MCPcopy Create free account

hub / github.com/ThisisBillhe/NAR / functions

Functions683 in github.com/ThisisBillhe/NAR

⨍ Functions 683◇ Types & classes 116
↓ 2 callersMethodget_emb
(self)
NAR-videos/models/larp_tokenizer.py:493
↓ 2 callersMethodget_feature_stats_for_dataset
( self, dataset: Dataset, bs=32, cache_stats=True,
NAR-videos/utils/fvd/fvd.py:368
↓ 2 callersMethodget_feature_stats_for_dataset
( self, dataset: Dataset, bs=32, cache_stats=True,
NAR-videos/utils/fid/fid.py:192
↓ 2 callersMethodinitialize
(self, input)
NAR-images/tokenizer/tokenizer_image/discriminator.py:91
↓ 2 callersMethodinitialize
(self, input)
NAR-images/tokenizer/tokenizer_image/discriminator_patchgan.py:82
↓ 2 callersFunctionload_feature_stats_from_multiple_files
(files)
NAR-videos/eval/calc_fvd_from_multiple_feature_stats.py:18
↓ 2 callersMethodmanifold_radii
(self, features: np.ndarray)
NAR-images/evaluations/c2i/evaluator.py:264
↓ 2 callersFunctionmd5_hash
(path)
NAR-images/tokenizer/tokenizer_image/lpips.py:36
↓ 2 callersFunctionnormalize_tensor
(x,eps=1e-10)
NAR-images/tokenizer/tokenizer_image/lpips.py:158
↓ 2 callersMethodpairwise_distances
Evaluate pairwise distances between two batches of feature vectors.
NAR-images/evaluations/c2i/evaluator.py:409
↓ 2 callersFunctionparse_args
(input_args=None)
NAR-videos/sample_one.py:44
↓ 2 callersFunctionparse_args
(input_args=None)
NAR-videos/sample.py:44
↓ 2 callersFunctionprecompute_freqs_cis_2d
(grid_size: int, n_elem: int, base: int = 10000, cls_token_num=120)
NAR-images/autoregressive/models/gpt.py:468
↓ 2 callersMethodproject_out
(self, z_cat)
NAR-videos/models/bottleneck.py:162
↓ 2 callersMethodread_activations
(self, npz_path: str)
NAR-images/evaluations/c2i/evaluator.py:154
↓ 2 callersMethodread_statistics
( self, npz_path: str, activations: Tuple[np.ndarray, np.ndarray] )
NAR-images/evaluations/c2i/evaluator.py:180
↓ 2 callersFunctionrequires_grad
Set requires_grad flag for all parameters in a model.
NAR-images/utils/ema.py:17
↓ 2 callersFunctionresize_video
(video)
NAR-videos/sample_one.py:325
↓ 2 callersFunctionresize_video
(video)
NAR-videos/sample.py:336
↓ 2 callersFunctionsample
(logits, temperature: float=1.0, top_k: int=0, top_p: float=1.0, sample_logits=True)
NAR-images/autoregressive/models/generate.py:59
↓ 2 callersFunctionsample
(all_logits, temperature: float=1.0, top_k: int=0, top_p: float=1.0, sample_logits=True)
NAR-videos/ar/generate.py:57
↓ 2 callersMethodsave_checkpoint
(self, filename, save_best=False, model_sd_only=False)
NAR-videos/trainers/base_trainer.py:794
↓ 2 callersMethodset_perceptual_eval
(self)
NAR-videos/models/loss.py:305
↓ 2 callersMethodsetup_proximity_mask
(self, mask, block_size)
NAR-images/autoregressive/models/gpt.py:324
↓ 2 callersMethodsync_ave_scalars_
(self, ave_scalars)
NAR-videos/trainers/base_trainer.py:597
↓ 2 callersMethodto_feature_stats
Convert input to FeatureStats object. x: FeatureStats, Dataset, or path-like object. If x is a path-like object, it can be a
NAR-videos/utils/fid/fid.py:238
↓ 2 callersFunctiontop_k_top_p_filtering
Filter a distribution of logits using top-k and/or nucleus (top-p) filtering Args: logits: logits distribution shape (batch size, vocabula
NAR-videos/ar/generate.py:16
↓ 2 callersFunctiontrace_sqrt_product
(sigma, sigma_v)
NAR-videos/utils/fvd/fvd.py:30
↓ 2 callersMethodtrainable_modules
(self)
NAR-videos/models/loss.py:314
↓ 2 callersMethodtrainable_requires_grad_
(self, requires_grad)
NAR-videos/models/loss.py:310
↓ 2 callersMethodunpatchify
x: (b, n, t_patch_size * s_patch_size**2 * c) videos: (b, c, t, h, w)
NAR-videos/models/larp_tokenizer.py:435
↓ 1 callersMethod__init__
(self, n_e, e_dim, beta)
NAR-images/tokenizer/vqgan/quantize.py:25
↓ 1 callersMethod__init__
(self, num_features, logdet=False, affine=True, allow_reverse_init=False)
NAR-images/tokenizer/tokenizer_image/discriminator_patchgan.py:71
↓ 1 callersMethod__init__
