Functions — github.com/bytedance/Shot2Story

Method__init__

(self, config)

code/lavis/models/blip_models/nlvr_encoder.py:594

Method__init__

(self, config)

code/lavis/models/blip_models/nlvr_encoder.py:609

Method__init__

(self, config)

code/lavis/models/blip_models/nlvr_encoder.py:626

Method__init__

(self, config)

code/lavis/models/blip_models/nlvr_encoder.py:646

Method__init__

(self, config, add_pooling_layer=True)

code/lavis/models/blip_models/nlvr_encoder.py:688

Method__init__

(self, image_encoder, text_encoder, embed_dim, max_txt_len=40)

code/lavis/models/blip_models/blip_feature_extractor.py:38

Method__init__

code/lavis/models/blip_models/blip_retrieval.py:51

Method__init__

(self, image_encoder, text_encoder, embed_dim=256, max_txt_len=35)

code/lavis/models/blip_models/blip_image_text_matching.py:38

Method__init__

( self, image_encoder, text_encoder, text_decoder, queue_size,

code/lavis/models/blip_models/blip_pretrain.py:40

Method__init__

(self, image_encoder, text_decoder, prompt=None, max_txt_len=40)

code/lavis/models/blip_models/blip_caption.py:40

Method__init__

(self)

code/lavis/models/blip_models/blip.py:21

Method__init__

(self, drop_prob=None)

code/lavis/models/timesformer/vit_utils.py:184

Method__init__

( self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_

code/lavis/models/timesformer/vit.py:87

Method__init__

( self, dim, num_heads, layer_num, mlp_ratio=4.0, qkv_bias=Fal

code/lavis/models/timesformer/vit.py:135

Method__init__

(self, img_size=224, patch_size=16, in_chans=3, embed_dim=768)

code/lavis/models/timesformer/vit.py:266

Method__init__

( self, img_size=224, patch_size=16, in_chans=3, num_classes=1000,

code/lavis/models/timesformer/vit.py:291

Method__init__

(self, cfg, **kwargs)

code/lavis/models/timesformer/vit.py:483

Method__init__

( self, image_size=224, patch_size=16, n_frms=8, attn_drop_rate=0.0,

code/lavis/models/timesformer/vit.py:529

Method__init__

(self, hooks, named_modules, out_map=None, default_hook_type="forward")

code/lavis/models/timesformer/features.py:92

Method__init__

( self, model, out_indices=(0, 1, 2, 3, 4), out_map=None, feature_conc

code/lavis/models/timesformer/features.py:178

Method__init__

( self, model, out_indices=(0, 1, 2, 3, 4), out_map=None, feature_conc

code/lavis/models/timesformer/features.py:231

Method__init__

( self, model, out_indices=(0, 1, 2, 3, 4), out_map=None, out_as_dict=

code/lavis/models/timesformer/features.py:261

Method__init__

( self, in_channels, out_channels, kernel_size, stride=1, padd

code/lavis/models/timesformer/conv2d_same.py:82

Method__init__

( self, image_encoder, text_encoder, text_decoder, use_distill=True,

code/lavis/models/albef_models/albef_vqa.py:42

Method__init__

( self, image_encoder, text_encoder, num_classes, momentum=0.995,

code/lavis/models/albef_models/albef_classification.py:31

Method__init__

( self, image_encoder, text_encoder, num_classes, momentum=0.995,

code/lavis/models/albef_models/albef_nlvr.py:29

Method__init__

( self, image_encoder, text_encoder, queue_size, embed_dim=256,

code/lavis/models/albef_models/albef_pretrain.py:41

Method__init__

( self, image_encoder, text_encoder, queue_size, embed_dim=256,

code/lavis/models/albef_models/albef_retrieval.py:45

Method__init__

(self, image_encoder, text_encoder, embed_dim=256, max_txt_len=30)

