MCPcopy Create free account

hub / github.com/SillyTavern/SillyTavern-Extras / functions

Functions4,572 in github.com/SillyTavern/SillyTavern-Extras

⨍ Functions 4,572◇ Types & classes 788↳ Endpoints 33
↓ 8 callersMethod__init__
(self, num_features: int)
talkinghead/tha3/nn/normalization.py:38
↓ 8 callersFunctionbase_gpt3_architecture
(args)
modules/voice_conversion/fairseq/models/transformer_lm.py:523
↓ 8 callersMethodcreate_alpha_block
(self)
talkinghead/tha3/nn/common/poser_args.py:31
↓ 8 callersFunctiondata_file_path
(prefix_path)
modules/voice_conversion/fairseq/data/indexed_dataset.py:142
↓ 8 callersMethodgcmvn_denormalize
(self, x)
modules/voice_conversion/fairseq/speech_generator.py:21
↓ 8 callersFunctionget_rank
(group)
modules/voice_conversion/fairseq/distributed/utils.py:399
↓ 8 callersMethodget_waveform
(self, feat)
modules/voice_conversion/fairseq/speech_generator.py:31
↓ 8 callersMethodisfile
(path: str)
modules/voice_conversion/fairseq/file_io.py:91
↓ 8 callersMethodload
Loads centroids and assignments from a given path Args: - path: folder use to load centroids and assignments
modules/voice_conversion/fairseq/modules/quantization/pq/em.py:191
↓ 8 callersMethodmax_positions
Maximum output length supported by the encoder.
modules/voice_conversion/fairseq/models/roberta/model.py:622
↓ 8 callersMethodsize
(self, index)
modules/voice_conversion/fairseq/data/audio/speech_to_text_dataset.py:367
↓ 7 callersMethod__init__
(self, tgt_dict)
modules/voice_conversion/fairseq/search.py:20
↓ 7 callersMethod__init__
( self, input_dim, clamp_grad=True, max_grad_value=256, eps=1e-5,
modules/voice_conversion/fairseq/models/speech_to_text/modules/emformer.py:64
↓ 7 callersFunction_collate_frames
Convert a list of 2D frames into a padded 3D tensor Args: frames (list): list of 2D frames of size L[i]*f_dim. Where L[i] is
modules/voice_conversion/fairseq/data/audio/speech_to_text_dataset.py:35
↓ 7 callersFunction_lang_token_index
(dictionary, lang: str)
modules/voice_conversion/fairseq/tasks/online_backtranslation.py:667
↓ 7 callersMethodadd_meter
(self, key, meter, priority)
modules/voice_conversion/fairseq/logging/meters.py:270
↓ 7 callersMethodalter_dataset_langtok
( self, lang_pair_dataset, src_eos=None, src_lang=None, tgt_eos=None,
modules/voice_conversion/fairseq/tasks/multilingual_translation.py:178
↓ 7 callersFunctionapply_color_change
(alpha, color_change, image: Tensor)
talkinghead/tha3/nn/image_processing_util.py:57
↓ 7 callersMethodattention
(self, query, key, value, mask=None)
modules/voice_conversion/rvc/infer_pack/attentions.py:226
↓ 7 callersMethodbuild_bpe
Build the tokenizer for this task.
modules/voice_conversion/fairseq/tasks/fairseq_task.py:629
↓ 7 callersFunctionget_active_aggregators
()
modules/voice_conversion/fairseq/logging/metrics.py:107
↓ 7 callersMethodget_criterion
Get the (non-wrapped) criterion instance.