( self, capture_all=False, capture_mean_cov=False, max_items=None, onl
NAR-videos/utils/fvd/fvd.py:47
↓ 1 callersMethod__init__
( self, hidden_size, n_heads, n_layers, input_size, temporal_p
NAR-videos/models/loss.py:120
↓ 1 callersMethod__init__
(self, hidden_size, temporal_patch_size, patch_size, out_channels)
NAR-videos/models/larp_tokenizer.py:31
↓ 1 callersMethod__init__
(self, dim, depth, n_head, head_dim, ff_dim=None, dropout=0.0)
NAR-videos/models/transformer.py:10
↓ 1 callersFunction_create_feature_graph
(input_batch)
NAR-images/evaluations/c2i/evaluator.py:602
↓ 1 callersFunction_create_softmax_graph
(input_batch)
NAR-images/evaluations/c2i/evaluator.py:619
↓ 1 callersMethod_norm
(self, x)
NAR-images/autoregressive/models/gpt.py:145
↓ 1 callersMethod_norm
(self, x)
NAR-videos/models/norm.py:13
↓ 1 callersFunction_numpy_partition
(arr, kth, **kwargs)
NAR-images/evaluations/c2i/evaluator.py:652
↓ 1 callersFunction_open_npy_file
(path: str, arr_name: str)
NAR-images/evaluations/c2i/evaluator.py:580
↓ 1 callersFunction_read_bytes
Copied from: https://github.com/numpy/numpy/blob/fb215c76967739268de71aa4bda55dd1b062bc2e/numpy/lib/format.py#L788-L886 Read from file-like
NAR-images/evaluations/c2i/evaluator.py:550
↓ 1 callersFunction_update_shapes
(pool3)
NAR-images/evaluations/c2i/evaluator.py:633
↓ 1 callersMethodadjust_learning_rate_stepwise
(self)
NAR-videos/trainers/base_trainer.py:557
↓ 1 callersMethodapply_lr_multiplier
(self, lr_mult)
NAR-videos/trainers/base_trainer.py:538
↓ 1 callersFunctionapply_spectral_norm
(module: nn.Module)
NAR-videos/models/loss.py:59
↓ 1 callersMethodbasic_clean
(text)
NAR-images/language/t5.py:91
↓ 1 callersMethodbefore_save_checkpoint
(self, checkpoint)
NAR-videos/trainers/base_trainer.py:904
↓ 1 callersFunctionbuild_coco
(args, transform)
NAR-images/dataset/coco.py:26
↓ 1 callersFunctionbuild_imagenet
(args, transform)
NAR-images/dataset/imagenet.py:53
↓ 1 callersFunctionbuild_imagenet_code
(args)
NAR-images/dataset/imagenet.py:56
↓ 1 callersFunctionbuild_openimage
(args, transform)
NAR-images/dataset/openimage.py:41
↓ 1 callersFunctionbuild_pexels
(args, transform)
NAR-images/dataset/pexels.py:3
↓ 1 callersFunctionbuild_t2i
(args, transform)
NAR-images/dataset/t2i.py:148
↓ 1 callersFunctionbuild_t2i_code
(args)
NAR-images/dataset/t2i.py:151
↓ 1 callersFunctionbuild_t2i_image
(args, transform)
NAR-images/dataset/t2i.py:145
↓ 1 callersFunctionbuild_tree
(tree_list)
NAR-videos/train.py:92
↓ 1 callersFunctioncalc_dataset_md5
(dataset)
NAR-videos/utils/fvd/fvd.py:36
↓ 1 callersFunctioncalc_dataset_md5
(dataset)
NAR-videos/utils/fid/fid.py:27
↓ 1 callersMethodcalculate_adaptive_weight
(self, nll_loss, g_loss, last_layer)
NAR-images/tokenizer/tokenizer_image/vq_loss.py:109
↓ 1 callersMethodcalculate_fid_smart
( self, gen: Union[FeatureStats, Dataset, os.PathLike, str], real: Union[FeatureStats,
NAR-videos/utils/fid/fid.py:273
↓ 1 callersMethodcalculate_fvd_with_dataset
(self, feats_gen, dataset_real, bs=32, cache_stats=True)
NAR-videos/utils/fvd/fvd.py:436
↓ 1 callersMethodcalculate_prior_loss_with_pred
(self, encode_output, **kwargs)
NAR-videos/models/larp_tokenizer.py:498
↓ 1 callersFunctioncenter_crop_arr
Center cropping implementation from ADM. https://github.com/openai/guided-diffusion/blob/8fb3ad9197f16bbc40620447b2742e13458d2831/guided_diff
NAR-images/tokenizer/vae/reconstruction_vae_ddp.py:60
↓ 1 callersFunctioncenter_crop_arr
Center cropping implementation from ADM. https://github.com/openai/guided-diffusion/blob/8fb3ad9197f16bbc40620447b2742e13458d2831/guided_diff
NAR-images/tokenizer/vqgan/reconstruction_vqgan_ddp.py:62
↓ 1 callersFunctioncenter_crop_arr
Center cropping implementation from ADM. https://github.com/openai/guided-diffusion/blob/8fb3ad9197f16bbc40620447b2742e13458d2831/guided_diff
NAR-images/tokenizer/validation/val_ddp.py:55
↓ 1 callersFunctioncenter_crop_arr
