MCPcopy Create free account

hub / github.com/FunAudioLLM/ThinkSound / functions

Functions638 in github.com/FunAudioLLM/ThinkSound

⨍ Functions 638◇ Types & classes 175
Functioncheckpoint
(function, *args, **kwargs)
ThinkSound/models/autoencoders.py:20
Functioncleanup
()
extract_latents.py:33
Methodclip_seq_len
(self)
ThinkSound/models/mmdit.py:524
Functioncollation_fn
(samples)
ThinkSound/data/dataset.py:966
Functioncompile
(function, *args, **kwargs)
ThinkSound/models/blocks.py:249
Methodcompile
(self)
data_utils/v2a_utils/feature_utils_224_audio.py:91
Methodcompute_latent_stats
(self)
ThinkSound/data/dataset.py:595
Functioncond_model_fn
(x, sigma, **kwargs)
ThinkSound/inference/sampling.py:97
Methodconfigure_optimizers
(self)
ThinkSound/training/diffusion.py:128
Methodconfigure_optimizers
(self)
ThinkSound/training/autoencoders.py:224
Functioncopy_state_dict
Load state_dict to model, but only for keys that match exactly. Args: model (nn.Module): model to load state_dict. state_dict (Or
ThinkSound/training/utils.py:62
Functioncreate_causal_mask
(i, j, device)
ThinkSound/models/transformer.py:33
Functioncreate_dataloader_from_config
(dataset_config, batch_size, sample_size, sample_rate, audio_channels=2, num_workers=4)
ThinkSound/data/dataset.py:1093
Methodcreate_dataset
(configs, random_crop)
ThinkSound/data/datamodule.py:139
Functioncreate_diffusion_infill_from_config
(config: tp.Dict[str, tp.Any])
ThinkSound/models/diffusion.py:778
Functioncreate_model_from_config_path
(model_config_path)
ThinkSound/models/factory.py:29
Methoddecode
(self, latents, steps=100)
ThinkSound/models/autoencoders.py:589
Methoddecode
(self, z, **kwargs)
ThinkSound/models/pretransforms.py:63
Methoddecode
(self, z)
ThinkSound/models/pretransforms.py:108
Methoddecode
(self, x)
ThinkSound/models/pretransforms.py:127
Methoddecode
(self, z)
ThinkSound/models/pretransforms.py:177
Methoddecode
(self, z)
ThinkSound/models/pretransforms.py:240
Methoddecode
(self, x)
ThinkSound/models/bottleneck.py:48
Methoddecode
(self, x)
ThinkSound/models/bottleneck.py:79
Methoddecode
(self, x)
ThinkSound/models/bottleneck.py:120
Methoddecode
(self, x)
ThinkSound/models/bottleneck.py:143
Methoddecode_tokens
Decode discrete tokens to audio Only works with discrete autoencoders
ThinkSound/models/autoencoders.py:364
Methoddecode_tokens
(self, tokens, **kwargs)
ThinkSound/models/pretransforms.py:84
Methoddecode_tokens
(self, tokens)
ThinkSound/models/pretransforms.py:190
Methoddecode_tokens
(self, tokens)
ThinkSound/models/pretransforms.py:256
Methoddecode_tokens
(self, codes, **kwargs)
ThinkSound/models/bottleneck.py:30
Methoddecode_tokens
(self, codes, **kwargs)
ThinkSound/models/bottleneck.py:170
Methoddecode_tokens
(self, codes, **kwargs)
ThinkSound/models/bottleneck.py:203
Methoddecode_tokens
(self, codes, **kwargs)
ThinkSound/models/bottleneck.py:256
Methoddecode_tokens
(self, codes, **kwargs)
ThinkSound/models/bottleneck.py:308
Methoddecode_tokens
(self, tokens, **kwargs)
ThinkSound/models/bottleneck.py:352
Methoddevice
(self)
data_utils/v2a_utils/feature_utils_224.py:177
Methoddevice
(self)
data_utils/v2a_utils/feature_utils_224_audio.py:128
Methoddtype
(self)
data_utils/v2a_utils/feature_utils_224.py:181
Methoddtype
(self)
data_utils/v2a_utils/feature_utils_224_audio.py:132
Methodencode
(self, x, **kwargs)
ThinkSound/models/pretransforms.py:50
Methodencode
(self, x)
ThinkSound/models/pretransforms.py:105
Methodencode
(self, x)
ThinkSound/models/pretransforms.py:119
Methodencode
(self, x)
ThinkSound/models/pretransforms.py:162
Methodencode
(self, x)
ThinkSound/models/pretransforms.py:223
Methodencode
(self, x, return_info=False)
ThinkSound/models/bottleneck.py:38
Methodencode
(self, x, return_info=False, **kwargs)
ThinkSound/models/bottleneck.py:65
Methodencode
(self, x, return_info=False)
ThinkSound/models/bottleneck.py:104
Methodencode
(self, x, return_info=False)
ThinkSound/models/bottleneck.py:133
Methodencode
(self, x, return_info=False, **kwargs)
ThinkSound/models/bottleneck.py:152
Methodencode
(self, x, return_info=False)
ThinkSound/models/bottleneck.py:181
