MCPcopy Create free account

hub / github.com/alipay/Ant-Multi-Modal-Framework / functions

Functions4,004 in github.com/alipay/Ant-Multi-Modal-Framework

⨍ Functions 4,004◇ Types & classes 941↳ Endpoints 1
↓ 1 callersMethodapply_system_template
(self, text)
prj/M2_omni/models/processing_m2omni.py:222
↓ 1 callersMethodapproximate_hamming_similarity
Approximate Hamming similarity.
antmmf/modules/losses/pairwise_loss.py:43
↓ 1 callersMethodassert_files
(self, folder)
antmmf/scripts/hm_convert.py:26
↓ 1 callersMethodattention
(self, x: torch.Tensor, attn_mask: torch.Tensor)
prj/dmae_vtp/roi_univl/univl/model/dmae_utils.py:601
↓ 1 callersFunctionaugment_list
()
prj/M2_Encoder/vlmo/transforms/randaug.py:163
↓ 1 callersFunctionavailable_models
Returns the names of available CNCLIP models
antmmf/modules/vision/backbone/clip/cn_model.py:352
↓ 1 callersFunctionaverage_blur
Averageing: blur is done by convolving an image with a normalized box filter. @param ksize: the kernel size of the boxFilter, which must be a
antmmf/utils/dataset_utils.py:398
↓ 1 callersMethodbackward
(ctx, grad_output)
antmmf/modules/functional/swish.py:25
↓ 1 callersFunctionbasic_clean
(text)
antmmf/modules/vision/backbone/clip/simple_tokenizer.py:74
↓ 1 callersFunctionbilateral_blur
Bilateral Filtering, which takes a Gaussian filter in space, but one more Gaussian filter which is a function of pixel difference.
antmmf/utils/dataset_utils.py:385
↓ 1 callersFunctionblock_read
(files, size=65536)
antmmf/datasets/utils.py:58
↓ 1 callersMethodbpe
(self, token)
antmmf/modules/vision/backbone/clip/simple_tokenizer.py:110
↓ 1 callersFunctionbroadcast_scalar
(scalar, src=0, device="cpu")
antmmf/utils/distributed_utils.py:72
↓ 1 callersFunctionbroadcast_tensor
(tensor, src=0)
antmmf/utils/distributed_utils.py:61
↓ 1 callersFunctionbroadcast_tensor
(tensor, src=0)
antmmf/utils/distributed.py:52
↓ 1 callersMethodbuild
(self)
prj/snps3_vtp/roi_univl/univl/model/univl_video_base.py:24
↓ 1 callersMethodbuild
(self)
prj/dmae_vtp/roi_univl/univl/model/univl_video_base.py:24
↓ 1 callersMethodbuild
(self)
prj/base_vtp/roi_univl/univl/model/univl_video_base.py:24
↓ 1 callersMethodbuild
(self)
prj/cnvid_vtp/roi_univl/univl/model/univl_video_base.py:24
↓ 1 callersMethodbuild
Function to be implemented by the child class, in case they need to build their model separately than ``__init__`` for training phrase.
antmmf/models/base_model.py:94
↓ 1 callersFunctionbuild_adversarial_training
(config, model)
antmmf/trainers/build.py:29
↓ 1 callersMethodbuild_attention_mask
(self)
antmmf/modules/vision/backbone/clip/model.py:402
↓ 1 callersMethodbuild_base
(self)
prj/snps3_vtp/roi_univl/univl/model/univl_base.py:62
↓ 1 callersMethodbuild_base
(self)
prj/dmae_vtp/roi_univl/univl/model/univl_base.py:62
↓ 1 callersMethodbuild_base
(self)
prj/base_vtp/roi_univl/univl/model/univl_base.py:62
↓ 1 callersMethodbuild_base
(self)
prj/cnvid_vtp/roi_univl/univl/model/univl_base.py:62
↓ 1 callersFunctionbuild_dataset_sampler
Args: dataset (antmmf.datasets.BaseDataset): instance of BaseDataset sampler_config(configNode): default_config (dict)
antmmf/datasets/build.py:19
↓ 1 callersFunctionbuild_decoder
(config, *args, **kwargs)
antmmf/modules/build.py:75
↓ 1 callersMethodbuild_decoder_layer
(self, args, depth, is_moe_layer=False, is_encoder_decoder=False)
prj/M2_Encoder/vlmo/torchscale/architecture/decoder.py:309
↓ 1 callersMethodbuild_encoder_attention
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/decoder.py:119
↓ 1 callersMethodbuild_ffn
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/decoder.py:97
↓ 1 callersMethodbuild_ffn
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/encoder.py:87
↓ 1 callersMethodbuild_for_test
(self)
antmmf/models/bert.py:48
↓ 1 callersMethodbuild_hier_tree
(label_schema, root=None)
antmmf/modules/utils.py:375
