MCPcopy Create free account

hub / github.com/ASLP-lab/OSUM / functions

Functions1,852 in github.com/ASLP-lab/OSUM

⨍ Functions 1,852◇ Types & classes 348
↓ 1 callersMethod_get_constant_lr
(self, step)
OSUM-EChat/wenet/utils/scheduler.py:374
↓ 1 callersMethod_get_constant_lr
(self, step)
OSUM/wenet/utils/scheduler.py:374
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:193
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:372
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM-EChat/wenet/utils/scheduler.py:198
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM-EChat/wenet/utils/scheduler.py:377
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM/wenet/utils/scheduler.py:198
↓ 1 callersMethod_get_lr
Simple const lr policy
OSUM/wenet/utils/scheduler.py:377
↓ 1 callersMethod_get_warmup_lr
(self, step)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:365
↓ 1 callersMethod_get_warmup_lr
(self, step)
OSUM-EChat/wenet/utils/scheduler.py:370
↓ 1 callersMethod_get_warmup_lr
(self, step)
OSUM/wenet/utils/scheduler.py:370
↓ 1 callersMethod_istft
(self, magnitude, phase)
OSUM-EChat/tts/cosyvoice/hifigan/generator.py:342
↓ 1 callersFunction_linear_warmup_with_cosine_annealing
(max_lr, warmup_steps, step, decay_steps, min_lr)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:397
↓ 1 callersFunction_linear_warmup_with_cosine_annealing
(max_lr, warmup_steps, step, decay_steps, min_lr)
OSUM-EChat/wenet/utils/scheduler.py:402
↓ 1 callersFunction_linear_warmup_with_cosine_annealing
(max_lr, warmup_steps, step, decay_steps, min_lr)
OSUM/wenet/utils/scheduler.py:402
↓ 1 callersFunction_load_json_cmvn
Load the json format cmvn stats file and calculate cmvn Args: json_cmvn_file: cmvn stats file in json format Returns: a num
OSUM-EChat/wenet/utils/cmvn.py:21
↓ 1 callersFunction_load_json_cmvn
Load the json format cmvn stats file and calculate cmvn Args: json_cmvn_file: cmvn stats file in json format Returns: a num
OSUM/wenet/utils/cmvn.py:21
↓ 1 callersFunction_load_kaldi_cmvn
Load the kaldi format cmvn stats file and calculate cmvn Args: kaldi_cmvn_file: kaldi text style global cmvn file, which is
OSUM-EChat/wenet/utils/cmvn.py:46
↓ 1 callersFunction_load_kaldi_cmvn
Load the kaldi format cmvn stats file and calculate cmvn Args: kaldi_cmvn_file: kaldi text style global cmvn file, which is
OSUM/wenet/utils/cmvn.py:46
↓ 1 callersFunction_load_paraformer_cmvn
(cmvn_file)
OSUM/wenet/paraformer/convert_paraformer_to_wenet_config_and_ckpt.py:17
↓ 1 callersMethod_nearest_embedding_idx
(self, xs: torch.Tensor)
OSUM/wenet/ssl/bestrq/bestrq_model.py:290
↓ 1 callersMethod_noam_annealing
(self, initial_lr, step)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:610
↓ 1 callersMethod_noam_annealing
(self, initial_lr, step)
OSUM-EChat/wenet/utils/scheduler.py:616
↓ 1 callersMethod_noam_annealing
(self, initial_lr, step)
OSUM/wenet/utils/scheduler.py:616
↓ 1 callersFunction_noam_hold_annealing
(initial_lr, step, warmup_steps, hold_steps, decay_rate, min_lr)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:433
↓ 1 callersFunction_noam_hold_annealing
(initial_lr, step, warmup_steps, hold_steps, decay_rate, min_lr)
OSUM-EChat/wenet/utils/scheduler.py:438
↓ 1 callersFunction_noam_hold_annealing
(initial_lr, step, warmup_steps, hold_steps, decay_rate, min_lr)
OSUM/wenet/utils/scheduler.py:438
↓ 1 callersMethod_norm
(self, x)
OSUM-EChat/wenet/transformer/norm.py:19
↓ 1 callersMethod_norm
(self, x)
OSUM/wenet/transformer/norm.py:19
↓ 1 callersFunction_read_compressed_mat
Read a compressed matrix, see: https://github.com/kaldi-asr/kaldi/blob/master/src/matrix/compressed-matrix.h methods: CompressedMatrix::R
OSUM/wenet/dataset/kaldi_io.py:487
↓ 1 callersFunction_read_mat_ascii
(fd)
OSUM/wenet/dataset/kaldi_io.py:472
↓ 1 callersFunction_read_mat_binary
(fd)
OSUM/wenet/dataset/kaldi_io.py:450
↓ 1 callersMethod_relative_indices
(self, keys: torch.Tensor)
OSUM-EChat/wenet/transformer/attention.py:546
↓ 1 callersMethod_relative_indices
(self, keys: torch.Tensor)
OSUM/wenet/transformer/attention.py:546
↓ 1 callersFunction_sampler
(pdf: torch.Tensor, num_samples: int, device=torch.device('cpu'))
