icefall

History

Improve diagnostics code memory-wise and accumulate more stats. (#373 )

* Update diagnostics, hopefully print more stats.

# Conflicts:
#	egs/librispeech/ASR/pruned_transducer_stateless4b/train.py

* Remove memory-limit options arg

* Remove unnecessary option for diagnostics code, collect on more batches

2022-05-19 11:45:59 +08:00

__init__.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

asr_datamodule.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

beam_search.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

conformer.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

decode.py

Ignore padding frames during RNN-T decoding. (#358 )

2022-05-13 07:39:14 +08:00

decoder.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

encoder_interface.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

export.py

Various fixes to support torch script. (#371 )

2022-05-16 21:46:59 +08:00

joiner.py

Various fixes to support torch script. (#371 )

2022-05-16 21:46:59 +08:00

model.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

pretrained.py

Ignore padding frames during RNN-T decoding. (#358 )

2022-05-13 07:39:14 +08:00

subsampling.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00

test_model.py

Various fixes to support torch script. (#371 )

2022-05-16 21:46:59 +08:00

train.py

Improve diagnostics code memory-wise and accumulate more stats. (#373 )

2022-05-19 11:45:59 +08:00

transformer.py

Support computing RNN-T loss with torchaudio (#316 )

2022-04-19 18:47:13 +08:00