MCPcopy Create free account

hub / github.com/baaivision/Emu / functions

Functions524 in github.com/baaivision/Emu

⨍ Functions 524◇ Types & classes 120↳ Endpoints 12
↓ 91 callersMethodjoin
Stop logger process
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:313
↓ 80 callersMethodappend
(self, trie, bos_token_id)
Emu1/mm_eval/models/trie.py:28
↓ 58 callersMethodappend
(self, r: Role, p: DataMeta)
Emu2/demo/frontend/libs/meta.py:232
↓ 41 callersMethodinfo
Print information about the VQA annotation file. :return:
Emu1/mm_eval/datasets/vqa_tools/vqa.py:71
↓ 35 callersMethodpop
(self)
Emu2/demo/frontend/libs/meta.py:287
↓ 26 callersMethodwrite
Write sample to disk
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_writer.py:257
↓ 24 callersMethodparams_num
(self, module: nn.Module)
Emu2/demo/backend/hf_model/emu/chat.py:314
↓ 23 callersMethodbuild
(cls, text=None, image=None, coordinate=None, frames=None, is_error=False, *, resize: bool = True)
Emu2/demo/frontend/libs/meta.py:145
↓ 18 callersMethodget
(self)
Emu2/demo/backend/hf_model/backend.py:78
↓ 18 callersMethodget
(self)
Emu2/demo/backend/pytorch_model/backend.py:81
↓ 18 callersMethodload
(cls, d, max_size=10**5)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:40
↓ 15 callersMethodget
(self, prefix_sequence: List[int])
Emu1/mm_eval/models/trie.py:36
↓ 14 callersMethodstack
(self)
Emu2/demo/frontend/libs/meta.py:214
↓ 13 callersMethodclose
(self)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_writer.py:290
↓ 13 callersMethodformat_chatbot
(self)
Emu2/demo/frontend/libs/meta.py:236
↓ 12 callersMethod__init__
(self, cfg, args=None)
Emu1/models/modeling_t5.py:76
↓ 12 callersMethodfrom_pretrained
(cls, path: str, **kwargs)
Emu1/models/pipeline.py:237
↓ 10 callersMethoddecode
Decode an entire sample. :param dict sample: The sample, a dictionary of key-value pairs. :return: Decoded sample. :
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/dataloader/custom_wds.py:57
↓ 10 callersMethoddevice
(self, module=None)
Emu2/emu/emu.py:67
↓ 10 callersMethodparams_num
(self, module: nn.Module)
Emu2/emu/mixin.py:84
↓ 10 callersMethodupdate
(self, counter)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:31
↓ 9 callersMethoddevice
(self, module=None)
Emu2/demo/backend/hf_model/emu/diffusion.py:140
↓ 9 callersFunctionprocess_img
(img_path=None, img=None, device=torch.device("cuda"))
Emu1/utils.py:17
↓ 8 callersMethod__init__
( self, in_features, hidden_features=None, out_features=None,
Emu1/models/eva_vit_model.py:142
↓ 8 callersMethoddump
(self)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:36
↓ 8 callersMethodfrom_pretrained
( cls, path: str, **kwargs, )
Emu2/demo/backend/hf_model/emu/diffusion.py:326
↓ 7 callersMethod__init__
( self, in_features, hidden_features=None, out_features=None, act_
Emu2/emu/eva_vit.py:83
↓ 7 callersMethodencode_image
(self, image: torch.Tensor, *, n_query=None)
Emu2/emu/emu.py:78
↓ 7 callersMethodrun
Run logger process
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:228
↓ 6 callersMethodclear
(self)
Emu2/demo/frontend/libs/meta.py:281
↓ 6 callersMethodforward_features
(self, x)
Emu1/models/eva_vit_model.py:636
↓ 6 callersFunctionget_seconds
(t)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/subsamplers/clipping_subsampler.py:15
↓ 6 callersMethodparallelize
(self, device_map=None)
Emu1/models/modeling_t5.py:1056
↓ 5 callersMethoddeparallelize
(self)
Emu1/models/modeling_t5.py:1083
↓ 5 callersMethodfrom_pretrained
( cls, path: str, **kwargs, )
Emu2/demo/backend/hf_model/emu/chat.py:250
↓ 5 callersMethodgenerate
( self, samples, do_sample=False, num_beams=5, max_new_tokens=50,
Emu1/models/modeling_emu.py:101
↓ 5 callersFunctionget_yt_meta
Return yt meta dict with meta data and/or subtitles yt_metadata_args is a dict of follwing format: yt_metadata_args = { 'writesubtitle
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_reader.py:55
↓ 5 callersMethodincrement
(self, key)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:20
↓ 5 callersMethodresize
(cls, image: Image.Image | List[Image.Image] | None, *, force: bool = True)
Emu2/demo/frontend/libs/meta.py:182
↓ 5 callersMethodset_input_embeddings
(self, new_embeddings)
Emu1/models/modeling_t5.py:1097
↓ 4 callersFunctionEmu_inference
(image_list, text_sequence, system='', instruct=True, max_new_tokens=128, beam_size=5, length_penalty=0.0)
Emu1/inference.py:64
↓ 4 callersMethod_bytes_feature
Returns a bytes_list from a string / byte.