code/lavis/models/albef_models/albef_feature_extractor.py:28

Method__init__

( self, image_question_matching_model, image_captioning_model, question_genera

code/lavis/models/img2prompt_models/img2prompt_vqa.py:51

Method__init__

( self, local_loss=False, gather_with_grad=False, cache_labels=False,

code/lavis/models/clip_models/loss.py:79

Method__init__

(self, bpe_path: str = default_bpe(), special_tokens=None)

code/lavis/models/clip_models/tokenizer.py:83

Method__init__

( self, max_size, interpolation=InterpolationMode.BICUBIC, fn="max", fill=0 )

code/lavis/models/clip_models/transform.py:29

Method__init__

( self, in_features: int, out_features: int = None, embed_dim: int = None,

code/lavis/models/clip_models/timm_model.py:139

Method__init__

( self, in_features: int, feat_size: Union[int, Tuple[int, int]], out_features

code/lavis/models/clip_models/timm_model.py:200

Method__init__

( self, max_res: int = 224, num_bands: int = 64, concat_grid=True, kee

code/lavis/models/clip_models/timm_model.py:387

Method__init__

(self, dim, max_res=224, linear_bands: bool = False)

code/lavis/models/clip_models/timm_model.py:488

Method__init__

(self, inplanes, planes, stride=1)

code/lavis/models/clip_models/model.py:53

Method__init__

( self, spacial_dim: int, embed_dim: int, num_heads: int, output_dim: int = None )

code/lavis/models/clip_models/model.py:110

Method__init__

(self, layers, output_dim, heads, image_size=224, width=64)

code/lavis/models/clip_models/model.py:164

Method__init__

(self, d_model: int, n_head: int, act_layer: Callable = nn.GELU)

code/lavis/models/clip_models/model.py:263

Method__init__

( self, width: int, layers: int, heads: int, act_layer: Callable = nn.GELU )

code/lavis/models/clip_models/model.py:289

Method__init__

( self, image_size: int, patch_size: int, width: int, layers: int,

code/lavis/models/clip_models/model.py:309

Method__init__

apply_lemmatizer: when set to True, postprocess predict_answers() result with lemmas.

code/lavis/models/blip2_models/blip2_t5_instruct.py:40

Method__init__

( self, vit_model="eva_clip_g", img_size=224, drop_path_rate=0, use_gr

code/lavis/models/blip2_models/blip2_qformer.py:45

Method__init__

apply_lemmatizer: when set to True, postprocess predict_answers() result with lemmas.

code/lavis/models/blip2_models/blip2_t5.py:40

Method__init__

( self, img_size=224, drop_path_rate=0, use_grad_checkpoint=False, vit

code/lavis/models/blip2_models/blip2_video_cascade_t5.py:38

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:54

Method__init__

(self, config, is_cross_attention)

code/lavis/models/blip2_models/Qformer.py:112

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:279

Method__init__

(self, config, is_cross_attention=False)

code/lavis/models/blip2_models/Qformer.py:293

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:350

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:365

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:488

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:593

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:608

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:625

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:645

Method__init__

(self, config, add_pooling_layer=False)

code/lavis/models/blip2_models/Qformer.py:687

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:973

Method__init__

(self, config)

code/lavis/models/blip2_models/Qformer.py:1136

Method__init__

LlamaRMSNorm is equivalent to T5LayerNorm

code/lavis/models/blip2_models/modeling_llama.py:75

Method__init__

(self, dim, max_position_embeddings=2048, base=10000, device=None)

code/lavis/models/blip2_models/modeling_llama.py:95

Method__init__

(self, config: LlamaConfig)

code/lavis/models/blip2_models/modeling_llama.py:163

Method__init__

(self, config: LlamaConfig)

code/lavis/models/blip2_models/modeling_llama.py:252

Method__init__

(self, config: LlamaConfig)

code/lavis/models/blip2_models/modeling_llama.py:439

Method__init__

(self, config)

code/lavis/models/blip2_models/modeling_llama.py:617

Method__init__

(self, config)

code/lavis/models/blip2_models/modeling_llama.py:796

Method__init__

( self, img_size=224, drop_path_rate=0, use_grad_checkpoint=False, vit

code/lavis/models/blip2_models/blip2_video_opt.py:40

Method__init__

( self, img_size=224, drop_path_rate=0, use_grad_checkpoint=False, vit

code/lavis/models/blip2_models/blip2_video_cascade_opt.py:38

Method__init__

apply_lemmatizer: when set to True, postprocess predict_answers() result with lemmas.

