MCPcopy Create free account

hub / github.com/ThisisBillhe/NAR / functions

Functions683 in github.com/ThisisBillhe/NAR

⨍ Functions 683◇ Types & classes 116
↓ 186 callersFunctionprint
(*args, **kwargs)
NAR-images/utils/distributed.py:13
↓ 120 callersMethodappend
(self, x)
NAR-videos/utils/fvd/fvd.py:91
↓ 65 callersMethoditem
(self)
NAR-videos/utils/common.py:99
↓ 39 callersMethodsave
(self, pkl_file)
NAR-videos/utils/fvd/fvd.py:151
↓ 37 callersMethodload
(cls, path: str, arr_name: str)
NAR-images/evaluations/c2i/evaluator.py:515
↓ 22 callersMethodload
(stats_path)
NAR-videos/utils/fvd/fvd.py:156
↓ 21 callersMethodupdate
(self, input_pos, k_val, v_val)
NAR-videos/models/larp_ar.py:157
↓ 19 callersMethodget_feature_stats_for_batch
(self, batch, feats=None, num_gpus=None)
NAR-videos/utils/fvd/fvd.py:339
↓ 15 callersFunctionget_1d_sincos_pos_embed_from_grid
embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D) scale_factor: the base fo
NAR-videos/models/embed.py:312
↓ 14 callersMethodfrom_pretrained
(cls, name="vgg_lpips")
NAR-images/tokenizer/tokenizer_image/lpips.py:75
↓ 14 callersMethodsample
( self, c, cfg_scale=2.0, cfg_interval=-1, temperature=1.0, t
NAR-videos/models/larp_ar.py:527
↓ 12 callersMethodupdate
(self, input_pos, k_val, v_val)
NAR-images/autoregressive/models/gpt.py:179
↓ 11 callersMethodadd
(self, v, n=1.0)
NAR-videos/utils/common.py:95
↓ 10 callersMethodencode
(self, x)
NAR-videos/models/larp_tokenizer.py:415
↓ 9 callersMethodencode
(self, x)
NAR-images/tokenizer/vqgan/model.py:69
↓ 9 callersMethodtrain
For epochs perform training, evaluation, and visualization. Note that ave_scalars update ignores the actual current batch_siz
NAR-videos/trainers/base_trainer.py:484
↓ 8 callersMethod__init__
(self, in_features, hidden_features, out_features)
NAR-images/autoregressive/models/gpt.py:121
↓ 8 callersMethodcalculate_fvd
(self, feats_gen, feats_real)
NAR-videos/utils/fvd/fvd.py:419
↓ 8 callersFunctioncreate_logger
Create a logger that writes to a log file and stdout.
NAR-images/utils/logger.py:4
↓ 8 callersMethodlog_temp_scalar
(self, k, v, t=None)
NAR-videos/trainers/base_trainer.py:585
↓ 7 callersMethod__init__
(self, config: ModelArgs)
NAR-images/tokenizer/tokenizer_image/vq_model.py:29
↓ 7 callersMethoddecode_code
(self, code_b, shape, channel_first=True)
NAR-images/tokenizer/vqgan/model.py:80
↓ 7 callersMethodfrom_checkpoint
(cls, ckpt, args, load_state_dict=True)
NAR-videos/models/larp_ar.py:549
↓ 6 callersMethod__init__
(self, in_features, hidden_features, out_features)
NAR-videos/models/larp_ar.py:72
↓ 6 callersFunction_assert
(cond, msg)
NAR-videos/models/embed.py:118
↓ 6 callersFunctionbuild_dataset
(args, **kwargs)
NAR-images/dataset/build.py:8
↓ 6 callersMethoddecode
(self, quant)
NAR-images/tokenizer/vqgan/model.py:75
↓ 6 callersMethoddecode_from_bottleneck
(self, bottleneck_rep)
NAR-videos/models/larp_tokenizer.py:475
↓ 6 callersMethoddevice
(self)
NAR-videos/models/larp_ar.py:313
↓ 6 callersMethodget_mean_cov
(self)
NAR-videos/utils/fvd/fvd.py:132
↓ 5 callersFunctionNormalize
(in_channels)
NAR-images/tokenizer/vqgan/layer.py:13
↓ 5 callersFunctionNormalize
(in_channels, norm_type='group')
NAR-images/tokenizer/tokenizer_image/vq_model.py:359
↓ 5 callersMethod__init__
(self, *, ch, out_ch, ch_mult=(1,2,4,8), num_res_blocks, attn_resolutions, dropout=0.0, resam
NAR-images/tokenizer/vqgan/layer.py:176
↓ 5 callersMethod__init__
