MCPcopy Create free account

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

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

↓ 1 callersFunctionconstruct_trie_tree
(word_list)
prj/Pink/pink/eval/model_vqav2.py:286
↓ 1 callersFunctionconstruct_trie_tree
(word_list)
prj/Pink/pink/eval/model_gqa.py:76
↓ 1 callersFunctionconstruct_trie_tree
(word_list)
prj/Pink/pink/eval/model_iconqa.py:68
↓ 1 callersMethodcontrastive_loss
(img_embed, text_embed, gamma=0.1)
antmmf/models/concat_bert.py:197
↓ 1 callersFunctionconvert_value_to_str
(value: Any)
antmmf/common/configuration.py:89
↓ 1 callersFunctionconvert_weights_to_fp16
Convert applicable model parameters to fp16
prj/Pink/pink/model/eva_vit.py:406
↓ 1 callersMethodcopy
Get a copy of the current SampleList Returns: SampleList: Copy of current SampleList.
antmmf/structures/sample.py:191
↓ 1 callersFunctioncos_amsoftmax_loss
(y_pred, y_true, margin, scale=30.0)
antmmf/modules/losses/cos_ams_softmax_loss.py:10
↓ 1 callersFunctioncount_parameters
(model)
antmmf/utils/general.py:160
↓ 1 callersMethodcount_words_belong_to_what
(self, caption)
prj/snps3_vtp/auxiliary_files/generate_ss_word_json/1st_ss_word_mining.py:92
↓ 1 callersFunctioncreate_common_coco_eval
(coco_eval, img_ids, eval_imgs)
antmmf/modules/metrics/evaluators/coco_eval.py:208
↓ 1 callersMethodcreate_saver
(self, obj)
antmmf/datasets/features/vision/feature_saver.py:227
↓ 1 callersMethodcsv_dump
(self, filepath)
antmmf/common/test_reporter.py:91
↓ 1 callersFunctioncustomed_forward
(module)
antmmf/utils/register_fp32.py:36
↓ 1 callersFunctioncv2ImgAddText
interface that supports chinese text visualization
antmmf/utils/visual_utils/visualization_utils.py:41
↓ 1 callersMethodde_normalize
(inputs, audio_mean, audio_std)
prj/M2_omni/models/feature_extraction_sanm.py:359
↓ 1 callersMethoddecode
Decode a list of string notations to specify blocks inside the network. Args: string_list (list[str]): A list of strings, each st
antmmf/modules/vision/backbone/efficientnet.py:554
↓ 1 callersFunctiondecode_whitespaces
(content)
prj/M2_Encoder/vlmo/tokenizer/tokenization_glm.py:213
↓ 1 callersMethoddecompress_zip
(self, dest, fname, password=None)
antmmf/scripts/hm_convert.py:137
↓ 1 callersFunctiondefault_bpe
()
antmmf/modules/vision/backbone/clip/simple_tokenizer.py:12
↓ 1 callersFunctiondefault_vocab
()
antmmf/modules/vision/backbone/clip/cn_tokenizer.py:21
↓ 1 callersMethoddequeue_and_enqueue
(self, vis_keys, txt_keys)
prj/snps3_vtp/roi_univl/univl/model/moco_utils.py:84
↓ 1 callersMethoddequeue_and_enqueue
(self, vis_keys, txt_keys)
prj/dmae_vtp/roi_univl/univl/model/moco_utils.py:84
↓ 1 callersMethoddequeue_and_enqueue
(self, vis_keys, txt_keys)
prj/base_vtp/roi_univl/univl/model/moco_utils.py:84
↓ 1 callersMethoddequeue_and_enqueue
(self, vis_keys, txt_keys)
prj/cnvid_vtp/roi_univl/univl/model/moco_utils.py:84
↓ 1 callersFunctiondownload
Download file using `requests`. If ``redownload`` is set to false, then will not download tar file again if it is present (default ``Tru
antmmf/utils/download.py:207
↓ 1 callersFunctiondownload_based_on_config
(config, resources, download_path, version)
antmmf/utils/download.py:439
↓ 1 callersMethoddownload_file
(self, download_path)
antmmf/utils/download.py:131
↓ 1 callersFunctiondownload_from_google_drive
Use the requests package to download a file from Google Drive.
