Guanbo Wang
ba245aa60f
Use default storage_type
2022-04-11 20:10:50 -04:00
Guanbo Wang
407998267c
Remove lazy_load option
2022-04-11 18:12:54 -04:00
Guanbo Wang
36ec10c41a
Merge remote-tracking branch 'upstream/master' into gigaspeech_recipe
2022-04-11 17:49:01 -04:00
Guanbo Wang
22f011e5ab
Update results
2022-04-11 21:46:40 +00:00
Guanbo Wang
f485b66d54
Decode with post-processing
2022-04-11 21:46:15 +00:00
Guanbo Wang
6d07cf9245
Decrease num_paths while CUDA OOM
2022-04-11 21:45:29 +00:00
Mingshuang Luo
118e195004
Update results for tedlium3 pruned RNN-T ( #307 )
...
* Update README.md
2022-04-11 22:19:26 +08:00
Mingshuang Luo
93c60a9d30
Code style check for librispeech pruned transducer stateless2 ( #308 )
2022-04-11 22:15:18 +08:00
Mingshuang Luo
8cb727e24a
Tedlium3 pruned transducer stateless ( #261 )
...
* update tedlium3-pruned-transducer-stateless-codes
* update README.md
* update README.md
* add fast beam search for decoding
* do a change for RESULTS.md
* do a change for RESULTS.md
* do a fix
* do some changes for pruned RNN-T
2022-04-11 17:08:53 +08:00
Wei Kang
7012fd65b5
Support mix precision training on the reworked model ( #305 )
...
* Add mix precision support
* Minor fixes
* Minor fixes
* Minor fixes
2022-04-11 16:49:54 +08:00
Daniel Povey
34aad74a2c
Merge pull request #303 from danpovey/fix_docs
...
Fix docs in optim.py
2022-04-11 15:14:06 +08:00
Daniel Povey
03c7c2613d
Fix docs in optim.py
2022-04-11 15:13:42 +08:00
Daniel Povey
6eb6d9b4cd
Merge pull request #288 from danpovey/reworked_model
...
Reworked model
2022-04-11 15:03:08 +08:00
Daniel Povey
5078332088
Fix adding learning rate to tensorboard
2022-04-11 14:58:15 +08:00
Daniel Povey
d5f9d49e53
Modify beam search to be efficient with current joienr
2022-04-11 12:35:29 +08:00
Daniel Povey
46d52dda10
Fix dir names
2022-04-11 12:03:41 +08:00
Wei Kang
f721a2fd7a
Minor fixes for logging ( #296 )
...
* Minor fixes for logging
* Minor fix
2022-04-10 23:34:18 +08:00
Zengwei Yao
08473a17aa
Modify init ( #301 )
...
* update icefall/__init__.py to import more common functions.
* update icefall/__init__.py
* make imports style consistent.
* exclude black check for icefall/__init__.py in pyproject.toml.
2022-04-10 23:29:28 +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
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
Fangjun Kuang
7c0070e6f6
Display torch version in the training log. ( #299 )
2022-04-08 11:39:54 +08:00
Guanbo Wang
3d2c261684
black
2022-04-06 21:05:42 -04:00
Wang, Guanbo
3e5436e5d4
Merge branch 'k2-fsa:master' into gigaspeech_recipe
2022-04-06 20:57:51 -04:00
Guanbo Wang
e56d32728d
Update RESULTS
2022-04-06 20:53:47 -04:00
Guanbo Wang
f857d5a9ea
Added README
2022-04-06 20:30:09 -04:00
Guanbo Wang
a4e1471d1d
Decrease num_paths while CUDA OOM
2022-04-06 19:47:44 -04:00
Guanbo Wang
d9addb7c43
update params in decode.py
2022-04-06 19:41:24 -04:00
Guanbo Wang
9a5340b21f
Use bpe_500
2022-04-06 19:06:54 -04:00
Guanbo Wang
64bb39bf17
Update params in train.py
2022-04-06 19:06:22 -04:00
Guanbo Wang
79211633ed
Use BucketingSampler for valid and test dataloader
2022-04-06 19:05:45 -04:00
Guanbo Wang
3ddcc7939b
Delete compute_fbank_gigaspeech.py
2022-04-06 18:57:53 -04:00
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
Zengwei Yao
ceeb95bcb8
update icefall/__init__.py to import more common functions. ( #294 )
2022-04-06 11:55:29 +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