modules/voice_conversion/fairseq/trainer.py:1195
↓ 7 callersFunctionget_features_or_waveform
Get speech features from .npy file or waveform from .wav/.flac file. The file may be inside an uncompressed ZIP file and is accessed via byte
modules/voice_conversion/fairseq/data/audio/audio_utils.py:169
↓ 7 callersFunctionindex_file_path
(prefix_path)
modules/voice_conversion/fairseq/data/indexed_dataset.py:138
↓ 7 callersMethodinference_step
( self, generator, models, sample, prefix_tokens=None, constraints=None )
modules/voice_conversion/fairseq/tasks/fairseq_task.py:538
↓ 7 callersFunctionmaybe_shorten_dataset
( dataset, split, shorten_data_split_list, shorten_method, tokens_per_sample, seed, )
modules/voice_conversion/fairseq/data/shorten_dataset.py:63
↓ 7 callersFunctionnumpy_linear_to_srgb
(x)
talkinghead/tha3/util.py:21
↓ 7 callersMethodoutput_layer
(self, features, masked_tokens=None, **unused)
modules/voice_conversion/fairseq/models/roberta/model.py:619
↓ 7 callersMethodpose
(self, image: Tensor, pose: Tensor, output_index: int = 0)
talkinghead/tha3/poser/poser.py:150
↓ 7 callersMethodrun
(self)
modules/voice_conversion/fairseq/data/iterators.py:671
↓ 7 callersMethodshuffle
Shuffle dataset samples before batching
modules/voice_conversion/fairseq/data/audio/data_cfg.py:61
↓ 7 callersMethodwrite
(self, sizes, pointers)
modules/voice_conversion/fairseq/data/huffman/huffman_mmap_indexed_dataset.py:43
↓ 6 callersFunctionLinear
Linear layer (input: N x T x C)
modules/voice_conversion/fairseq/models/lstm.py:691
↓ 6 callersMethod_get_input_buffer
(self, incremental_state)
modules/voice_conversion/fairseq/modules/dynamic_convolution.py:292
↓ 6 callersFunction_skip
Getting sliced (dim=0) tensor by mask. Supporting tensor and list/dict of tensors.
modules/voice_conversion/fairseq/models/nat/levenshtein_utils.py:231
↓ 6 callersFunction_skip_encoder_out
(encoder, encoder_out, mask)
modules/voice_conversion/fairseq/models/nat/levenshtein_utils.py:256
↓ 6 callersMethodadd_datapoint
(self, data: float)
talkinghead/tha3/app/util.py:307
↓ 6 callersFunctionbase_multitask_text_transformer_decoder_arch
(args)
modules/voice_conversion/fairseq/models/speech_to_speech/s2s_transformer.py:582
↓ 6 callersFunctionbroadcast
(tensor, src, group)
modules/voice_conversion/fairseq/distributed/utils.py:507
↓ 6 callersFunctioncheck_type
(module, expected_type)
modules/voice_conversion/fairseq/models/fairseq_model.py:30
↓ 6 callersFunctioncheckpoint_wrapper
A friendlier wrapper for performing activation checkpointing. Compared to the PyTorch version, this version: - wraps an nn.Module, so th
modules/voice_conversion/fairseq/modules/checkpoint_activations.py:14
↓ 6 callersMethodcreate
(constraints: List[List[int]])
modules/voice_conversion/fairseq/token_generation_constraints.py:146
↓ 6 callersMethodcreate_color_change_block
(self)
talkinghead/tha3/nn/common/poser_args.py:53
↓ 6 callersFunctioncreate_conv3_from_block_args
(in_channels: int, out_channels: int, bias: bool = False,
talkinghead/tha3/nn/conv.py:44
↓ 6 callersMethodexit
Terminate the output encoder thread. Called automatically when the process exits.
talkinghead/tha3/app/app.py:1224
↓ 6 callersFunctionexpand_dim
(t, dim, k)
modules/voice_conversion/fairseq/modules/kmeans_attention.py:82
↓ 6 callersMethodfinalize
Sort symbols by frequency in descending order, ignoring special ones. Args: - threshold defines the minimum word count
modules/voice_conversion/fairseq/data/dictionary.py:151
↓ 6 callersFunctionfsdp_wrap
Helper to wrap layers/modules in FSDP. This falls back to a no-op if fairscale is not available. Args: module (nn.Module): modul
modules/voice_conversion/fairseq/distributed/fully_sharded_data_parallel.py:124
↓ 6 callersFunctionget_available_dataset_impl
()
modules/voice_conversion/fairseq/data/indexed_dataset.py:38
↓ 6 callersMethodget_lang_tag_idx
(cls, lang: str, dictionary: Dictionary)
modules/voice_conversion/fairseq/data/audio/speech_to_text_dataset.py:208
↓ 6 callersMethodget_normalized_probs
Get normalized probabilities (or log probs) from a net's output.