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_writer.py:206
↓ 4 callersMethodadd
(self, sequence: List[int])
Emu1/mm_eval/models/trie.py:32
↓ 4 callersMethodgenerate_image
( self, text: List[str], image: Optional[torch.Tensor] = None, placeholder: st
Emu2/emu/emu.py:93
↓ 4 callersMethodgetQuesIds
Get question ids that satisfy given filter conditions. default skips that filter :param imgIds (int array) : get question ids f
Emu1/mm_eval/datasets/vqa_tools/vqa.py:79
↓ 4 callersMethodimage2str
(self, image)
Emu2/demo/frontend/libs/meta.py:87
↓ 4 callersMethodpop_error
(self)
Emu2/demo/frontend/libs/meta.py:292
↓ 4 callersMethodput
(self, pipeline)
Emu2/demo/backend/hf_model/backend.py:86
↓ 3 callersMethod__init__
Log only every if min_interval (seconds) have elapsed since last log
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:49
↓ 3 callersMethod_to_hook
(self, x, d)
Emu2/emu/mixin.py:67
↓ 3 callersMethod_to_hook
(self, x, d)
Emu2/demo/backend/hf_model/emu/mixin.py:81
↓ 3 callersMethodevaluate
(self, quesIds=None)
Emu1/mm_eval/datasets/vqa_tools/vqa_eval.py:194
↓ 3 callersMethodfinish
Finish logger process
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:306
↓ 3 callersMethodflush
Write the buffer to disk
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_writer.py:39
↓ 3 callersMethodformat_prompt
(self)
Emu2/demo/frontend/libs/meta.py:106
↓ 3 callersMethodfrom_pretrained
( cls, model_path: str, config_path: str = osp.join(osp.dirname(__file__), "conf", "di
Emu2/emu/diffusion.py:252
↓ 3 callersMethodget_frames
(self, reader, n_frames, stride, **kwargs)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/dataloader/video_decode.py:83
↓ 3 callersFunctionget_video_dataset
Generates a webdataset given the specified parameters. Parameters: urls (str): The path to the dataset. batch_size (int): Th
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/dataloader/dataloader.py:35
↓ 3 callersMethodloadRes
Load result file and return a result object. :param resFile (str) : file name of result file :return: res (obj)
Emu1/mm_eval/datasets/vqa_tools/vqa.py:174
↓ 3 callersFunctionprocess_img
(img_path=None, img=None, device=torch.device("cuda"))
Emu1/mm_eval/models/emu.py:8
↓ 3 callersMethodput
(self, pipeline)
Emu2/demo/backend/pytorch_model/backend.py:89
↓ 3 callersFunctionrotate_half
(x)
Emu1/models/rope.py:23
↓ 3 callersFunctionshort_answer
(answer)
Emu1/mm_eval/datasets/utils.py:1
↓ 2 callersMethod__init__
( self, dim, pt_seq_len, ft_seq_len=None, custom_freqs = None,
Emu1/models/rope.py:31
↓ 2 callersMethod__init__
( self, dim, num_heads=8, qkv_bias=True, kv_dim=No
Emu1/models/transformer.py:111
↓ 2 callersMethod_add_to_trie
(sequence: List[int], trie_dict: Dict)
Emu1/mm_eval/models/trie.py:49
↓ 2 callersMethod_forward_hook
( self, module, forward, *, pre=True, post=False, )
Emu2/emu/mixin.py:44
↓ 2 callersMethod_forward_hook
( self, module, forward, *, pre=True, post=False, )
Emu2/demo/backend/hf_model/emu/mixin.py:58
↓ 2 callersMethod_initiatlize_buffer
(self)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_writer.py:25
↓ 2 callersMethod_keep_most_frequent
(self)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:25
↓ 2 callersMethod_predict
( self, samples, candidates, n_segments=1, )
