Fangjun Kuang da5dc4c504

Merge a133571dc7d073290b927877b54ca4cad06b0ad9 into f2387fe523c6f89987f3723bfa967095e8de5127

2021-10-14 20:20:43 +08:00

..

dataset

Add recipe for the yes_no dataset. (#16)

2021-08-23 11:36:29 +08:00

shared

Refactoring (#4)

2021-08-04 14:53:02 +08:00

__init__.py

Minor fixes after review.

2021-07-20 10:02:20 +08:00

bpe_graph_compiler.py

Fix code style and add copyright. (#18)

2021-08-23 10:43:59 +08:00

checkpoint.py

Fix code style and add copyright. (#18)

2021-08-23 10:43:59 +08:00

decode.py

Support pure ctc decoding requiring neither a lexicon nor an n-gram LM (#58)

2021-09-26 14:21:49 +08:00

dist.py

WIP: Support multi-node multi-GPU training.

2021-09-30 01:08:09 +08:00

graph_compiler.py

Refactor decode.py to make it more readable and more modular. (#44)

2021-09-20 15:44:54 +08:00

lexicon.py

Use new APIs with k2.RaggedTensor (#38)

2021-09-08 14:55:30 +08:00

utils.py

Use LossRecord to record and print the loss for the training process (#62)

2021-10-12 15:58:03 +08:00