OSUM/wenet/ssl/bestrq/mask.py:5
↓ 1 callersMethod_sampler
(self, encoder_out, encoder_out_mask, ys_pad, ys_pad_lens, pre_acoustic_embeds)
OSUM/wenet/paraformer/paraformer.py:236
↓ 1 callersFunction_square_annealing
(initial_lr, step, max_steps, min_lr)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:384
↓ 1 callersFunction_square_annealing
(initial_lr, step, max_steps, min_lr)
OSUM-EChat/wenet/utils/scheduler.py:389
↓ 1 callersFunction_square_annealing
(initial_lr, step, max_steps, min_lr)
OSUM/wenet/utils/scheduler.py:389
↓ 1 callersFunction_squareroot_annealing
(initial_lr, step, max_steps, min_lr)
OSUM-EChat/tts/cosyvoice/utils/scheduler.py:377
↓ 1 callersFunction_squareroot_annealing
(initial_lr, step, max_steps, min_lr)
OSUM-EChat/wenet/utils/scheduler.py:382
↓ 1 callersFunction_squareroot_annealing
(initial_lr, step, max_steps, min_lr)
OSUM/wenet/utils/scheduler.py:382
↓ 1 callersMethod_stack_features
(self, input: torch.Tensor, input_lens: torch.Tensor)
OSUM/wenet/ssl/bestrq/bestrq_model.py:257
↓ 1 callersMethod_stft
(self, x)
OSUM-EChat/tts/cosyvoice/hifigan/generator.py:334
↓ 1 callersFunction_to_wenet_cmvn
(cmvn_file)
OSUM/wenet/paraformer/convert_paraformer_to_wenet_config_and_ckpt.py:54
↓ 1 callersFunctionadd_dataset_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:101
↓ 1 callersFunctionadd_dataset_args
(parser)
OSUM/wenet/utils/train_utils.py:101
↓ 1 callersFunctionadd_ddp_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:187
↓ 1 callersFunctionadd_ddp_args
(parser)
OSUM/wenet/utils/train_utils.py:184
↓ 1 callersFunctionadd_deepspeed_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:204
↓ 1 callersFunctionadd_deepspeed_args
(parser)
OSUM/wenet/utils/train_utils.py:201
↓ 1 callersFunctionadd_fsdp_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:227
↓ 1 callersFunctionadd_fsdp_args
(parser)
OSUM/wenet/utils/train_utils.py:224
↓ 1 callersFunctionadd_lora_args
Configure parameters for LoRA fine-tuning. Set use_lora and only_optimize_lora to true to enable LoRA functionality. LoRA will be inject
OSUM-EChat/wenet/utils/train_utils.py:126
↓ 1 callersFunctionadd_lora_args
Configure parameters for LoRA fine-tuning. Set use_lora and only_optimize_lora to true to enable LoRA functionality. LoRA will be inject
OSUM/wenet/utils/train_utils.py:123
↓ 1 callersFunctionadd_model_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:59
↓ 1 callersFunctionadd_model_args
(parser)
OSUM/wenet/utils/train_utils.py:59
↓ 1 callersFunctionadd_trace_args
(parser)
OSUM-EChat/wenet/utils/train_utils.py:89
↓ 1 callersFunctionadd_trace_args
(parser)
OSUM/wenet/utils/train_utils.py:89
↓ 1 callersMethodalign
(self, audio_file: str, label: str)
OSUM/wenet/cli/model.py:156
↓ 1 callersFunctionamp2db
(amp)
OSUM/wenet/dataset/wav_distortion.py:27
↓ 1 callersFunctionapply_fsdp_checkpointing
(model, ckpt_layer_types: tuple)
OSUM-EChat/wenet/utils/fsdp_utils.py:100
↓ 1 callersFunctionapply_fsdp_checkpointing
(model, ckpt_layer_types: tuple)
OSUM/wenet/utils/fsdp_utils.py:100
↓ 1 callersFunctionarrange
(s, e, length, keep_length, mask_idc)
OSUM/wenet/ssl/bestrq/mask.py:112
↓ 1 callersFunctionattention_beam_search
( model, encoder_out: torch.Tensor, encoder_mask: torch.Tensor, beam_size: int = 10, lengt
OSUM-EChat/wenet/transformer/search.py:250
↓ 1 callersFunctionattention_beam_search
( model, encoder_out: torch.Tensor, encoder_mask: torch.Tensor, beam_size: int = 10, lengt
OSUM/wenet/transformer/search.py:250
↓ 1 callersFunctionattention_rescoring
Args: ctc_prefix_results(List[DecodeResult]): ctc prefix beam search results
OSUM-EChat/wenet/transformer/search.py:369
↓ 1 callersFunctionbasic_greedy_search
( model: torch.nn.Module, encoder_out: torch.Tensor, encoder_out_lens: torch.Tensor, n_steps:
OSUM/wenet/transducer/search/greedy_search.py:6
↓ 1 callersFunctionbatch_backward
(model, scaler, info_dict)
OSUM-EChat/wenet/utils/train_utils.py:724
↓ 1 callersFunctionbatch_backward
(model, scaler, info_dict)
OSUM/wenet/utils/train_utils.py:718
↓ 1 callersMethodbatch_to_cache
Args: cache: [state_m, state_c] state_ms: [1*n_layers, bs, ...] state_cs: [1*n_layers, bs, ...]