antmmf/utils/download.py:406
↓ 1 callersFunctiondownload_resource
(resource, download_path)
antmmf/utils/download.py:351
↓ 1 callersFunctiondraw
Args: image_path: BGR format numpy or image_path box_list(Box): list of Box object Returns: VisTool object, with .c
antmmf/utils/visual_utils/visualization_utils.py:154
↓ 1 callersMethoddraw
(self, coords, color=[255, 0, 0], prefix=None, canvas=None)
antmmf/utils/visual_utils/visualization_utils.py:91
↓ 1 callersFunctiondrop_connect
Drop connect. Args: input (tensor: BCWH): Input of this structure. p (float: 0.0~1.0): Probability of drop connection. tr
antmmf/modules/utils.py:91
↓ 1 callersFunctiondrop_path
(x, drop_prob: float = 0.0, training: bool = False)
antmmf/modules/vision/backbone/video_swin.py:49
↓ 1 callersMethoddryrun
(self)
antmmf/common/metrics_reporter.py:138
↓ 1 callersFunctionduplicate_interleave
A simple version of `torch.repeat_interleave` for duplicating a matrix while interleaving the copy.
prj/M2_Encoder/vlmo/torchscale/component/xpos_relative_position.py:22
↓ 1 callersFunctionefficientnet
Create BlockArgs and GlobalParams for efficientnet model. Args: width_coefficient (float) depth_coefficient (float) image
antmmf/modules/vision/backbone/efficientnet.py:610
↓ 1 callersMethodencode
(self, text)
antmmf/modules/vision/backbone/clip/simple_tokenizer.py:151
↓ 1 callersMethodencode_audio_post_processing
(self, audio_embeds: torch.Tensor, n_query=None, audio_compress=3)
prj/M2_omni/models/modeling_m2omni.py:640
↓ 1 callersMethodencode_image
(self, images)
prj/Pink/pink/model/pink.py:181
↓ 1 callersMethodencode_image
(self, image)
antmmf/modules/vision/backbone/clip/model.py:414
↓ 1 callersMethodencode_label_str
hier_label and hier_param may not have consistent shape for all samples in the batch, using mask_padding padding to the same shape.
antmmf/modules/utils.py:208
↓ 1 callersMethodencode_multilabel_str
label_str has multiple labels. Now TreeNode.ParamGroup must be 1, and TreeNode.ParamGroup > 1 will be supported in subsequent version
antmmf/modules/utils.py:235
↓ 1 callersMethodencode_text
(self, text)
antmmf/modules/vision/backbone/clip/model.py:417
↓ 1 callersFunctionencode_whitespaces
(content)
prj/M2_Encoder/vlmo/tokenizer/tokenization_glm.py:208
↓ 1 callersFunctioneuclidean_dist
(x, y)
antmmf/modules/losses/eet_loss.py:14
↓ 1 callersMethodeuclidean_distance
This is the squared Euclidean distance.
antmmf/modules/losses/pairwise_loss.py:38
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_vsr.py:93
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_seed.py:130
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_pointingvqa_looktwice.py:103
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_object365.py:155
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_pointingvqa_local.py:103
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_vg_base_batch.py:141
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_vqav2.py:319
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_gqa.py:128
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_okvqa.py:94
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_v7wgrounding.py:83
↓ 1 callersFunctioneval_model
(args)
prj/Pink/pink/eval/model_iconqa.py:125
↓ 1 callersFunctionevaluate
(args)
prj/Pink/pink/eval/eval_vqav2.py:67
↓ 1 callersFunctionevaluate
Run per image evaluation on given images and store results (a list of dict) in self.evalImgs :return: None
antmmf/modules/metrics/evaluators/coco_eval.py:224
↓ 1 callersMethodevaluate
(self, quesIds=None)
prj/Pink/pink/eval/vqa_tools/vqa_result.py:210
↓ 1 callersMethodevaluate_single_batch
(self)
antmmf/trainers/base_trainer.py:805
↓ 1 callersMethodexecute
(self)
antmmf/scripts/features/lmdb_conversion.py:139
↓ 1 callersFunctionexpand2square
(pil_img, background_color=(255, 255, 255))
prj/Pink/pink/eval/model_vsr.py:50
↓ 1 callersMethodexport
(self, attr_info, instance)
antmmf/modules/interpret/saliency_interpreter.py:97