( self, spatial_vid_size: Optional[int] = 224, temporal_vid_size: Optional
NAR-videos/models/embed.py:43
↓ 5 callersFunctioncenter_crop_arr
Center cropping implementation from ADM. https://github.com/openai/guided-diffusion/blob/8fb3ad9197f16bbc40620447b2742e13458d2831/guided_diff
NAR-images/dataset/augmentation.py:8
↓ 5 callersFunctionget_orig_module
(module)
NAR-videos/trainers/base_trainer.py:59
↓ 5 callersFunctioninit_distributed_mode
(args)
NAR-images/utils/distributed.py:20
↓ 5 callersFunctionleaky_relu
(p=0.2)
NAR-images/tokenizer/tokenizer_image/discriminator.py:250
↓ 5 callersFunctionleaky_relu
(p=0.2)
NAR-images/tokenizer/tokenizer_image/discriminator_stylegan.py:96
↓ 5 callersFunctionnonlinearity
(x)
NAR-images/tokenizer/vqgan/layer.py:8
↓ 5 callersMethodrun
(self)
NAR-videos/trainers/base_trainer.py:233
↓ 5 callersMethodsetup_diagonal_mask
(self, mask, cond_len)
NAR-videos/models/larp_ar.py:330
↓ 4 callersFunctionVideoTransform
(crop_size=128, scale=1.00, ratio=1.00, eval_tfm=False, rand_flip='no')
NAR-videos/datasets/video_dataset.py:42
↓ 4 callersMethod__init__
(self)
NAR-images/tokenizer/tokenizer_image/discriminator.py:239
↓ 4 callersMethod__init__
Build pretrained InceptionV3 Parameters ---------- output_blocks : list of int Indices of blocks to return featur
NAR-videos/utils/fid/inception.py:32
↓ 4 callersMethoddummy_data
(self)
NAR-images/dataset/t2i.py:88
↓ 4 callersFunctiongenerate
(model, cond, max_new_tokens, emb_masks=None, cfg_scale=1.0, cfg_interval=-1, **sampling_kwargs)
NAR-images/autoregressive/models/generate.py:166
↓ 4 callersMethodget_text_embeddings
(self, texts)
NAR-images/language/t5.py:58
↓ 4 callersFunctionnonlinearity
(x)
NAR-images/tokenizer/tokenizer_image/vq_model.py:354
↓ 4 callersMethodreset_caches
(self)
NAR-videos/models/larp_ar.py:366
↓ 4 callersFunctionupdate_ema
Step the EMA model towards the current model.
NAR-images/utils/ema.py:5
↓ 4 callersMethodupdate_ema
Step the EMA model towards the current model.
NAR-videos/trainers/base_trainer.py:783
↓ 3 callersMethod__init__
(self, use_dropout=True)
NAR-images/tokenizer/tokenizer_image/lpips.py:55
↓ 3 callersMethod__init__
( self, bottleneck_dim: int, input_dim: int, output_dim: int, token_nu
NAR-videos/models/bottleneck.py:67
↓ 3 callersMethoddecode
(self, z)
NAR-videos/models/larp_tokenizer.py:450
↓ 3 callersMethodencode
(self, x)
NAR-images/tokenizer/tokenizer_image/vq_model.py:41
↓ 3 callersFunctionget_3d_sincos_pos_embed
(embed_dim, grid_size, frame_num)
NAR-videos/models/embed.py:269
↓ 3 callersFunctionget_model_cls
(name)
NAR-videos/models/__init__.py:10
↓ 3 callersMethodinit_from_ckpt
(self, path, ignore_keys=list(), logging=True)
NAR-images/tokenizer/vqgan/model.py:55
↓ 3 callersMethodmake_loss
(self, loss_spec=None, load_sd=False)
NAR-videos/trainers/base_trainer.py:434
↓ 3 callersMethodmake_model
(self, model_spec=None, load_sd=False)
NAR-videos/trainers/base_trainer.py:355
↓ 3 callersMethodset_vq_eval_deterministic
(self, deterministic=True)
NAR-videos/models/larp_tokenizer.py:327
↓ 3 callersFunctiontime_text
(secs)
NAR-videos/utils/common.py:120
↓ 3 callersMethodtrainable_parameters
(self)
NAR-videos/models/loss.py:317
↓ 2 callersMethod__init__
(self)
NAR-images/tokenizer/tokenizer_image/discriminator_stylegan.py:85
↓ 2 callersMethod__init__
(self, config: GPTCConfig)
NAR-videos/models/gptc.py:108
↓ 2 callersFunction_batch_pairwise_distances
Compute pairwise distances between two batches of feature vectors.
