MCPcopy Create free account

hub / github.com/binbinjiang/CVT-SLR / functions

Functions220 in github.com/binbinjiang/CVT-SLR

⨍ Functions 220◇ Types & classes 38
↓ 14 callersMethodconvert_to_string
(self, tokens, vocab, seq_len)
ctcdecode/tests/test_decode.py:34
↓ 8 callersMethodDecoderState
ctcdecode/ctcdecode/src/ctc_beam_search_decoder.cpp:17
↓ 8 callersMethoddecode
Conducts the beamsearch on model outputs and return results. Args: probs (Tensor) - A rank 3 tensor representing model output
ctcdecode/ctcdecode/__init__.py:189
↓ 7 callersMethodis_character_based
retrun true if the language model is character based
ctcdecode/ctcdecode/src/scorer.h:60
↓ 7 callersMethodto
(self, device)
utils/optimizer.py:54
↓ 6 callersMethodprint_log
(self, str, path=None, print_time=True)
utils/record.py:15
↓ 5 callersMethoddecode
Conducts the beamsearch on model outputs and return results. Args: probs (Tensor) - A rank 3 tensor representing model output
ctcdecode/ctcdecode/__init__.py:53
↓ 5 callersFunctionlog_sum_exp
ctcdecode/ctcdecode/src/decoder_utils.h:48
↓ 4 callersMethod__init__
(self, d_model, max_len=5000)
gloss_encoder.py:84
↓ 4 callersMethodget_path_vec
ctcdecode/ctcdecode/src/path_trie.cpp:109
↓ 3 callersFunctioncompile_test
(header, library)
ctcdecode/setup.py:41
↓ 3 callersMethodget_dict_size
return the dictionary size of language model
ctcdecode/ctcdecode/src/scorer.h:57
↓ 3 callersMethodget_max_order
return the max order
ctcdecode/ctcdecode/src/scorer.h:54
↓ 3 callersMethodinsert
(ind, label, binfo)
utils/video_augmentation.py:59
↓ 3 callersMethodnormalize
(self, video, label, file_id=None)
dataset/dataloader_video_14.py:68
↓ 2 callersMethodBeamSearch
CTCBeamDecoder Shape: - Input: nn_output (B, T, N), which should be passed through a softmax layer - Output:
utils/decode.py:30
↓ 2 callersMethodMaxDecode
(self, nn_output, vid_lgt, grad_cam_mode=False)
utils/decode.py:60
↓ 2 callersMethod__init__
(self)
cvtslr_model.py:12
↓ 2 callersFunctionbeam_decode
ctcdecode/ctcdecode/src/binding.cpp:35
↓ 2 callersMethoddata_to_device
(self, data)
utils/device.py:30
↓ 2 callersMethoddecode
(self, nn_output, vid_lgt, batch_first=True, probs=False)
utils/decode.py:21
↓ 2 callersMethoddecode
ctcdecode/ctcdecode/src/ctc_beam_search_decoder.cpp:164
↓ 2 callersFunctiondownload_extract
(url, dl_path)
ctcdecode/setup.py:13
↓ 2 callersMethodeval
(self)
run_demo.py:33
↓ 2 callersMethodget_log_cond_prob
ctcdecode/ctcdecode/src/scorer.cpp:74
↓ 2 callersFunctionimport_class
(name)
run_demo.py:13
↓ 2 callersMethodload_state_dict
(self, state_dict)
utils/optimizer.py:51
↓ 2 callersMethodmake_ngram
ctcdecode/ctcdecode/src/scorer.cpp:163
↓ 2 callersMethodnext
ctcdecode/ctcdecode/src/ctc_beam_search_decoder.cpp:56
↓ 2 callersMethodnormalize
(self, video, label, file_id=None)
dataset/dataloader_video_14T.py:68
↓ 2 callersMethodpad
(tensor, length)
cvtslr_model.py:76
↓ 2 callersFunctionseq_eval
(cfg, loader, model, device, mode, epoch, work_dir, recoder)
seq_scripts.py:5
↓ 2 callersFunctionsplit_utf8_str
ctcdecode/ctcdecode/src/decoder_utils.cpp:83
↓ 1 callersMethod_cat_directions
If the encoder is bidirectional, do the following transformation. Ref: https://github.com/IBM/pytorch-seq2seq/blob/master/seq2seq/models/
modules/BiLSTM.py:67
↓ 1 callersMethod_get_PIL_interp
(self, interp)
utils/video_augmentation.py:273
↓ 1 callersMethod_get_PIL_interp
(self, interp)
utils/video_augmentation.py:319
↓ 1 callersFunctionadd_word_to_dictionary
ctcdecode/ctcdecode/src/decoder_utils.cpp:164
↓ 1 callersFunctionadd_word_to_fst
ctcdecode/ctcdecode/src/decoder_utils.cpp:147
↓ 1 callersFunctionbeam_decode_with_given_state
ctcdecode/ctcdecode/src/binding.cpp:153
↓ 1 callersMethodbuild_dataloader
(self, dataset, mode, train_flag)
run_demo.py:96
↓ 1 callersFunctioncalculate_stats
(gt, lstm_pred)
evaluation/wer_calculation.py:123
↓ 1 callersFunctionctc_beam_search_decoder_batch
ctcdecode/ctcdecode/src/ctc_beam_search_decoder.cpp:245
↓ 1 callersFunctionctc_beam_search_decoder_batch_with_states
ctcdecode/ctcdecode/src/ctc_beam_search_decoder.cpp:288
↓ 1 callersMethoddefine_lr_scheduler
(self, optimizer, milestones)
utils/optimizer.py:35
