform (line 20) | def parse_transform(self, transform_type):
    method get_composed_transform (line 36) | def get_composed_transform(self, aug=False):
  class DataManager (line 47) | class DataManager:
    method get_data_loader (line 49) | def get_data_loader(self, data_file, aug):
  class SimpleDataManager (line 53) | class SimpleDataManager(DataManager):
    method __init__ (line 54) | def __init__(self, image_size, batch_size):
    method get_data_loader (line 59) | def get_data_loader(self, data_file, aug, num_workers=12, tiered_mini=...
  class SetDataManager (line 70) | class SetDataManager(DataManager):
    method __init__ (line 71) | def __init__(self, image_size, n_way, n_support, n_query, n_eposide=100):
    method get_data_loader (line 80) | def get_data_loader(self, data_file, aug, debug=False):  # parameters ...

FILE: SIB/data/dataset.py
  class SimpleDataset (line 14) | class SimpleDataset:
    method __init__ (line 15) | def __init__(self, data_file, transform, target_transform=identity):
    method __getitem__ (line 21) | def __getitem__(self, i):
    method __len__ (line 28) | def __len__(self):
  class SimpleTieredDataset (line 32) | class SimpleTieredDataset:
    method __init__ (line 33) | def __init__(self, setname, transform):
    method __len__ (line 66) | def __len__(self):
    method __getitem__ (line 69) | def __getitem__(self, i):
  class SetDataset (line 75) | class SetDataset:
    method __init__ (line 76) | def __init__(self, data_file, batch_size, transform):
    method __getitem__ (line 98) | def __getitem__(self, i):
    method __len__ (line 101) | def __len__(self):
  class SubDataset (line 105) | class SubDataset:
    method __init__ (line 106) | def __init__(self, sub_meta, cl, transform=transforms.ToTensor(), targ...
    method __getitem__ (line 112) | def __getitem__(self, i):
    method __len__ (line 120) | def __len__(self):
  class EpisodicBatchSampler (line 124) | class EpisodicBatchSampler(object):
    method __init__ (line 125) | def __init__(self, n_classes, n_way, n_episodes):
    method __len__ (line 130) | def __len__(self):
    method __iter__ (line 133) | def __iter__(self):

FILE: SIB/data/feature_loader.py
  class SimpleHDF5Dataset (line 5) | class SimpleHDF5Dataset:
    method __init__ (line 6) | def __init__(self, file_handle = None):
    method __getitem__ (line 18) | def __getitem__(self, i):
    method __len__ (line 21) | def __len__(self):
  function init_loader (line 24) | def init_loader(filename, get_path=False, path_file=None):

FILE: SIB/data/get_cifarfs.py
  function download_file (line 24) | def download_file(url, filename):

FILE: SIB/dataloader.py
  function PilLoaderRGB (line 26) | def PilLoaderRGB(imgPath) :
  class EpisodeSampler (line 30) | class EpisodeSampler():
    method __init__ (line 44) | def __init__(self, imgDir, nClsEpisode, nSupport, nQuery, transform, u...
    method getEpisode (line 61) | def getEpisode(self):
  class BatchSampler (line 107) | class BatchSampler():
    method __init__ (line 122) | def __init__(self, imgDir, nClsEpisode, nSupport, nQuery, transform, u...
    method getBatch (line 136) | def getBatch(self):
  class ValImageFolder (line 159) | class ValImageFolder(data.Dataset):
    method __init__ (line 170) | def __init__(self, episodeJson, imgDir, inputW, inputH, valTransform, ...
    method __getitem__ (line 194) | def __getitem__(self, index):
    method __len__ (line 221) | def __len__(self):
  function ValLoader (line 228) | def ValLoader(episodeJson, imgDir, inputW, inputH, valTransform, useGPU) :
  function TrainLoader (line 235) | def TrainLoader(batchSize, imgDir, trainTransform) :

FILE: SIB/dataset.py
  function dataset_setting (line 19) | def dataset_setting(dataset, nSupport, image_size=80):