modules/voice_conversion/fairseq/models/roberta/model.py:321
↓ 6 callersFunctionget_padding
(kernel_size, dilation=1)
modules/voice_conversion/fairseq/models/text_to_speech/hifigan.py:16
↓ 6 callersFunctionget_parser
(desc, default_task="translation")
modules/voice_conversion/fairseq/options.py:228
↓ 6 callersMethodget_target_dictionary
(self, lang)
modules/voice_conversion/fairseq/data/multilingual/multilingual_data_manager.py:456
↓ 6 callersFunctionget_world_size
(group)
modules/voice_conversion/fairseq/distributed/utils.py:408
↓ 6 callersFunctionlang_token
(lang)
modules/voice_conversion/fairseq/tasks/multilingual_language_modeling.py:44
↓ 6 callersMethodmax_positions
Maximum input length supported by the encoder.
modules/voice_conversion/fairseq/models/lstm.py:344
↓ 6 callersMethodmaybe_layer_norm
(self, before: bool = False, after: bool = False)
modules/voice_conversion/fairseq/models/lightconv.py:973
↓ 6 callersFunctionmerge
(key, left_pad, move_eos_to_beginning=False, pad_to_length=None)
modules/voice_conversion/fairseq/data/denoising_dataset.py:28
↓ 6 callersMethodmodel
(self)
modules/voice_conversion/fairseq/trainer.py:254
↓ 6 callersMethodsize
Return an example's size as a float or tuple. This value is used when filtering a dataset with ``--max-positions``.
modules/voice_conversion/fairseq/data/audio/raw_audio_dataset.py:205
↓ 6 callersFunctionsplit_exists
(split, src, tgt, lang, data_path)
modules/voice_conversion/fairseq/tasks/translation.py:62
↓ 6 callersMethodstart
(self)
modules/voice_conversion/fairseq/logging/meters.py:204
↓ 6 callersMethodstep
Take a single search step. Args: step: the current search step, starting at 0 lprobs: (bsz x input_beam_size x vocab_
modules/voice_conversion/fairseq/search.py:30
↓ 6 callersMethodtokenize
(cls, tokenizer, text: str)
modules/voice_conversion/fairseq/data/audio/speech_to_text_dataset.py:187
↓ 5 callersFunctionEmbedding
(num_embeddings, embedding_dim, padding_idx)
modules/voice_conversion/fairseq/models/lstm.py:668
↓ 5 callersMethod__init__
(self, *args, **kwargs)
modules/voice_conversion/fairseq/models/fairseq_model.py:360
↓ 5 callersMethod__init__
( self, embed_dim, n_heads, hidden_dim, kernel_size, dropout, attention_dropout )
modules/voice_conversion/fairseq/models/text_to_speech/fastspeech2.py:74
↓ 5 callersMethod_load_samples_from_tsv
(cls, root: str, split: str)
modules/voice_conversion/fairseq/data/audio/speech_to_text_dataset.py:645
↓ 5 callersMethod_map_index
(self, key, index)
modules/voice_conversion/fairseq/data/round_robin_zip_datasets.py:46
↓ 5 callersMethod_map_index
If dataset A has length N and dataset B has length M then index 1 maps to index 1 of dataset A, and index N + 1 maps to index
modules/voice_conversion/fairseq/data/multi_corpus_dataset.py:145
↓ 5 callersFunction_set_legacy_defaults
Helper to set default arguments based on *add_args*.
modules/voice_conversion/fairseq/dataclass/utils.py:196
↓ 5 callersMethodadd_args
Add optimizer-specific arguments to the parser.