code/lavis/models/blip2_models/blip2_opt.py:42

Method__init__

Construct a layernorm module in the T5 style. No bias and no subtraction of mean.

code/lavis/models/blip2_models/modeling_t5.py:255

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:299

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:315

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:333

Method__init__

(self, config: T5Config, has_relative_attention_bias=False)

code/lavis/models/blip2_models/modeling_t5.py:351

Method__init__

(self, config, has_relative_attention_bias=False)

code/lavis/models/blip2_models/modeling_t5.py:624

Method__init__

(self, config)

code/lavis/models/blip2_models/modeling_t5.py:660

Method__init__

(self, config, embed_tokens=None)

code/lavis/models/blip2_models/modeling_t5.py:952

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:1458

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:1659

Method__init__

(self, config: T5Config)

code/lavis/models/blip2_models/modeling_t5.py:1966

Method__init__

( self, vit_model="eva_clip_g", img_size=224, drop_path_rate=0, use_gr

code/lavis/models/blip2_models/blip2_vicuna_instruct.py:34

Method__init__

( self, vit_model="eva_clip_g", q_former_model="https://storage.googleapis.com/sfr-vis

code/lavis/models/blip2_models/mini_gpt4.py:24

Method__init__

(self, num_embeddings: int, embedding_dim: int)

code/lavis/models/blip2_models/modeling_opt.py:114

Method__init__

( self, embed_dim: int, num_heads: int, dropout: float = 0.0, is_decod

code/lavis/models/blip2_models/modeling_opt.py:140

Method__init__

(self, config: OPTConfig)

code/lavis/models/blip2_models/modeling_opt.py:309

Method__init__

(self, config: OPTConfig)

code/lavis/models/blip2_models/modeling_opt.py:526

Method__init__

(self, config)

code/lavis/models/blip2_models/modeling_opt.py:900

Method__init__

( self, vit_model="eva_clip_g", q_former_model="https://storage.googleapis.com/sfr-vis

code/lavis/models/blip2_models/video_minigpt4.py:26

Method__init__

( self, vit_model="eva_clip_g", img_size=224, drop_path_rate=0, use_gr

code/lavis/models/blip2_models/blip2_image_text_matching.py:27

Method__init__

(self, config, model_path)

code/lavis/models/pnp_vqa_models/pnp_unifiedqav2_fid.py:24

Method__init__

(self, image_question_matching_model, image_captioning_model, question_answering_model, offlo

code/lavis/models/pnp_vqa_models/pnp_vqa.py:45

Method__init__

(self, config, len_video_ft=4224)

code/lavis/models/gpt_models/gpt_dialogue.py:22

Method__init__

( self, visual_encoder, text_encoder, hidden_size, num_classes, max_txt_len=40 )

code/lavis/models/alpro_models/alpro_qa.py:31

Method__init__

( self, visual_encoder, text_encoder, vision_width=768, text_width=768

code/lavis/models/alpro_models/alpro_retrieval.py:36

Method__init__

(self, mean=None, std=None, n_frms=MAX_INT)

code/lavis/processors/alpro_processors.py:22

Method__init__

(self)

code/lavis/processors/alpro_processors.py:34

Method__init__

(self, target_size, interpolation_mode="bilinear")

code/lavis/processors/alpro_processors.py:63

Method__init__

( self, image_size=384, mean=None, std=None, min_scale=0.5, ma

code/lavis/processors/alpro_processors.py:83

Functions1,483 in github.com/bytedance/Shot2Story