MCPcopy Create free account

hub / github.com/MCG-NJU/VideoMAE / functions

Functions287 in github.com/MCG-NJU/VideoMAE

Function_shear_level_to_arg
(level, _hparams)
rand_augment.py:220
Function_solarize_add_level_to_arg
(level, _hparams)
rand_augment.py:275
Function_solarize_increasing_level_to_arg
(level, _hparams)
rand_augment.py:269
Function_translate_abs_level_to_arg
(level, hparams)
rand_augment.py:227
Function_translate_rel_level_to_arg
(level, hparams)
rand_augment.py:234
Functionauto_contrast
(img, **__)
rand_augment.py:142
Functionauto_load_model
(args, model, model_without_ddp, optimizer, loss_scaler, model_ema=None)
utils.py:430
Methodavg
(self)
utils.py:58
Functionbrightness
(img, factor, **__)
rand_augment.py:187
Functionclip_boxes_to_image
Clip an array of boxes to an image with the given height and width. Args: boxes (ndarray): bounding boxes to perform clipping.
video_transforms.py:254
Functioncolor
(img, factor, **__)
rand_augment.py:183
Functioncolor_jitter
Perfrom a color jittering on the input images. The channels of images should be in order BGR. Args: images (tensor): images to pe
video_transforms.py:314
Functioncolor_normalization
Perform color nomration on the given images. Args: images (tensor): images to perform color normalization. Dimension is `
video_transforms.py:455
Functioncompute_video
(lst)
engine_for_finetuning.py:275
Functioncontrast
(img, factor, **__)
rand_augment.py:179
Functioncosine_scheduler
(base_value, final_value, epochs, niter_per_ep, warmup_epochs=0, start_warmup_value=0, wa
utils.py:385
Functioncreate_ds_config
(args)
utils.py:482
Functioncreate_random_augment
Get video randaug transform. Args: input_size: The size of the input video in tuple. auto_augment: Parameters for randaug. A
video_transforms.py:621
Functioncrop_clip
(clip, min_h, min_w, h, w)
functional.py:12
Functionequalize
(img, **__)
rand_augment.py:150
Methodextra_repr
(self)
modeling_finetune.py:31
Methodforward
(self, x)
modeling_finetune.py:28
Methodforward
(self, x)
modeling_finetune.py:45
Methodforward
(self, x)
modeling_finetune.py:79
Methodforward
(self, x)
modeling_finetune.py:124
Methodforward
(self, x, **kwargs)
modeling_finetune.py:150
Methodforward
(self, x)
modeling_finetune.py:282
Methodforward
(self, x, mask)
modeling_pretrain.py:107
Methodforward
(self, x, return_token_num)
modeling_pretrain.py:163
Methodforward
(self, x, mask)
modeling_pretrain.py:271
Methodget_classifier
(self)
modeling_finetune.py:254
Methodget_classifier
(self)
modeling_pretrain.py:81
Methodget_classifier
(self)
modeling_pretrain.py:156
Methodget_layer_id
(self, var_name)
optim_factory.py:45
Methodget_num_layers
(self)
modeling_pretrain.py:74
Methodget_num_layers
(self)
modeling_pretrain.py:149
Methodget_num_layers
(self)
modeling_pretrain.py:264
Methodget_scale
(self, layer_id)
optim_factory.py:42
Methodglobal_avg
(self)
utils.py:63
Functionhorizontal_flip
Perform horizontal flip on the given images and corresponding boxes. Args: prob (float): probility to flip the images. images
video_transforms.py:156
Functioninit_distributed_mode
(args)
utils.py:249
Functioninvert
(img, **__)
rand_augment.py:146
Functionlighting_jitter
Perform AlexNet-style PCA jitter on the given images. Args: images (tensor): images to perform lighting jitter. Dimension is
video_transforms.py:407
Functionload_state_dict
(model, state_dict, prefix='', ignore_missing="relative_position_index")
utils.py:293
Methodmedian
(self)
utils.py:53
Functionmultiple_samples_collate
Collate function for repeated augmentation. Each instance in the batch has more than one sample. Args: batch (tuple or list): dat
utils.py:513
Methodno_weight_decay
(self)
modeling_pretrain.py:78
Methodno_weight_decay