↓ 1 callersMethodbuild_image_encoder
(self)
antmmf/modules/encoders/visual_encoder.py:586
↓ 1 callersFunctionbuild_interpreter
(config: Configuration, predictor)
antmmf/modules/build.py:38
↓ 1 callersFunctionbuild_model
(state_dict: dict, model_type: str = "all")
antmmf/modules/vision/backbone/clip/model.py:516
↓ 1 callersMethodbuild_model
(self)
antmmf/modules/encoders/visual_encoder.py:218
↓ 1 callersMethodbuild_model
(self)
antmmf/modules/vision/backbone/pvt.py:61
↓ 1 callersMethodbuild_model
(self)
antmmf/modules/vision/backbone/efficientnet.py:69
↓ 1 callersMethodbuild_model
(self, model_name, config, pretrained)
antmmf/modules/vision/backbone/clip/cn_model.py:244
↓ 1 callersMethodbuild_model
(self, model_name, config, pretrained)
antmmf/modules/vision/backbone/clip/cn_model.py:268
↓ 1 callersMethodbuild_model
(self)
antmmf/modules/vision/backbone/clip/model.py:466
↓ 1 callersMethodbuild_model
(self)
antmmf/modules/vision/backbone/clip/model.py:492
↓ 1 callersMethodbuild_multimodal_embedding_input
( self, input_txt, attention_mask, text_token_type, cls_id, se
prj/snps3_vtp/roi_univl/roi/model.py:101
↓ 1 callersMethodbuild_multimodal_embedding_input
( self, input_txt, attention_mask, text_token_type, cls_id, se
prj/dmae_vtp/roi_univl/roi/model.py:101
↓ 1 callersMethodbuild_multimodal_embedding_input
( self, input_txt, attention_mask, text_token_type, cls_id, se
prj/base_vtp/roi_univl/roi/model.py:101
↓ 1 callersMethodbuild_multimodal_embedding_input
( self, input_txt, attention_mask, text_token_type, cls_id, se
prj/cnvid_vtp/roi_univl/roi/model.py:101
↓ 1 callersMethodbuild_multimodal_embedding_input
( self, input_txt, attention_mask, text_token_type, cls_id, se
antmmf/modules/encoders/multimodal_encoder.py:96
↓ 1 callersFunctionbuild_online_predictor
(model_dir=None, config_yaml=None)
antmmf/predictors/build.py:30
↓ 1 callersMethodbuild_output_projection
( self, args, )
prj/M2_Encoder/vlmo/torchscale/architecture/decoder.py:293
↓ 1 callersMethodbuild_output_projection
( self, args, )
prj/M2_Encoder/vlmo/torchscale/architecture/encoder.py:298
↓ 1 callersFunctionbuild_phoc
antmmf/utils/phoc/src/cphoc.c:12
↓ 1 callersFunctionbuild_predictor
(config)
antmmf/predictors/build.py:10
↓ 1 callersFunctionbuild_predictor_from_args
(args, *rest, **kwargs)
antmmf/predictors/build.py:18
↓ 1 callersMethodbuild_self_attention
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/decoder.py:108
↓ 1 callersMethodbuild_self_attention
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/encoder.py:98
↓ 1 callersMethodbuild_self_attention
(self, embed_dim, args)
prj/M2_Encoder/vlmo/torchscale/architecture/encoder.py:286
↓ 1 callersFunctionbuild_trainer
( config: Configuration, args: Union[Namespace, Configuration], )
antmmf/trainers/build.py:12
↓ 1 callersMethodbuild_transformer_input
( self, visual_embed_dict, text_embed_dict, caption_input )
prj/snps3_vtp/roi_univl/univl/model/univl_video_base.py:206
↓ 1 callersMethodbuild_transformer_input
(self, visual_embed_dict, text_embed_dict)
prj/snps3_vtp/roi_univl/univl/model/univl_base.py:262
↓ 1 callersMethodbuild_transformer_input
( self, visual_embed_dict, text_embed_dict, caption_input )
prj/dmae_vtp/roi_univl/univl/model/univl_video_base.py:206
↓ 1 callersMethodbuild_transformer_input
(self, visual_embed_dict, text_embed_dict)
prj/dmae_vtp/roi_univl/univl/model/univl_base.py:262
↓ 1 callersMethodbuild_transformer_input
( self, visual_embed_dict, text_embed_dict, caption_input )
prj/base_vtp/roi_univl/univl/model/univl_video_base.py:206
↓ 1 callersMethodbuild_transformer_input
(self, visual_embed_dict, text_embed_dict)
prj/base_vtp/roi_univl/univl/model/univl_base.py:262
↓ 1 callersMethodbuild_transformer_input
( self, visual_embed_dict, text_embed_dict, caption_input )
prj/cnvid_vtp/roi_univl/univl/model/univl_video_base.py:206