Emu1/models/prediction_mixin.py:36
↓ 2 callersMethod_prepare_inputs
( self, inputs: List[Image.Image | str], device: torch.device = torch.device("cpu"),
Emu2/emu/chat.py:121
↓ 2 callersMethod_prepare_inputs
( self, inputs: List[Image.Image | str], device: torch.device = torch.device("cpu"),
Emu2/demo/backend/hf_model/emu/chat.py:160
↓ 2 callersMethod_shift_right
(self, input_ids)
Emu1/models/modeling_t5.py:998
↓ 2 callersFunctionbenchmark_train_dl
(num_frames, num_workers, bs=1, num_threads=4, resize_size=None, crop_size=None)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/benchmark/benchmark_dataloader.py:11
↓ 2 callersFunctionbroadcat
(tensors, dim = -1)
Emu1/models/rope.py:7
↓ 2 callersFunctionconfig_logger
(logger_name)
Emu2/demo/frontend/libs/utils.py:47
↓ 2 callersFunctionconfig_logger
(logger_name)
Emu2/demo/backend/hf_model/utils.py:47
↓ 2 callersFunctionconfig_logger
(logger_name)
Emu2/demo/backend/pytorch_model/utils.py:47
↓ 2 callersFunctionconfig_logger
(logger_name)
Emu2/demo/common/utils.py:47
↓ 2 callersMethodcreateIndex
(self)
Emu1/mm_eval/datasets/vqa_tools/vqa.py:53
↓ 2 callersMethoddevice
(self, module=None)
Emu2/emu/diffusion.py:67
↓ 2 callersMethoddo_log
(self, *args, **kwargs)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:72
↓ 2 callersMethoddtype
(self, module=None)
Emu2/emu/diffusion.py:72
↓ 2 callersMethoddtype
(self, module)
Emu2/demo/backend/hf_model/emu/diffusion.py:145
↓ 2 callersMethodfrom_pretrained
( cls, path: str, instruct: bool = False, dtype: torch.dtype = torch.bfloat16,
Emu2/emu/chat.py:198
↓ 2 callersFunctiongen_id
()
Emu2/demo/frontend/libs/utils.py:35
↓ 2 callersMethodgenerate
( self, text: List[str], image: Optional[torch.Tensor] = None, video: Optional
Emu2/emu/emu.py:156
↓ 2 callersFunctionget_scenes_from_scene_manager
Returns a list of cuts from a scene manager given a cut detection mode
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/subsamplers/cut_detection_subsampler.py:10
↓ 2 callersFunctionget_web_file_info
returns info about the url (currently extension and modality)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/data_reader.py:103
↓ 2 callersFunctionmake_path_absolute
(path)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/main.py:76
↓ 2 callersMethodmost_common
(self, k)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/logger.py:28
↓ 2 callersMethodnumpy_to_pil
Convert a numpy image or a batch of images to a PIL image.
Emu1/models/pipeline.py:188
↓ 2 callersMethodnumpy_to_pil
Convert a numpy image or a batch of images to a PIL image.
Emu2/emu/diffusion.py:221
↓ 2 callersMethodnumpy_to_pil
Convert a numpy image or a batch of images to a PIL image.
Emu2/demo/backend/hf_model/emu/diffusion.py:295
↓ 2 callersMethodparallel
( self, mp_rule: MutableMapping = {}, )
Emu2/emu/mixin.py:14
↓ 2 callersFunctionprepare_model
(model_name, args)
Emu1/inference.py:33
↓ 2 callersMethodprocessPunctuation
(self, inText)
Emu1/mm_eval/datasets/vqa_tools/vqa_eval.py:252
↓ 2 callersMethodprune_heads
(self, heads)
Emu1/models/modeling_t5.py:439
↓ 2 callersFunctionretrier
(runf, failed_shards, max_shard_retry)
Emu1/data/yt-sb-1b/video2dataset-1.1.0/video2dataset/distributor.py:10
next →1–100 of 524, ranked by callers