↓ 1 callersMethodextract
(self)
antmmf/scripts/extract_vocabulary.py:24
↓ 1 callersMethodextract
(self)
antmmf/scripts/features/lmdb_conversion.py:103
↓ 1 callersMethodextract_and_save
(self, image_paths, batch_size=16)
antmmf/datasets/features/vision/feature_saver.py:178
↓ 1 callersMethodextract_audio_feature
( self, pixel_values_audios, attention_mask_audio, n_query=None, audio
prj/M2_omni/models/modeling_m2omni.py:661
↓ 1 callersFunctionextract_bboxes
(bottom_up_csv_file)
antmmf/scripts/features/extract_features.py:215
↓ 1 callersFunctionextract_dataset_pool5
(image_dir, save_dir, prefix, ext_filter)
antmmf/scripts/features/extract_resnet_features.py:66
↓ 1 callersFunctionextract_dataset_pool5
(image_dir, save_dir, total_group, group_id, ext_filter)
antmmf/scripts/features/extract_resnet152_feat.py:71
↓ 1 callersFunctionextract_feats
(model, texts, images, args)
prj/M2_Encoder/eval_retrieval.py:47
↓ 1 callersMethodextract_features
(self)
antmmf/scripts/features/extract_features_vmb.py:227
↓ 1 callersFunctionextract_image_feat
(img_file)
antmmf/scripts/features/extract_resnet_features.py:52
↓ 1 callersFunctionextract_image_feat
(img_file)
antmmf/scripts/features/extract_resnet152_feat.py:52
↓ 1 callersFunctionextract_vision_info
(conversations: list[dict] | list[list[dict]])
prj/M2_omni/models/m2omni_utils.py:388
↓ 1 callersMethodfeature_io
(self, image_feat_path)
antmmf/datasets/features/feature_readers.py:162
↓ 1 callersFunctionfetch_audio
(ele: dict[str, str | torch.Tensor], return_tensor="pt")
prj/M2_omni/models/m2omni_utils.py:367
↓ 1 callersFunctionfetch_video
(ele: dict, image_factor: int = IMAGE_FACTOR)
prj/M2_omni/models/m2omni_utils.py:320
↓ 1 callersMethodfinalize
(self)
antmmf/common/checkpoint.py:368
↓ 1 callersFunctionfix_audio_encoder_bf16_
Fix numpy not recognizing torch.bfloat16 in funasr's attention implementation for audio encoder.
prj/M2_omni/models/sanm_audio.py:39
↓ 1 callersMethodfix_init_weight
(self)
prj/Pink/pink/model/eva_vit.py:309
↓ 1 callersFunctionfixed_pos_embedding
(x)
prj/M2_Encoder/vlmo/torchscale/component/xpos_relative_position.py:8
↓ 1 callersMethodflatten_for_bert
(self, sample_list, **kwargs)
antmmf/models/visual_bert.py:456
↓ 1 callersFunctionflatten_list
(k)
antmmf/utils/general.py:526
↓ 1 callersFunctionformat_coco_detection_result
Args: image_ids(list): coco image id list pred_boxes(np.ndArray): Nx6, (x1, y1, x2, y2, conf, cls), absolute coords Returns:
antmmf/modules/metrics/evaluators/coco_eval.py:275
↓ 1 callersMethodformat_for_evalai
(self, report)
antmmf/datasets/base_dataset.py:64
↓ 1 callersMethodformat_for_overwrite_obs_labels
(self, report)
antmmf/datasets/mm_dataset.py:147
↓ 1 callersMethodformat_state_key
Can be implemented if something special needs to be done to the key when pretrained model is being loaded. This will adapt and return
antmmf/models/base_model.py:159
↓ 1 callersMethodforward
(self, x)
prj/M2_omni/models/modeling_llama_3d.py:252
↓ 1 callersMethodforward
(self, forward_params)
prj/M2_Encoder/ms_wrapper.py:32
↓ 1 callersMethodforward
(self, x)
prj/Pink/pink/model/eva_vit.py:62
↓ 1 callersMethodforward
(self, video_segments)
antmmf/modules/encoders/visual_encoder.py:490
↓ 1 callersMethodforward
(self, x, H, W)
antmmf/modules/vision/backbone/pvt.py:310
↓ 1 callersMethodforward
EfficientNet's forward function. Calls extract_features to extract features, applies final linear layer, and returns logits. Args:
antmmf/modules/vision/backbone/efficientnet.py:284
↓ 1 callersMethodforward
(self, image, text)
antmmf/modules/vision/backbone/clip/model.py:433
↓ 1 callersMethodforward
(self, sample_list)
antmmf/models/mm_adversarial.py:140
↓ 1 callersMethodforward2embed
(self, x)
antmmf/models/spkResNet.py:248
↓ 1 callersMethodforward_decoder
( self, tgt: torch.Tensor, memory: torch.Tensor, tgt_mask: Optional[torch.Tens
antmmf/modules/transformers/position_enhance.py:84
← previousnext →1,101–1,200 of 4,004, ranked by callers