FILE: SIB/deconfound/DSIB.py
  class DeconfoundedSIB (line 13) | class DeconfoundedSIB(nn.Module):
    method __init__ (line 14) | def __init__(self, n_way, pretrain, n_splits, is_cosine_feature, d_fea...
    method get_feat_dim (line 62) | def get_feat_dim(self):
    method fuse_features (line 72) | def fuse_features(self, x1, x2):
    method normalize (line 80) | def normalize(self, x, dim=1):
    method fuse_proba (line 85) | def fuse_proba(self, p1, p2):
    method forward (line 97) | def forward(self, support, labels, query, _):

FILE: SIB/deconfound/meta_toolkits.py
  class FeatureProcessor (line 9) | class FeatureProcessor():
    method __init__ (line 10) | def __init__(self, pretrain, n_splits, is_cosine_feature=False, d_feat...
    method get_split_features (line 36) | def get_split_features(self, x, preprocess=False, center=None, preproc...
    method nn_preprocess (line 54) | def nn_preprocess(self, data, center=None, preprocessing="l2n"):
    method calc_pd (line 65) | def calc_pd(self, x, clf_idx):
    method normalize (line 70) | def normalize(self, x, dim=1):
    method get_d_feature (line 75) | def get_d_feature(self, x):
    method get_features (line 93) | def get_features(self, support, query):

FILE: SIB/io_utils.py
  function parse_args (line 20) | def parse_args(script):
  function get_assigned_file (line 53) | def get_assigned_file(checkpoint_dir,num):
  function get_resume_file (line 57) | def get_resume_file(checkpoint_dir):
  function get_best_file (line 68) | def get_best_file(checkpoint_dir):
  function print_accuracy (line 75) | def print_accuracy(acc):
  function print_with_carriage_return (line 79) | def print_with_carriage_return(line):
  function end_carriage_return_print (line 83) | def end_carriage_return_print():
  function append_to_file (line 86) | def append_to_file(file, line):
  function get_result_file (line 91) | def get_result_file(test_name, method_name):
  function calc_recall_precision (line 101) | def calc_recall_precision(y, pred):

FILE: SIB/main.py
  class Params (line 35) | class Params:
    method __init__ (line 36) | def __init__(self):

FILE: SIB/main_feat.py
  class ClassifierEval (line 42) | class ClassifierEval(nn.Module):
    method __init__ (line 47) | def __init__(self, nKnovel, nFeat):
    method apply_classification_weights (line 57) | def apply_classification_weights(self, features, cls_weights):
    method forward (line 67) | def forward(self, features_supp, features_query):
  class ClassifierTrain (line 80) | class ClassifierTrain(nn.Module):
    method __init__ (line 81) | def __init__(self, nCls, nFeat=640, scaleCls = 10.):
    method getWeight (line 101) | def getWeight(self):
    method applyWeightCosine (line 104) | def applyWeightCosine(self, feature, weight, bias, scaleCls):
    method forward (line 113) | def forward(self, feature):
  class BaseTrainer (line 119) | class BaseTrainer:
    method __init__ (line 120) | def __init__(self, trainLoader, valLoader, nbCls, nClsEpisode, nFeat,
    method LrWarmUp (line 145) | def LrWarmUp(self, totalIter, lr):
    method train (line 209) | def train(self, epoch):
    method test (line 242) | def test(self, epoch):

