400 Commits

Author	SHA1	Message	Date
yaozengwei	8129470586	first upload the conv_emformer_transducer recipe, integrating convolution module into emformer layers.	2022-04-10 20:24:20 +08:00
Daniel Povey	962cf868c9	Fix import	2022-04-10 15:31:46 +08:00
Daniel Povey	d1e4ae788d	Refactor how learning rate is set.	2022-04-10 15:25:27 +08:00
Daniel Povey	82d58629ea	Implement 2p version of learning rate schedule.	2022-04-10 13:50:31 +08:00
Daniel Povey	da50525ca5	Make lrate rule more symmetric	2022-04-10 13:25:40 +08:00
Daniel Povey	4d41ee0caa	Implement 2o schedule	2022-04-09 18:37:03 +08:00
Daniel Povey	db72aee1f0	Set 2n rule..	2022-04-09 18:15:56 +08:00
yaozengwei	3e131891a2	update test_emformer.py	2022-04-08 20:43:54 +08:00
yaozengwei	d58002c414	update emformer.py	2022-04-08 20:31:32 +08:00
Daniel Povey	0f8ee68af2	Fix bug	2022-04-08 16:53:42 +08:00
Daniel Povey	f587cd527d	Change exponential part of lrate to be epoch based	2022-04-08 16:24:21 +08:00
Daniel Povey	6ee32cf7af	Set new scheduler	2022-04-08 16:10:06 +08:00
Fangjun Kuang	78b8792d1d	Fix potential bugs in PyTorch that exist in label_smoothing. (#300)	2022-04-08 13:41:33 +08:00
yaozengwei	2d1b90f758	update the docs of Emformer class in emformer.py	2022-04-08 10:59:39 +08:00
yaozengwei	374eacdd5c	First upload emformer_pruned_transducer_stateless recipe, refator emformer codes from torchaudio.	2022-04-07 21:32:59 +08:00
Daniel Povey	61486a0f76	Remove initial_speed	2022-04-06 13:17:26 +08:00
Daniel Povey	a41e93437c	Change some defaults in LR-setting rule.	2022-04-06 12:36:58 +08:00
yaozengwei	9423b3899f	Update emformer_pruned_transducer_stateless/emformer.py and upload emformer_pruned_transducer_stateless/test_emformer.py.	2022-04-05 22:28:46 +08:00
yaozengwei	fe43c1349e	First upload emformer_pruned_transducer_stateless/emformer.py, modified from torchaudio.	2022-04-05 22:28:46 +08:00
yaozengwei	b0bce20e21	Modify subsampling.py to make T'=T//4 strictly	2022-04-05 22:28:46 +08:00
Daniel Povey	2545237eb3	Changing initial_speed from 0.25 to 01	2022-04-05 18:00:54 +08:00
Daniel Povey	25724b5ce9	Bug-fix RE sign of target_rms	2022-04-05 13:49:35 +08:00
Daniel Povey	d1a669162c	Fix bug in lambda	2022-04-05 13:31:52 +08:00
Daniel Povey	ed8eba91e1	Reduce model_warm_step from 4k to 3k	2022-04-05 13:24:09 +08:00
Daniel Povey	c3169222ae	Simplified optimizer, rework somet things..	2022-04-05 13:23:02 +08:00
Daniel Povey	0f5957394b	Fix to reading scheudler from optim	2022-04-05 12:58:43 +08:00
Daniel Povey	1548cc7462	Fix checkpoint-writing	2022-04-05 11:19:40 +08:00
Wei Kang	cb3ba16f2b	Fix aishell prepare.sh when using pre-download data (#291)	2022-04-05 10:22:49 +08:00
Daniel Povey	47d49f29d7	Fix weight decay formula by adding 1/1-beta	2022-04-05 00:31:55 +08:00
Daniel Povey	2b0727a355	Fix weight decay formula by adding 1/1-beta	2022-04-05 00:31:28 +08:00
Daniel Povey	234366e51c	Fix type of parameter	2022-04-05 00:18:36 +08:00
Daniel Povey	179d0605ea	Change initialization to 0.25	2022-04-04 23:34:39 +08:00
Daniel Povey	d1f2f93460	Some fixes..	2022-04-04 22:40:18 +08:00
Daniel Povey	72f4a673b1	First draft of new approach to learning rates + init	2022-04-04 20:21:34 +08:00
Daniel Povey	4929e4cf32	Change how warm-step is set	2022-04-04 17:09:25 +08:00
Daniel Povey	a5bbcd7b71	Make training more efficient, avoid redoing some projections.	2022-04-04 14:14:03 +08:00
Daniel Povey	99e9d6c4b8	Some cleanups	2022-04-04 13:37:10 +08:00
Daniel Povey	0fd0828f79	Fix to joiner to allow different dims	2022-04-04 13:34:43 +08:00
Fangjun Kuang	87cf9231ea	Support specifying iteration number of checkpoints for decoding. (#289)	2022-04-03 13:02:08 +08:00
Daniel Povey	9f62a0296c	Revert transducer_stateless/ to state in upstream/master	2022-04-02 21:16:39 +08:00
Daniel Povey	807fcada68	Change learning speed of simple_lm_proj	2022-04-02 20:15:11 +08:00
Daniel Povey	34500afc43	Various bug fixes	2022-04-02 20:06:43 +08:00
Daniel Povey	8be10d3d6c	First draft of model rework	2022-04-02 20:03:21 +08:00
Daniel Povey	eec597fdd5	Merge changes from master	2022-04-02 18:45:20 +08:00
Daniel Povey	e0ba4ef3ec	Make layer dropout rate 0.075, was 0.1.	2022-04-02 17:48:54 +08:00
Fangjun Kuang	189ca555b1	Use Emformer as RNN-T encoder. (#278) ... * Add emformer model. * Copy files. * Use Emformer model as RNN-T encoder. * Support streaming decoding. * Minor fixes. * Add RNN-T Emformer for Aishell.	2022-04-02 13:37:39 +08:00
Daniel Povey	45f872c27d	Remove final dropout	2022-04-01 19:33:20 +08:00
Daniel Povey	92ec2e356e	Fix test-mode	2022-04-01 12:22:12 +08:00
Fangjun Kuang	e7493ede90	Don't use a lambda for dataloader's worker_init_fn. (#284) ... * Don't use a lambda for dataloader's worker_init_fn.	2022-03-31 20:32:00 +08:00
Daniel Povey	8caa18e2fe	Bug fix to warmup_scale	2022-03-31 17:30:51 +08:00