↓ 1 callersMethoddelete
(ind, label, binfo)
utils/video_augmentation.py:50
↓ 1 callersMethodeval_network
(self, x, len_x)
cvtslr_model.py:93
↓ 1 callersFunctionevaluate
(prefix="./", mode="dev", evaluate_dir=None, evaluate_prefix=None,output_file=None)
evaluation/wer_calculation.py:166
↓ 1 callersFunctionget_beam_search_result
ctcdecode/ctcdecode/src/decoder_utils.cpp:48
↓ 1 callersMethodget_path_trie
ctcdecode/ctcdecode/src/path_trie.cpp:38
↓ 1 callersFunctionget_pruned_log_probs
ctcdecode/ctcdecode/src/decoder_utils.cpp:10
↓ 1 callersMethodget_sent_log_prob
ctcdecode/ctcdecode/src/scorer.cpp:95
↓ 1 callersFunctionget_utf8_str_len
ctcdecode/ctcdecode/src/decoder_utils.cpp:75
↓ 1 callersFunctionget_wer_delsubins
(ref, hyp, merge_same=False, align_results=False, penalty={'ins': 1, 'del': 1, 'sub': 1}
evaluation/wer_calculation.py:27
↓ 1 callersFunctionis_character_based
ctcdecode/ctcdecode/src/binding.cpp:271
↓ 1 callersMethodis_empty
ctcdecode/ctcdecode/src/path_trie.h:40
↓ 1 callersMethoditerate_to_vec
ctcdecode/ctcdecode/src/path_trie.cpp:128
↓ 1 callersMethodload_data
(self)
run_demo.py:83
↓ 1 callersFunctionload_groundtruth
(fpath)
evaluation/wer_calculation.py:5
↓ 1 callersMethodload_model_weights
(self, model, weight_path)
run_demo.py:73
↓ 1 callersFunctionload_prediction
(fpath)
evaluation/wer_calculation.py:15
↓ 1 callersMethodloading
(self)
run_demo.py:57
↓ 1 callersMethodmasked_bn
(self, inputs, len_x)
cvtslr_model.py:75
↓ 1 callersMethodoccupy_gpu
make program appear on nvidia-smi.
utils/device.py:47
↓ 1 callersMethodone_operation
(self, *inputs)
utils/video_augmentation.py:40
↓ 1 callersMethodread_features
(self, index)
dataset/dataloader_video_14T.py:62
↓ 1 callersMethodread_features
(self, index)
dataset/dataloader_video_14.py:62
↓ 1 callersMethodread_video
(self, index, num_glosses=-1)
dataset/dataloader_video_14T.py:49
↓ 1 callersMethodread_video
(self, index, num_glosses=-1)
dataset/dataloader_video_14.py:49
↓ 1 callersMethodrecord_time
(self)
utils/record.py:27
↓ 1 callersMethodrecord_time
(self)
dataset/dataloader_video_14T.py:132
↓ 1 callersMethodrecord_time
(self)
dataset/dataloader_video_14.py:132
↓ 1 callersMethodrecord_timer
(self, key)
utils/record.py:40
↓ 1 callersMethodremove
ctcdecode/ctcdecode/src/path_trie.cpp:144
↓ 1 callersMethodrepara
(self, z_mean, z_log_var, sample=False)
cvtslr_model.py:86
↓ 1 callersMethodreset_params
ctcdecode/ctcdecode/src/scorer.cpp:122
↓ 1 callersMethodsave_arg
(self)
run_demo.py:49
↓ 1 callersFunctionsent_evaluation
(**kwargs)
evaluation/wer_calculation.py:136
↓ 1 callersMethodset_device
(self, device)
utils/device.py:11
↓ 1 callersMethodset_dictionary
ctcdecode/ctcdecode/src/path_trie.cpp:165
↓ 1 callersMethodset_matcher
ctcdecode/ctcdecode/src/path_trie.cpp:172
↓ 1 callersMethodsplit_labels
ctcdecode/ctcdecode/src/scorer.cpp:135
↓ 1 callersFunctionsplit_str
ctcdecode/ctcdecode/src/decoder_utils.cpp:102
↓ 1 callersMethodsplit_time
(self)
utils/record.py:31
↓ 1 callersMethodsubstitute
(ind, label, binfo)
utils/video_augmentation.py:71
↓ 1 callersFunctionsum_dict
(dict_list)
evaluation/wer_calculation.py:143
↓ 1 callersMethodtransform
(self)
dataset/dataloader_video_14T.py:73
↓ 1 callersMethodtransform
(self)
dataset/dataloader_video_14.py:73
↓ 1 callersMethodupdate_lgt
(self, lgt)
modules/tconv.py:41
↓ 1 callersFunctionwer_calculation
(gt_path, primary_pred)
evaluation/wer_calculation.py:150
↓ 1 callersFunctionwrite2file
(path, info, output)
seq_scripts.py:30
MethodAdd
ctcdecode/ctcdecode/src/scorer.h:26
FunctionPYBIND11_MODULE
ctcdecode/ctcdecode/src/binding.cpp:290
MethodPathTrie
ctcdecode/ctcdecode/src/path_trie.cpp:11
MethodRetriveStrEnumerateVocab
ctcdecode/ctcdecode/src/scorer.h:24
MethodScorer
ctcdecode/ctcdecode/src/scorer.cpp:16
Method__call__
(self, image, label, file_info=None)
utils/video_augmentation.py:14
Method__call__
(self, video, label, file_info)
utils/video_augmentation.py:28
Method__call__
(self, video)
utils/video_augmentation.py:83
Method__call__
(self, clip)
utils/video_augmentation.py:112
Method__call__
(self, clip)
utils/video_augmentation.py:150
next →1–100 of 220, ranked by callers