(self)
modeling_pretrain.py:153
Methodno_weight_decay
(self)
modeling_pretrain.py:268
Functionnormalize
(clip, mean, std, inplace=False)
functional.py:77
Functionposterize
(img, bits_to_keep, **__)
rand_augment.py:173
Functionpretrain_videomae_base_patch16_224
(pretrained=False, **kwargs)
modeling_pretrain.py:311
Functionpretrain_videomae_huge_patch16_224
(pretrained=False, **kwargs)
modeling_pretrain.py:359
Functionpretrain_videomae_large_patch16_224
(pretrained=False, **kwargs)
modeling_pretrain.py:335
Functionpretrain_videomae_small_patch16_224
(pretrained=False, **kwargs)
modeling_pretrain.py:287
Functionrandom_crop
Perform random spatial crop on the given images and corresponding boxes. Args: images (tensor): images to perform random crop. The di
video_transforms.py:120
Functionrandom_resized_crop
Crop the given images to random size and aspect ratio. A crop of random size (default: of 0.08 to 1.0) of the original size and a random aspe
video_transforms.py:541
Functionrandom_resized_crop_with_shift
This is similar to random_resized_crop. However, it samples two different boxes (for cropping) for the first and last frame. It then linearly
video_transforms.py:576
Functionrandom_short_side_scale_jitter
Perform a spatial short scale jittering on the given images and corresponding boxes. Args: images (tensor): images to perform sca
video_transforms.py:44
Functionrandom_sized_crop_img
Performs Inception-style cropping (used for training).
video_transforms.py:657
Methodreset_classifier
(self, num_classes, global_pool='')
modeling_finetune.py:257
Methodreset_classifier
(self, num_classes, global_pool='')
modeling_pretrain.py:84
Methodreset_classifier
(self, num_classes, global_pool='')
modeling_pretrain.py:159
Functionresize_clip
(clip, size, interpolation='bilinear')
functional.py:26
Functionrotate
(img, degrees, **kwargs)
rand_augment.py:108
Functionsave_model
(args, epoch, model, model_without_ddp, optimizer, loss_scaler, model_ema=None)
utils.py:405
Functionseed_worker
(worker_id)
utils.py:190
Functionsharpness
(img, factor, **__)
rand_augment.py:191
Functionshear_x
(img, factor, **kwargs)
rand_augment.py:64
Functionshear_y
(img, factor, **kwargs)
rand_augment.py:71
Functionsolarize
(img, thresh, **__)
rand_augment.py:154
Functionsolarize_add
(img, add, thresh=128, **__)
rand_augment.py:158
Methodsynchronize_between_processes
Warning: does not synchronize the deque!
utils.py:39
Functiontransforms_imagenet_train
If separate==True, the transforms are returned as a tuple of 3 separate transforms for use in a mixing dataset that passes * all data th
video_transforms.py:799
Functiontranslate_x_abs
(img, pixels, **kwargs)
rand_augment.py:94
Functiontranslate_x_rel
(img, pct, **kwargs)
rand_augment.py:78
Functiontranslate_y_abs
(img, pixels, **kwargs)
rand_augment.py:101
Functiontranslate_y_rel
(img, pct, **kwargs)
rand_augment.py:86
Functionuniform_crop
Perform uniform spatial sampling on the images and corresponding boxes. Args: images (tensor): images to perform uniform crop. The di
video_transforms.py:191
Methodvalue
(self)
utils.py:71
Functionvit_base_patch16_224
(pretrained=False, **kwargs)
modeling_finetune.py:298
Functionvit_base_patch16_384
(pretrained=False, **kwargs)
modeling_finetune.py:307
Functionvit_huge_patch16_224
(pretrained=False, **kwargs)
modeling_finetune.py:343
Functionvit_large_patch16_224
(pretrained=False, **kwargs)
modeling_finetune.py:316
Functionvit_large_patch16_384
(pretrained=False, **kwargs)
modeling_finetune.py:325
Functionvit_large_patch16_512
(pretrained=False, **kwargs)
modeling_finetune.py:334
Functionvit_small_patch16_224
(pretrained=False, **kwargs)
modeling_finetune.py:289
← previous201–287 of 287, ranked by callers