↓ 1 callersMethodbuild_transformer_input
(self, visual_embed_dict, text_embed_dict)
prj/cnvid_vtp/roi_univl/univl/model/univl_base.py:262
↓ 1 callersMethodbuild_tree_from_file
(label_schema_file)
antmmf/modules/utils.py:397
↓ 1 callersMethodbuild_video_encoder
(self)
antmmf/modules/encoders/visual_encoder.py:590
↓ 1 callersFunctionbuilt
Check if '.built' flag has been set for that task. If a version_string is provided, this has to match, or the version is regarded as not
antmmf/utils/download.py:165
↓ 1 callersMethodcal_itm_loss
(self, sample_list, model_output)
prj/snps3_vtp/roi_univl/roi/model.py:510
↓ 1 callersMethodcal_itm_loss
(self, sample_list, model_output)
prj/dmae_vtp/roi_univl/roi/model.py:510
↓ 1 callersMethodcal_itm_loss
(self, sample_list, model_output)
prj/base_vtp/roi_univl/roi/model.py:510
↓ 1 callersMethodcal_itm_loss
(self, sample_list, model_output)
prj/cnvid_vtp/roi_univl/roi/model.py:510
↓ 1 callersMethodcal_mlm
(self, sample_list, model_output)
prj/snps3_vtp/roi_univl/roi/model.py:521
↓ 1 callersMethodcal_mlm
(self, sample_list, model_output)
prj/dmae_vtp/roi_univl/roi/model.py:521
↓ 1 callersMethodcal_mlm
(self, sample_list, model_output)
prj/base_vtp/roi_univl/roi/model.py:521
↓ 1 callersMethodcal_mlm
(self, sample_list, model_output)
prj/cnvid_vtp/roi_univl/roi/model.py:521
↓ 1 callersMethodcal_region_loss
(self, sample_list, model_output)
prj/snps3_vtp/roi_univl/roi/model.py:485
↓ 1 callersMethodcal_region_loss
(self, sample_list, model_output)
prj/dmae_vtp/roi_univl/roi/model.py:485
↓ 1 callersMethodcal_region_loss
(self, sample_list, model_output)
prj/base_vtp/roi_univl/roi/model.py:485
↓ 1 callersMethodcal_region_loss
(self, sample_list, model_output)
prj/cnvid_vtp/roi_univl/roi/model.py:485
↓ 1 callersMethodcalculate
Calculate accuracy and return it back. Args: sample_list (SampleList): SampleList provided by DataLoader for
antmmf/modules/metrics/bleu4.py:33
↓ 1 callersMethodcalculate
Calculate rouge score and return it back. Args: sample_list (SampleList): SampleList provided by DataLoader for
antmmf/modules/metrics/rouge_antmmf.py:36
↓ 1 callersFunctioncalu_recall
(txt_feats, img_feats, txt2img_gt, img2txt_gt)
prj/M2_Encoder/eval_retrieval.py:71
↓ 1 callersMethodchange_dataset
(self)
antmmf/tasks/base_task.py:148
↓ 1 callersMethodchange_dataset
(self)
antmmf/tasks/base_task.py:343
↓ 1 callersFunctioncheck_integrity
(videopath, md5)
prj/cnvid_vtp/download_cnvid/download_cnvid_video.py:36
↓ 1 callersMethodchecksum
(self, file, hashes)
antmmf/scripts/hm_convert.py:121
↓ 1 callersMethodcircle_loss
(img_embed, text_embed, m=0.45, gamma=32)
antmmf/models/concat_bert.py:214
↓ 1 callersFunctionclip_gradients
(model, i_iter, writer, config)
antmmf/utils/general.py:47
↓ 1 callersMethodcollate_fn
(sample_list: List[Tuple])
antmmf/datasets/features/vision/feature_saver.py:117
↓ 1 callersMethodcombine_embeddings
(self, *args)
antmmf/models/ant_mmf.py:277
↓ 1 callersMethodcompare_hier_label
Args: pred_hier_label: 输出的最细粒度的层次分类标签 target_hier_label: 细粒度或粗粒度标签 Returns:
antmmf/modules/utils.py:353
↓ 1 callersMethodcompute_answers_scores
Generate VQA based answer scores for answers_indices. Args: answers_indices (torch.LongTensor): tensor containing indices of the a
antmmf/datasets/processors/mm_processors.py:224
↓ 1 callersMethodcompute_bias
(self, qlen, klen, step=None)
prj/M2_Encoder/vlmo/torchscale/component/relative_position_bias.py:41
↓ 1 callersFunctioncompute_mask
(D, H, W, window_size, shift_size, device)
antmmf/modules/vision/backbone/video_swin.py:383
↓ 1 callersMethodconcat
(self, e1_embed, rel_embed)
antmmf/modules/graph.py:781
↓ 1 callersFunctionconfig_setting
(custom_config: dict)
prj/M2_Encoder/vlmo/utils/beit_utils.py:38
↓ 1 callersFunctionconstruct_trie_tree
(word_list)
prj/Pink/pink/eval/model_seed.py:68
← previousnext →1,001–1,100 of 4,004, ranked by callers