NAR-images/evaluations/c2i/evaluator.py:430
↓ 2 callersFunction_download_inception_model
()
NAR-images/evaluations/c2i/evaluator.py:589
↓ 2 callersMethod_gen_vis_result
(self, tag, vislist)
NAR-videos/trainers/larp_tokenizer_trainer.py:395
↓ 2 callersFunction_inception_v3
Wraps `torchvision.models.inception_v3`
NAR-videos/utils/fid/inception.py:165
↓ 2 callersMethod_iter_step
(self, data, is_train)
NAR-videos/trainers/larp_tokenizer_trainer.py:232
↓ 2 callersMethod_iter_step
(self, data, is_train)
NAR-videos/trainers/larp_ar_fp_trainer.py:206
↓ 2 callersMethod_iter_step
(self, data, is_train)
NAR-videos/trainers/larp_ar_trainer.py:174
↓ 2 callersFunction_symmetric_matrix_square_root
(mat, eps=1e-10)
NAR-videos/utils/fvd/fvd.py:24
↓ 2 callersFunctionadopt_weight
(weight, global_step, threshold=0, value=0.)
NAR-images/tokenizer/tokenizer_image/vq_loss.py:43
↓ 2 callersFunctionadopt_weight
(weight, global_step, threshold=0, value=0.0)
NAR-videos/models/loss.py:98
↓ 2 callersMethodappend_torch
(self, x, num_gpus=1)
NAR-videos/utils/fvd/fvd.py:111
↓ 2 callersFunctionapply_rotary_emb
(x: torch.Tensor, freqs_cis: torch.Tensor)
NAR-images/autoregressive/models/gpt.py:484
↓ 2 callersMethodar_predict
(self, x: torch.Tensor)
NAR-videos/models/gptc.py:179
↓ 2 callersMethodcalculate_logits_and_ar_pred_cont
(self, prior_model_output)
NAR-videos/models/larp_tokenizer.py:538
↓ 2 callersMethodcalculate_stats_for_dataset
(self, dataset, bs=32, num_workers=4)
NAR-videos/utils/fvd/fvd.py:412
↓ 2 callersMethodcalculate_stats_for_dataset
(self, dataset: ImageDataset, bs=32, num_workers=4)
NAR-videos/utils/fid/fid.py:184
↓ 2 callersMethodclean_caption
(self, caption)
NAR-images/language/t5.py:96
↓ 2 callersMethodcompute_activations
Compute image features for downstream evals. :param batches: a iterator over NHWC numpy arrays in [0, 255]. :return: a tuple
NAR-images/evaluations/c2i/evaluator.py:158
↓ 2 callersMethodconfigure_optimizers
(self, config, load_sd=False)
NAR-videos/trainers/base_trainer.py:437
↓ 2 callersMethodconfigure_scalers
(self, sd=None, load_sd=False)
NAR-videos/trainers/base_trainer.py:440
↓ 2 callersFunctionconvert
(type, x)
NAR-videos/train.py:105
↓ 2 callersFunctioncreat_optimizer
(model, weight_decay, learning_rate, betas, logger)
NAR-images/autoregressive/train/train_c2i.py:30
↓ 2 callersMethoddecode
(self, quant)
NAR-images/tokenizer/tokenizer_image/vq_model.py:47
↓ 2 callersMethoddecode
(self, bottleneck_rep)
NAR-videos/models/bottleneck.py:166
↓ 2 callersMethoddecode_code
(self, code_b, shape=None, channel_first=True)
NAR-images/tokenizer/tokenizer_image/vq_model.py:52
↓ 2 callersMethoddecoder_parameters
(self)
NAR-videos/models/larp_tokenizer.py:338
↓ 2 callersMethodenable_wandb_if_needed
(self, wandb_run_id=None)
NAR-videos/trainers/base_trainer.py:168
↓ 2 callersMethodevaluate_pr
Evaluate precision and recall efficiently. :param features_1: [N1 x D] feature vectors for reference batch. :param radii_1:
NAR-images/evaluations/c2i/evaluator.py:341
↓ 2 callersFunctionfind_multiple
(n: int, k: int)
NAR-images/autoregressive/models/gpt.py:18
↓ 2 callersFunctionfind_multiple
(n: int, k: int)
NAR-videos/models/larp_ar.py:28
↓ 2 callersMethodforward
(self, x, targets=None)
NAR-videos/models/gptc.py:140
↓ 2 callersMethodfrechet_distance
Compute the Frechet distance between two sets of statistics.
NAR-images/evaluations/c2i/evaluator.py:88
↓ 2 callersMethodget_all
(self)
NAR-videos/utils/fvd/fvd.py:125
↓ 2 callersFunctionget_ckpt_path
(name, root, check=False)
NAR-images/tokenizer/tokenizer_image/lpips.py:42
next →1–100 of 683, ranked by callers