Functions — github.com/TencentYoutuResearch/SceneSegmentation-SCRL

hub / github.com/TencentYoutuResearch/SceneSegmentation-SCRL / functions

Functions115 in github.com/TencentYoutuResearch/SceneSegmentation-SCRL

⨍ Functions 115◇ Types & classes 17

↓ 12 callersFunctionto_log

(cfg, content, echo=False, gpu_print_id=0)

utils.py:53

↓ 9 callersFunction_resnet

( arch: str, block: Type[Union[BasicBlock, Bottleneck]], layers: List[int], pretrained: bool,

models/backbones/visual/resnet.py:284

↓ 7 callersFunctionto_log

(args, content, echo=False)

SceneSeg/main.py:182

↓ 5 callersMethodupdate

(self, val, n=1)

utils.py:18

↓ 4 callersMethod_make_layer

(self, block: Type[Union[BasicBlock, Bottleneck]], planes: int, blocks: int, stride: int =

models/backbones/visual/resnet.py:205

↓ 4 callersFunctionencoder_resnet50

(input_channel:int = 9, weight_path: str = '', progress: bool = True, num_classes=2048, **kwargs: Any)

models/backbones/visual/resnet.py:432

↓ 3 callersMethod__init__

( self, block: Type[Union[BasicBlock, Bottleneck]], layers: List[int], input_c

models/backbones/visual/resnet.py:145

↓ 3 callersFunction_process

(data)

data/data_preparation.py:88

↓ 3 callersFunctionconcat_all_gather

Performs all_gather operation on the provided tensors. *** Warning ***: torch.distributed.all_gather has no gradient.

models/core/SCRL_MoCo.py:261

↓ 3 callersFunctionconv1x1

1x1 convolution

models/backbones/visual/resnet.py:31

↓ 3 callersFunctionconv3x3

3x3 convolution with padding

models/backbones/visual/resnet.py:25

↓ 3 callersFunctionprocess_raw_label

(_T = 'train', raw_root_dir = './')

data/data_preparation.py:53

↓ 2 callersMethod_batch_shuffle_ddp

Batch shuffle, for making use of BatchNorm. *** Only support DistributedDataParallel (DDP) model. ***

models/core/SCRL_MoCo.py:82

↓ 2 callersMethod_batch_unshuffle_ddp

Undo batch shuffle. *** Only support DistributedDataParallel (DDP) model. ***

models/core/SCRL_MoCo.py:111

↓ 2 callersMethod_dequeue_and_enqueue

(self, keys)

models/core/SCRL_MoCo.py:65

↓ 2 callersMethod_get_clip_by_idx

(self, idx, length)

SceneSeg/movienet_seg_data.py:92

↓ 2 callersMethod_momentum_update_key_encoder

Momentum update of the key encoder

models/core/SCRL_MoCo.py:57

↓ 2 callersMethod_shuffle_offset

(self)

SceneSeg/movienet_seg_data.py:37

↓ 2 callersFunctionget_train_loader

(cfg)

data/movienet_data.py:97

↓ 2 callersFunctionto_log

(cfg, content, echo=True)

extract_embeddings.py:160

↓ 1 callersMethod__init__

(self, input_feature_dim=2048, fc_dim=1024, hidden_size=512, input_drop_rate=0.3, lstm_drop_rate=0.6,

SceneSeg/BiLSTM_protocol.py:7

↓ 1 callersMethod_forward_impl

(self, x: Tensor, is_fc: bool)

models/backbones/visual/resnet.py:230

↓ 1 callersFunction_generate_shot_num

(new_shot_info='./MovieNet_shot_num.json')

data/data_preparation.py:34

↓ 1 callersMethod_get_batch_fmtstr

(self, num_batches)

utils.py:41

↓ 1 callersMethod_get_randomly_cat_clip

(self, idx)

SceneSeg/movienet_seg_data.py:43

↓ 1 callersMethod_load_from_weight

(self, weight_path: str)

models/backbones/visual/resnet.py:267

↓ 1 callersMethod_padding

(self, data)

SceneSeg/movienet_seg_data.py:138

↓ 1 callersMethod_process

(self, idx)

extract_embeddings.py:34

↓ 1 callersMethod_process_puzzle

(self, idx)

data/movienet_data.py:72

↓ 1 callersMethod_seg_shuffle

(self, data, label)

SceneSeg/movienet_seg_data.py:65

↓ 1 callersMethod_transform

(self, img_list)

data/movienet_data.py:61

↓ 1 callersFunctionaccuracy

Computes the accuracy over the k top predictions for the specified values of k

utils.py:62

↓ 1 callersFunctionadjust_learning_rate

Decay the learning rate based on schedule

pretrain_main.py:95

↓ 1 callersFunctionadjust_learning_rate

Decay the learning rate based on schedule

SceneSeg/main.py:188

↓ 1 callersFunctionconcate_pic

(shot_info, img_path, save_path, row=16)

data/data_preparation.py:9

↓ 1 callersMethoddisplay

(self, batch)

utils.py:35

↓ 1 callersFunctionextract_features

(cfg)

extract_embeddings.py:133

↓ 1 callersFunctionget_ap

(gts_raw,preds_raw,is_list=True)

SceneSeg/main.py:196

↓ 1 callersFunctionget_config

()

extract_embeddings.py:166

↓ 1 callersFunctionget_config

()

pretrain_main.py:121

↓ 1 callersFunctionget_config

()