Methodencode
(self, x, return_info=False, **kwargs)
ThinkSound/models/bottleneck.py:216
Methodencode
(self, x, return_info=False, n_quantizers: int = None)
ThinkSound/models/bottleneck.py:268
Methodencode
(self, x, return_info=False)
ThinkSound/models/bottleneck.py:321
Methodencode_audio
(self, x)
data_utils/v2a_utils/feature_utils_224_audio.py:123
Methodencode_t5_text
(self, text: list[str])
data_utils/v2a_utils/feature_utils_224_audio.py:111
Methodencode_text
(self, text: list[str])
data_utils/v2a_utils/feature_utils_224_audio.py:103
Functionerror_avoidance_collate
(batch)
extract_latents.py:17
Functionerror_avoidance_collate
(batch)
data_utils/extract_training_audio.py:27
Functionerror_avoidance_collate
(batch)
data_utils/extract_training_video.py:27
Methodexport_model
(self, path, use_safetensors=False)
ThinkSound/training/diffusion.py:453
Methodexport_model
(self, path, use_safetensors=False)
ThinkSound/training/autoencoders.py:381
Methodextra_repr
(self)
ThinkSound/models/blocks.py:217
Methodextra_repr
(self)
ThinkSound/models/blocks.py:295
Methodforward
(self, info, *args, **kwargs)
ThinkSound/training/losses/losses.py:13
Methodforward
(self, info)
ThinkSound/training/losses/losses.py:22
Methodforward
(self, info)
ThinkSound/training/losses/losses.py:34
Methodforward
(self, info)
ThinkSound/training/losses/losses.py:53
Methodforward
(self, info)
ThinkSound/training/losses/losses.py:79
Methodforward
(self, info)
ThinkSound/training/losses/losses.py:90
Methodforward
Calculate forward propagation. Args: x (Tensor): Predicted signal (B, #channels, #samples). Returns: Tensor:
ThinkSound/training/losses/auraloss.py:50
Methodforward
Calculate forward propagation. Args: input (Tensor): Predicted signal (B, #channels, #samples). target (Tensor): Groun
ThinkSound/training/losses/auraloss.py:155
Methodforward
(self, x_mag, y_mag)
ThinkSound/training/losses/auraloss.py:180
Methodforward
(self, x_mag, y_mag)
ThinkSound/training/losses/auraloss.py:219
Methodforward
(self, input: torch.Tensor, target: torch.Tensor)
ThinkSound/training/losses/auraloss.py:395
Methodforward
(self, x, y)
ThinkSound/training/losses/auraloss.py:517
Methodforward
This loss function assumes batched input of stereo audio in the time domain. Args: input (torch.Tensor): Input tensor with shape
ThinkSound/training/losses/auraloss.py:587
Methodforward
This loss function assumes batched input of stereo audio in the time domain. Args: input (torch.Tensor): Input tensor with shape
ThinkSound/training/losses/auraloss.py:662
Methodforward
latent: (B, N, C) vf: (B, T, C_V) t: (B,)
ThinkSound/models/mmdit.py:380
Methodforward
(self, input)
ThinkSound/models/blocks.py:19
Methodforward
(self, input)
ThinkSound/models/blocks.py:58
Methodforward
(self, input)
ThinkSound/models/blocks.py:81
Methodforward
(self, input)
ThinkSound/models/blocks.py:91
Methodforward
(self, x)
ThinkSound/models/blocks.py:120
Methodforward
(self, x)
ThinkSound/models/blocks.py:142
Methodforward
(self, x, cond)
ThinkSound/models/blocks.py:220
Methodforward
(self, x)
ThinkSound/models/blocks.py:234
Methodforward
(self, x)
ThinkSound/models/blocks.py:281
Methodforward
(self, x)
ThinkSound/models/blocks.py:298
Methodforward
(self, x)
ThinkSound/models/blocks.py:331
Methodforward
(self, x: torch.Tensor)
ThinkSound/models/blocks.py:343
Methodforward
(self, x)
ThinkSound/models/blocks.py:429
Methodforward
(self, x, mask = None, cond = None, cross_attn_cond = None, cross_attn_cond_mask = None, prepend_cond = None)
ThinkSound/models/local_attention.py:69
Methodforward
(self, x)
ThinkSound/models/local_attention.py:130
Methodforward
(self, x)
ThinkSound/models/local_attention.py:172
Methodforward
(self, x)
ThinkSound/models/local_attention.py:225
Methodforward
(self, x)
ThinkSound/models/local_attention.py:272
Methodforward
(self, x, t, **kwargs)
ThinkSound/models/diffusion.py:39
Methodforward
(self, x, t, **kwargs)
ThinkSound/models/diffusion.py:65
Methodforward
(self, x: torch.Tensor, t: torch.Tensor, cross_attn_cond: torc
ThinkSound/models/diffusion.py:82
← previousnext →401–500 of 638, ranked by callers