modules/voice_conversion/fairseq/optim/sgd.py:18
↓ 5 callersFunctionadd_dataset_args
(parser, train=False, gen=False)
modules/voice_conversion/fairseq/options.py:312
↓ 5 callersFunctionadd_distributed_training_args
(parser, default_world_size=None)
modules/voice_conversion/fairseq/options.py:319
↓ 5 callersMethodbackward
(ctx, grad_output)
modules/voice_conversion/fairseq/modules/base_layer.py:156
↓ 5 callersFunctionbase_architecture
(args)
modules/voice_conversion/fairseq/models/fconv.py:700
↓ 5 callersFunctionbase_architecture
(args)
modules/voice_conversion/fairseq/models/transformer/transformer_legacy.py:169
↓ 5 callersMethodbuild_embedding
(dictionary, embed_dim)
modules/voice_conversion/fairseq/models/speech_to_text/s2t_transformer.py:241
↓ 5 callersMethodbuild_optimizer
Args: cfg (omegaconf.DictConfig): fairseq args params (iterable): iterable of parameters to optimize
modules/voice_conversion/fairseq/optim/amp_optimizer.py:30
↓ 5 callersFunctioncheckpoint_paths
Retrieves all checkpoints found in `path` directory. Checkpoints are identified by matching filename to the specified pattern. If the pattern
modules/voice_conversion/fairseq/checkpoint_utils.py:527
↓ 5 callersMethodclose
(self)
modules/voice_conversion/fairseq/nan_detector.py:106
↓ 5 callersFunctioncreate_conv1
(in_channels: int, out_channels: int, initialization_method: Union[str, Callable[[Module], Mo
talkinghead/tha3/nn/conv.py:54
↓ 5 callersFunctioncreate_conv3_block_from_block_args
( in_channels: int, out_channels: int, block_args: Optional[BlockArgs] = None)
talkinghead/tha3/nn/conv.py:116
↓ 5 callersFunctioncreate_poser
( device: torch.device, module_file_names: Optional[Dict[str, str]] = None, eyebrow_mo
talkinghead/tha3/poser/modes/separable_half.py:257
↓ 5 callersMethodforward_length
(self, normalize, encoder_out)
modules/voice_conversion/fairseq/models/nat/nonautoregressive_transformer.py:235
↓ 5 callersMethodforward_length_prediction
(self, length_out, encoder_out, tgt_tokens=None)
modules/voice_conversion/fairseq/models/nat/nonautoregressive_transformer.py:368
↓ 5 callersMethodget_batch_iterator
Get an iterator that yields batches of data from the given dataset. Args: dataset (~fairseq.data.FairseqDataset): datase
modules/voice_conversion/fairseq/tasks/fairseq_task.py:207
↓ 5 callersMethodget_category
(self)
talkinghead/tha3/poser/poser.py:68
↓ 5 callersMethodget_decoder_langtok
(self, tgt_lang)
modules/voice_conversion/fairseq/tasks/multilingual_translation.py:173
↓ 5 callersFunctionget_from_selector
(selector: str)
modules/websearch/script.py:87
↓ 5 callersMethodget_image_size
(self)
talkinghead/tha3/poser/poser.py:134
↓ 5 callersMethodget_normalized_probs_scriptable
Get normalized probabilities (or log probs) from a net's output.
modules/voice_conversion/fairseq/models/fairseq_decoder.py:72
↓ 5 callersMethodget_range
(self)
talkinghead/tha3/poser/poser.py:59
↓ 5 callersMethodget_transforms
Split-specific feature transforms. Allowing train set wildcard `_train`, evaluation set wildcard `_eval` and general wildcard `*` for
modules/voice_conversion/fairseq/data/audio/data_cfg.py:136
↓ 5 callersFunctionindex_put
(tensor, indices, value)
modules/voice_conversion/fairseq/utils.py:736
↓ 5 callersMethodis_discrete
(self)
talkinghead/tha3/poser/poser.py:56
↓ 5 callersFunctionis_xla_tensor
(tensor)
modules/voice_conversion/fairseq/utils.py:732
↓ 5 callersFunctionload_indexed_dataset
A helper function for loading indexed datasets. Args: path (str): path to indexed dataset (e.g., 'data-bin/train') dictionary (~f
modules/voice_conversion/fairseq/data/data_utils.py:73
↓ 5 callersFunctionload_langpair_dataset
( data_path, split, src, src_dict, tgt, tgt_dict, combine, dataset_impl, u
modules/voice_conversion/fairseq/tasks/translation.py:39
← previousnext →101–200 of 4,572, ranked by callers