MCPcopy Create free account

hub / github.com/SuperSupermoon/MedViLL / functions

Functions425 in github.com/SuperSupermoon/MedViLL

⨍ Functions 425◇ Types & classes 96
Functionsave_on_master
(*args, **kwargs)
utils/utils.py:294
Methodselect_beam_items
(x, ids)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/model.py:1330
Functionset_seed
(seed)
CNN_BERT/main_cnn_bert.py:22
Methodstep
Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/optimization.py:202
Functionstore_preds_to_disk
(tgts, preds, args)
utils/utils.py:51
Functionswish
(x)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/model.py:109
Functionswish
(x)
downstream_task/retrieval/cxrbert_origin.py:132
Functionswish
(x)
models/MedViLL_origin.py:133
Methodsynchronize_between_processes
Warning: does not synchronize the deque!
utils/utils.py:93
Methodsynchronize_between_processes
(self)
utils/utils.py:173
Methodsynchronize_between_processes
Warning: does not synchronize the deque!
downstream_task/report_generation_and_vqa/utils.py:195
Methodsynchronize_between_processes
(self)
downstream_task/report_generation_and_vqa/utils.py:375
Functiontest
(args)
downstream_task/classification/cls.py:294
Methodtokenize
(self, text)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/tokenization.py:106
Methodtokenize
(self, text)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/tokenization.py:177
Methodtokenize
Tokenizes a piece of text.
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/tokenization.py:193
Methodtokenize
Tokenizes a piece of text into its word pieces. This uses a greedy longest-match-first algorithm to perform tokenization using the
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/tokenization.py:304
Functiontrunc_normal_
(tensor, mean=0., std=1., a=-2., b=2.)
downstream_task/report_generation_and_vqa/utils.py:482
Functiontruncate_seq_pair
tokens_a : IMG patches tokens_b : TXT tokens max_length: bert-base(512) using all img patches, only truncate txt tokens if exceed max
utils/utils.py:35
Functiontruncate_seq_pair
Truncates a sequence pair in place to the maximum length. Copied from https://github.com/huggingface/pytorch-pretrained-BERT
downstream_task/classification/utils/utils.py:32
Methodupdate
(self, **kwargs)
utils/utils.py:142
Methodvalue
(self)
utils/utils.py:125
Functionwarmup_constant
(x, warmup=0.002)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/optimization.py:39
Functionwarmup_cosine
(x, warmup=0.002)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/optimization.py:33
Functionwrapper
(url: str, *args, **kwargs)
downstream_task/report_generation_and_vqa/pytorch_pretrained_bert/file_utils.py:124
← previous401–425 of 425, ranked by callers