FILE: SIB/networks.py
  class ConvBlock (line 21) | class ConvBlock(nn.Module):
    method __init__ (line 22) | def __init__(self, in_planes, out_planes):
    method forward (line 34) | def forward(self, x):
  class ConvNet_4_64 (line 38) | class ConvNet_4_64(nn.Module):
    method __init__ (line 39) | def __init__(self, inputW=80, inputH=80):
    method forward (line 58) | def forward(self, x):
  class BasicBlock (line 64) | class BasicBlock(nn.Module):
    method __init__ (line 65) | def __init__(self, in_planes, out_planes, stride, dropRate=0.0):
    method forward (line 82) | def forward(self, x):
  class NetworkBlock (line 101) | class NetworkBlock(nn.Module):
    method __init__ (line 102) | def __init__(self, nb_layers, in_planes, out_planes, block, stride, dr...
    method _make_layer (line 106) | def _make_layer(self, block, in_planes, out_planes, nb_layers, stride,...
    method forward (line 114) | def forward(self, x):
  class WideResNet (line 118) | class WideResNet(nn.Module):
    method __init__ (line 119) | def __init__(self, depth=28, widen_factor=10, dropRate=0.0, userelu=Tr...
    method forward (line 150) | def forward(self, x):
  function label_to_1hot (line 166) | def label_to_1hot(label, K):
  class dni_linear (line 174) | class dni_linear(nn.Module):
    method __init__ (line 175) | def __init__(self, input_dims, dni_hidden_size=1024):
    method forward (line 192) | def forward(self, x):
  class LinearDiag (line 199) | class LinearDiag(nn.Module):
    method __init__ (line 200) | def __init__(self, num_features, bias=False):
    method forward (line 211) | def forward(self, X):
  class FeatExemplarAvgBlock (line 219) | class FeatExemplarAvgBlock(nn.Module):
    method __init__ (line 220) | def __init__(self, nFeat):
    method forward (line 223) | def forward(self, features_train, labels_train):
  function get_featnet (line 232) | def get_featnet(architecture, inputW=80, inputH=80):

FILE: SIB/sib.py
  class ClassifierSIB (line 21) | class ClassifierSIB(nn.Module):
    method __init__ (line 36) | def __init__(self, nKnovel, nFeat, q_steps):
    method apply_classification_weights (line 54) | def apply_classification_weights(self, features, cls_weights):
    method init_theta (line 73) | def init_theta(self, features_supp, labels_supp_1hot):
    method refine_theta (line 89) | def refine_theta(self, theta, features_query, lr=1e-3):
    method get_classification_weights (line 118) | def get_classification_weights(self, features_supp, labels_supp_1hot, ...
    method forward (line 142) | def forward(self, features_supp, labels_supp, features_query, lr):

FILE: SIB/simple_shot_models/Conv4.py
  function conv_block (line 6) | def conv_block(in_channels: int, out_channels: int) -> nn.Module:
  class Conv4 (line 15) | class Conv4(nn.Module):
    method __init__ (line 16) | def __init__(self, num_classes, remove_linear=False):
    method forward (line 27) | def forward(self, x, feature=False):

FILE: SIB/simple_shot_models/DenseNet.py
  class _DenseLayer (line 10) | class _DenseLayer(nn.Sequential):
    method __init__ (line 11) | def __init__(self, num_input_features, growth_rate, bn_size, drop_rate):
    method forward (line 23) | def forward(self, x):
  class _DenseBlock (line 30) | class _DenseBlock(nn.Sequential):
    method __init__ (line 31) | def __init__(self, num_layers, num_input_features, bn_size, growth_rat...
  class _Transition (line 38) | class _Transition(nn.Sequential):
    method __init__ (line 39) | def __init__(self, num_input_features, num_output_features):
  class DenseNet (line 48) | class DenseNet(nn.Module):
    method __init__ (line 62) | def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16),
    method forward (line 103) | def forward(self, x, feature=False):
  function densenet121 (line 120) | def densenet121(**kwargs):
  function densenet169 (line 129) | def densenet169(**kwargs):
  function densenet201 (line 138) | def densenet201(**kwargs):
  function densenet161 (line 147) | def densenet161(**kwargs):

FILE: SIB/simple_shot_models/MobileNet.py
  class Block (line 12) | class Block(nn.Module):
    method __init__ (line 15) | def __init__(self, in_planes, out_planes, stride=1):
    method forward (line 23) | def forward(self, x):
  class MobileNet (line 29) | class MobileNet(nn.Module):
    method __init__ (line 33) | def __init__(self, num_classes=10, remove_linear=False):
    method _make_layers (line 44) | def _make_layers(self, in_planes):
    method forward (line 53) | def forward(self, x, feature=False):