SceneSeg/main.py:217

↓ 1 callersFunctionget_criterion

(cfg)

models/factory.py:46

↓ 1 callersFunctionget_encoder

(model_name='resnet50', weight_path='', input_channel=9)

extract_embeddings.py:83

↓ 1 callersFunctionget_loader

(cfg, _Type='train')

extract_embeddings.py:57

↓ 1 callersFunctionget_model

(cfg)

models/factory.py:7

↓ 1 callersFunctionget_optimizer

(cfg, model)

models/factory.py:57

↓ 1 callersFunctionget_save_embeddings

(model, loader, shot_num, filename, log_interval=100)

extract_embeddings.py:106

↓ 1 callersMethodinitialize

(self, X)

cluster/Group.py:69

↓ 1 callersFunctionmain

()

pretrain_main.py:131

↓ 1 callersFunctionmain

(args)

SceneSeg/main.py:17

↓ 1 callersFunctionprocess_scene_seg_lable

(scene_seg_path = './CVPR20SceneSeg/data/scene318/label318', scene_seg_label_json_name = './movie1K.scene_

data/data_preparation.py:85

↓ 1 callersMethodreset

(self)

utils.py:12

↓ 1 callersFunctionsave_checkpoint

(cfg, state, is_best, filename='checkpoint.pth.tar')

pretrain_main.py:111

↓ 1 callersFunctionsave_checkpoint

(state, is_best, fpath='checkpoint.pth.tar')

SceneSeg/main.py:210

↓ 1 callersFunctionset_log

(cfg)

utils.py:46

↓ 1 callersFunctionset_log

(args)

SceneSeg/main.py:170

↓ 1 callersFunctionsetup_seed

(seed)

SceneSeg/main.py:162

↓ 1 callersFunctionsetup_worker

(seed, gpu)

pretrain_main.py:43

↓ 1 callersFunctionstart_training

data/movienet_data.py:17

Method__call__

(self, x)

data/movienet_data.py:92

(self, tensor_input, debug=False)

cluster/Group.py:75

Method__getitem__

(self, idx)

extract_embeddings.py:54

Method__getitem__

(self, idx)

SceneSeg/movienet_seg_data.py:106

Method__getitem__

(self, idx)

SceneSeg/movienet_seg_data.py:151

Method__getitem__

(self, idx)

data/movienet_data.py:83

Method__init__

(self, img_path, shot_info_path, transform, frame_per_shot = 3, _Type='train')

extract_embeddings.py:13

Method__init__

(self, name, fmt=':f')

utils.py:7

Method__init__

(self, num_batches, meters, prefix="")

utils.py:30

Method__init__

(self, pkl_path, frame_size=3, shot_num=1, sampled_shot_num=10, shuffle_p=0.5,random_cat=False)

SceneSeg/movienet_seg_data.py:8

Method__init__

(self, pkl_path, frame_size=3, shot_num=1, sampled_shot_num=100)

SceneSeg/movienet_seg_data.py:119

Method__init__

(self, p=0.2)

SceneSeg/BiLSTM_protocol.py:71

Method__init__

(self, base_transform_a, base_transform_b, fixed_aug_shot=True)

data/movienet_data.py:11

Method__init__

(self, img_path, shot_info_path, transform, shot_len = 16, frame_per_shot = 3, _Type='train')

data/movienet_data.py:38

Method__init__

(self, sigma=[.1, 2.])

data/movienet_data.py:89

Method__init__

( self, inplanes: int, planes: int, stride: int = 1, downsample: Optio

models/backbones/visual/resnet.py:39

Method__init__

( self, inplanes: int, planes: int, stride: int = 1, downsample: Optio

models/backbones/visual/resnet.py:94

Method__init__

(self, input_channel:int, block: Type[Union[BasicBlock, Bottleneck]], layers: List[int], weigh

models/backbones/visual/resnet.py:253

Method__init__

(self, base_encoder, dim=2048, K=65536, m=0.999, T=0.07, mlp=False, encoder_pretrained_path: s

models/core/SCRL_MoCo.py:11

Method__init__

(self, num_clusters, shift_threshold=1e-2, max_iter=20, device=torch.devic

cluster/Group.py:12

Method__init__

(self, num_clusters, shift_threshold, max_iter,

cluster/Group.py:55

Method__len__

(self)

extract_embeddings.py:31

Method__len__

(self)

SceneSeg/movienet_seg_data.py:115

Method__len__

(self)

SceneSeg/movienet_seg_data.py:161

Method__len__

(self)

data/movienet_data.py:57

models/backbones/visual/resnet.py:277

Methodforward

(self, x, y)

SceneSeg/BiLSTM_protocol.py:38

Methodforward

(self, x, y)

SceneSeg/BiLSTM_protocol.py:75

Methodforward

(self, x: Tensor)

models/backbones/visual/resnet.py:66

Methodforward

(self, x: Tensor)

models/backbones/visual/resnet.py:120

Methodforward

(self, x: Tensor, is_fc=True)

models/backbones/visual/resnet.py:248

Methodforward

Input: query , key (images) Output: logits, targets

models/core/SCRL_MoCo.py:146

Methodforward_SCRL

(self, img_q, img_k)

models/core/SCRL_MoCo.py:156

Methodforward_moco_old

Input: im_q: a batch of query images im_k: a batch of key images Output: logits, targets

models/core/SCRL_MoCo.py:211

Functionget_encoder

(model_name='resnet50', weight_path='', modal='v', input_channel=9, ssl_type='moco')

models/backbones/visual/resnet.py:438

Functionget_loader

(cfg)

models/factory.py:41

Methodget_q_and_k_index_cluster

(self, embeddings, return_group=False)

models/core/SCRL_MoCo.py:130

Functionget_training_stuff

(cfg, gpu, ngpus_per_node)

models/factory.py:78

next →1–100 of 115, ranked by callers