Functions — github.com/OmkarThawakar/composed-video-retrieval

hub / github.com/OmkarThawakar/composed-video-retrieval / functions

Functions232 in github.com/OmkarThawakar/composed-video-retrieval

⨍ Functions 232◇ Types & classes 52

Method__init__

Args: img_size (int, tuple): input image size patch_size (int, tuple): patch size in_chans (int): number

src/model/vit.py:162

Method__init__

Args: med_config (str): path for the mixture of encoder-decoder model's configuration file image_size (int): input im

src/model/blip_cir.py:15

Method__init__

Args: med_config (str): path for the mixture of encoder-decoder model's configuration file image_size (int): input im

src/model/blip_embs.py:9

Method__init__

Args: med_config (str): path for the mixture of encoder-decoder model's configuration file image_size (int): input im

src/model/blip_cir_text.py:14

Method__init__

Args: med_config (str): path for the mixture of encoder-decoder model's configuration file image_size (int): input im

src/model/blip_cir_visual.py:14

Method__init__

Args: med_config (str): path for the mixture of encoder-decoder model's configuration file image_size (int): input im

(self, config, is_cross_attention)

(self, config, is_cross_attention=False)

(self, config, add_pooling_layer=True)

(self, image_size=384, min_scale=0.5)

src/data/transforms.py:12

Method__init__

(self, image_size=384)

src/data/transforms.py:48

Method__init__

( self, batch_size: int, num_workers: int = 4, pin_memory: bool = True,

src/data/webvid_covr.py:19

Method__init__

( self, batch_size: int, annotation: str, vid_dirs: str, emb_dirs: str

src/data/webvid_covr.py:103

Method__init__

(self, transform, frames_video=1, method="middle")

src/data/utils.py:137

Method__init__

( self, batch_size: int, num_workers: int = 4, pin_memory: bool = True,

src/data/webvid_covr_rulebased.py:20

Method__init__

( self, batch_size: int, num_workers: int = 4, pin_memory: bool = True,

src/data/cirr.py:16

Method__init__

( self, batch_size: int, annotation: str, img_dirs: str, emb_dirs: str

src/data/cirr.py:79

Method__init__

( self, image_dir, img_ext: str = "png", save_dir=None, )

src/data/embs.py:27

Method__init__

( self, video_dir, todo_ids=None, shard_id=0, num_shards=1, fr

src/data/embs.py:64

Method__init__

( self, csv_path, max_words=30, )

src/data/embs.py:127

Method__init__

(self, N=2, M=10, isPIL=False, augs=[])

src/data/randaugment.py:321

Method__init__

( self, batch_size: int, num_workers: int = 4, pin_memory: bool = True,

src/data/fashioniq.py:16

Method__init__

( self, batch_size: int, annotation: str, targets: str, img_dirs: str,

src/data/fashioniq.py:77

Method__init__

(self, remove_self_similarity=True)

src/test/webvid_covr_exp.py:14

Method__init__

(self)

src/test/webvid_covr_exp.py:108

Method__init__

(self, remove_self_similarity=True)

src/test/webvid_covr.py:14

(self, category: str)

src/test/fashioniq.py:17

Method__len__

(self)

src/data/webvid_covr.py:297

Method__len__

(self)

src/data/webvid_covr_rulebased.py:224

src/data/fashioniq.py:181

Initialize the weights

src/model/med.py:623

Method_prune_heads

Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base class PreTrainedModel

src/model/med.py:664

Method_reorder_cache

(self, past, beam_idx)

src/model/med.py:1067

Functionautocontrast_func

same output as PIL.ImageOps.autocontrast

src/data/randaugment.py:12

Functionblip_cir

(model, ckpt_path, **kwargs)

src/model/blip_cir.py:162

Functionblip_cir_text

(model, ckpt_path, **kwargs)

src/model/blip_cir_text.py:99

Functionblip_cir_visual

(model, ckpt_path, **kwargs)

src/model/blip_cir_visual.py:84

Functionblip_decoder

(pretrained="", **kwargs)

src/model/blip.py:217

Functionblip_feature_extractor

(pretrained="", **kwargs)

src/model/blip.py:225

Functionbrightness_func

same output as PIL.ImageEnhance.Contrast

src/data/randaugment.py:123

Functioncolor_func

same output as PIL.ImageEnhance.Color

src/data/randaugment.py:90

Functioncontrast_func

same output as PIL.ImageEnhance.Contrast

src/data/randaugment.py:109

Methodcreate_custom_forward

(img, pad_size, replace=(0, 0, 0))

src/data/randaugment.py:203

Functioncutout_level_to_args

(cutout_const, MAX_LEVEL, replace_value)

src/data/randaugment.py:244

Functionequalize_func

same output as PIL.ImageOps.equalize PIL's implementation is different from cv2.equalize

src/data/randaugment.py:45

Functionevaluate

(model, data_loader, fabric)

src/test/utils.py:11

Methodfeed_forward_chunk

(self, attention_output)

src/model/med.py:442

Methodforward

(self, tar_img_feat: torch.Tensor, query_feat: torch.Tensor, temp)

src/model/loss.py:14

Methodforward

Args: video_embds: (batch_size, video_embd_dim) text_embds: (batch_size, text_embd_dim)

(self, x, register_hook=False)

src/model/vit.py:83

Methodforward

(self, x, register_hook=False)

src/model/vit.py:150

Methodforward

(self, x, register_blk=-1)

src/model/vit.py:260

Methodforward

(self, batch, fabric)

src/model/blip_cir.py:60

Methodforward

(self, batch, fabric)

src/model/blip_cir_text.py:61

Methodforward

(self, batch, fabric)

src/model/blip_cir_visual.py:60

Methodforward

(self, image, caption, mode)

src/model/blip.py:50

Methodforward

(self, image, caption)

src/model/blip.py:120

Methodforward

( self, input_ids=None, position_ids=None, inputs_embeds=None, past_ke

src/model/med.py:74

Methodforward

( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi

src/model/med.py:164

Methodforward

(self, hidden_states, input_tensor)

src/model/med.py:279

Methodforward

( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi

src/model/med.py:316

Methodforward

(self, hidden_states)

src/model/med.py:351

Methodforward

(self, hidden_states, input_tensor)

src/model/med.py:364

Methodforward

( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi

src/model/med.py:386

Methodforward

( self, hidden_states, attention_mask=None, head_mask=None, encoder_hi

src/model/med.py:457

Methodforward

(self, hidden_states)

src/model/med.py:557

Methodforward

(self, hidden_states)

src/model/med.py:576

Methodforward

(self, hidden_states)

src/model/med.py:597

Methodforward

(self, sequence_output)

src/model/med.py:608

Methodforward

r""" encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`): Seq

src/model/med.py:750

Methodforward

r""" encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`): Seq

src/model/med.py:939

Methodgenerate

( self, image, sample=False, num_beams=3, max_length=30, min_l

src/model/blip.py:153

Methodget_attention_map

(self)

src/model/vit.py:80

Methodget_attention_map

(self)

src/model/med.py:153

Methodget_attn_gradients

(self)

src/model/vit.py:74

Methodget_attn_gradients

(self)

src/model/med.py:147

Methodget_input_embeddings

(self)

src/model/med.py:658

Functionget_middle_frame

(reference_vid_pth)

src/data/utils.py:40

Methodget_output_embeddings

(self)

src/model/med.py:933

Functionget_random_frame

(reference_vid_pth)

src/data/utils.py:81

Functionidentity_func

(img)

src/data/randaugment.py:8

← previous next →101–200 of 232, ranked by callers