FILE: SIB/simple_shot_models/ProtoNet.py
  function get_metric (line 6) | def get_metric(metric_type):
  class ProtoNet (line 16) | class ProtoNet(nn.Module):
    method __init__ (line 18) | def __init__(self, feature_net, args=None):
    method forward (line 28) | def forward(self, data, _=False):

FILE: SIB/simple_shot_models/ResNet.py
  function conv3x3 (line 7) | def conv3x3(in_planes, out_planes, stride=1):
  function conv1x1 (line 13) | def conv1x1(in_planes, out_planes, stride=1):
  class BasicBlock (line 18) | class BasicBlock(nn.Module):
    method __init__ (line 21) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 31) | def forward(self, x):
  class Bottleneck (line 50) | class Bottleneck(nn.Module):
    method __init__ (line 53) | def __init__(self, inplanes, planes, stride=1, downsample=None):
    method forward (line 65) | def forward(self, x):
  class ResNet (line 88) | class ResNet(nn.Module):
    method __init__ (line 90) | def __init__(self, block, layers, num_classes=1000, zero_init_residual...
    method _make_layer (line 125) | def _make_layer(self, block, planes, blocks, stride=1):
    method forward (line 141) | def forward(self, x, feature=False):
  function resnet10 (line 165) | def resnet10(**kwargs):
  function resnet18 (line 172) | def resnet18(**kwargs):
  function resnet34 (line 179) | def resnet34(**kwargs):
  function resnet50 (line 186) | def resnet50(**kwargs):
  function resnet101 (line 193) | def resnet101(**kwargs):
  function resnet152 (line 200) | def resnet152(**kwargs):

FILE: SIB/simple_shot_models/WideResNet.py
  function conv3x3 (line 9) | def conv3x3(in_planes, out_planes, stride=1):
  function conv_init (line 13) | def conv_init(m):
  class wide_basic (line 23) | class wide_basic(nn.Module):
    method __init__ (line 24) | def __init__(self, in_planes, planes, dropout_rate, stride=1):
    method forward (line 38) | def forward(self, x):
  class Wide_ResNet (line 46) | class Wide_ResNet(nn.Module):
    method __init__ (line 47) | def __init__(self, depth, widen_factor, dropout_rate, num_classes, rem...
    method _wide_layer (line 75) | def _wide_layer(self, block, planes, num_blocks, dropout_rate, stride):
    method forward (line 85) | def forward(self, x, feature=False):
  function wideres (line 105) | def wideres(num_classes=64, remove_linear=False):

FILE: SIB/utils/config.py
  function create_dirs (line 21) | def create_dirs(dirs):
  function get_config_from_json (line 38) | def get_config_from_json(json_file):
  function get_config_from_yaml (line 54) | def get_config_from_yaml(yaml_file):
  function get_args (line 69) | def get_args():
  function get_config (line 104) | def get_config():

FILE: SIB/utils/outils.py
  class AverageMeter (line 30) | class AverageMeter(object):
    method __init__ (line 32) | def __init__(self):
    method reset (line 35) | def reset(self):
    method update (line 41) | def update(self, val, n=1):
  function getCi (line 48) | def getCi(accLog):
  function accuracy (line 57) | def accuracy(output, target, topk=(1,), diff_scores=None):
  function get_mean_and_std (line 79) | def get_mean_and_std(dataset):
  function init_params (line 94) | def init_params(net):
  function progress_bar (line 118) | def progress_bar(current, total, msg=None):
  function format_time (line 162) | def format_time(seconds):

FILE: SIB/utils/utils.py
  function set_random_seed (line 26) | def set_random_seed(seed=3):
  function to_device (line 33) | def to_device(input, device):
  function fast_hist (line 46) | def fast_hist(label_pred, label_true, n_class):
  function convert_state_dict (line 53) | def convert_state_dict(state_dict):
  function get_logger (line 67) | def get_logger(logdir, name):