Center cropping implementation from ADM. https://github.com/openai/guided-diffusion/blob/8fb3ad9197f16bbc40620447b2742e13458d2831/guided_diff
NAR-images/tokenizer/consistencydecoder/reconstruction_cd_ddp.py:60
↓ 1 callersFunctioncheck_image
(image_path)
NAR-images/tools/openimage_json.py:11
↓ 1 callersFunctioncompute_clip_score
( dataset: DataLoader, clip_model="ViT-B/32", device="cuda", how_many=5000)
NAR-images/evaluations/t2i/evaluation.py:130
↓ 1 callersFunctioncompute_entropy_loss
(affinity, loss_type="softmax", temperature=0.01)
NAR-images/tokenizer/tokenizer_image/vq_model.py:399
↓ 1 callersFunctioncompute_fid
(fake_dir: Path, gt_dir: Path, resize_size=None, feature_extractor="clip")
NAR-images/evaluations/t2i/evaluation.py:180
↓ 1 callersMethodcompute_inception_score
(self, activations: np.ndarray, split_size: int = 5000)
NAR-images/evaluations/c2i/evaluator.py:195
↓ 1 callersMethodcompute_prec_recall
( self, activations_ref: np.ndarray, activations_sample: np.ndarray )
NAR-images/evaluations/c2i/evaluator.py:210
↓ 1 callersMethodcompute_statistics
(self, activations: np.ndarray)
NAR-images/evaluations/c2i/evaluator.py:190
↓ 1 callersFunctioncount_samples_in_tar
Count the number of files (samples) in a tar file.
NAR-images/scripts/analyze_tar.py:14
↓ 1 callersFunctioncreat_optimizer_by_name
(model, weight_decay, learning_rate, betas, global_rank, logger)
NAR-images/autoregressive/train/train_c2i_fsdp.py:69
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/tokenizer/vae/reconstruction_vae_ddp.py:41
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/tokenizer/vqgan/reconstruction_vqgan_ddp.py:43
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/tokenizer/validation/val_ddp.py:36
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/tokenizer/tokenizer_image/reconstruction_vq_ddp.py:25
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/tokenizer/consistencydecoder/reconstruction_cd_ddp.py:41
↓ 1 callersFunctioncreate_npz_from_sample_folder
Builds a single .npz file from a folder of .png samples.
NAR-images/autoregressive/sample/sample_c2i_ddp.py:23
↓ 1 callersMethoddecode_eval
(self, z, num_x_tokens=None)
NAR-videos/models/larp_tokenizer.py:462
↓ 1 callersFunctiondecode_n_tokens
( model, cur_token: torch.Tensor, input_pos: torch.Tensor, cfg_scale: float, cfg_interv
NAR-images/autoregressive/models/generate.py:111
↓ 1 callersFunctiondecode_n_tokens
( model, cur_token: torch.Tensor, input_pos: torch.Tensor, num_new_tokens: int, cfg_scale: float, cfg
NAR-videos/ar/generate.py:124
↓ 1 callersFunctiondecode_one_token
( model, x: torch.Tensor, coordinate: list, input_pos: torch.Tensor, cfg_scale: float, cfg_flag: bool,
NAR-videos/ar/generate.py:95
↓ 1 callersFunctiondecode_proximity_tokens
( model, x: torch.Tensor, input_pos: torch.Tensor, cfg_scale: float, cfg_flag: bool, a
NAR-images/autoregressive/models/generate.py:87
↓ 1 callersFunctiondeserialize_to_obj
(obj_tensor)
NAR-videos/utils/common.py:157
↓ 1 callersMethoddist_all_reduce_mean_
(self, x)
NAR-videos/trainers/base_trainer.py:593
↓ 1 callersFunctiondownload
(url, local_path, chunk_size=1024)
NAR-images/tokenizer/tokenizer_image/lpips.py:24
↓ 1 callersFunctiondrop_path
Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for E
NAR-images/utils/drop_path.py:4
↓ 1 callersFunctiondrop_path
Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for E
NAR-videos/models/larp_ar.py:92
↓ 1 callersMethoddump_csv
(self, cfg)
NAR-videos/trainers/base_trainer.py:451
↓ 1 callersMethodemb_drop
(self, embs, force_drop_ids=None)
NAR-videos/models/embed.py:211
↓ 1 callersMethodencode_eval
(self, x)
NAR-videos/models/larp_tokenizer.py:424
↓ 1 callersFunctionensure_path
(path, replace=True)
NAR-videos/utils/common.py:17
↓ 1 callersFunctionentropy_loss
Calculates the entropy loss using PyTorch.
NAR-videos/models/bottleneck.py:12
↓ 1 callersMethodepoch_done
(self)
NAR-videos/utils/common.py:110
← previousnext →101–200 of 683, ranked by callers