OSUM/wenet/transducer/predictor.py:128
↓ 1 callersMethodbuild_graph
Build the ContextGraph from a list of token list. It first build a trie from the given token lists, then fill the fail arc for each tr
OSUM-EChat/wenet/utils/context_graph.py:144
↓ 1 callersMethodbuild_graph
Build the ContextGraph from a list of token list. It first build a trie from the given token lists, then fill the fail arc for each tr
OSUM/wenet/utils/context_graph.py:144
↓ 1 callersMethodcache_to_batch
Args: cache : [[state_m_1, state_c_1], [state_m_1, state_c_1]...] Returns: new_caceh: [state_ms, state_cs],
OSUM/wenet/transducer/predictor.py:150
↓ 1 callersMethodcalculate
(self, lab, rec)
OSUM/tools/compute-cer.py:102
↓ 1 callersMethodcalculate
(self, lab, rec)
OSUM/tools/compute-wer.py:96
↓ 1 callersMethodcalculate_downsampling_factor
(self, i: int)
OSUM/wenet/squeezeformer/encoder.py:251
↓ 1 callersMethodcalculate_downsampling_factor
(self, i: int)
OSUM/wenet/bin/export_onnx_gpu.py:253
↓ 1 callersMethodcheck_ascending_list
(self)
OSUM/wenet/squeezeformer/encoder.py:243
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:100
↓ 1 callersMethodcheck_equal
(self)
OSUM/wenet/bin/export_onnx_bpu.py:143
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:187
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:287
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:347
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:473
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:532
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:590
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:700
↓ 1 callersMethodcheck_equal
(self, module)
OSUM/wenet/bin/export_onnx_bpu.py:829
↓ 1 callersFunctioncheck_gradient_checkpoint
(model)
OSUM-EChat/wenet/utils/fsdp_utils.py:83
↓ 1 callersFunctioncheck_gradient_checkpoint
(model)
OSUM/wenet/utils/fsdp_utils.py:83
↓ 1 callersFunctioncheck_modify_and_save_config
(args, configs)
OSUM-EChat/tts/cosyvoice/utils/train_utils.py:72
↓ 1 callersFunctioncheck_modify_and_save_config
(args, configs, symbol_table)
OSUM-EChat/wenet/utils/train_utils.py:283
↓ 1 callersFunctioncheck_modify_and_save_config
(args, configs, symbol_table)
OSUM/wenet/utils/train_utils.py:280
↓ 1 callersFunctioncheck_wer
(enc, ctc, args, conf, tokenizer)
OSUM/tools/onnx2horizonbin.py:152
↓ 1 callersFunctioncif
(hidden: torch.Tensor, alphas: torch.Tensor, threshold: float)
OSUM/wenet/paraformer/cif.py:250
↓ 1 callersFunctioncif_without_hidden
(alphas: torch.Tensor, threshold: float)
OSUM/wenet/paraformer/cif.py:225
↓ 1 callersMethodcompute_feats
(self, audio_file: str)
OSUM/wenet/cli/model.py:61
↓ 1 callersMethodcompute_loss
Computes diffusion loss Args: x1 (torch.Tensor): Target shape: (batch_size, n_feats, mel_timesteps) m
OSUM-EChat/tts/cosyvoice/flow/flow_matching.py:145
↓ 1 callersMethodcompute_preds
(self, x, y, negatives)
OSUM/wenet/ctl_model/asr_model_ctl.py:241
← previousnext →401–500 of 1,852, ranked by callers