42 Commits

Author SHA1 Message Date
tuna2134
b8ce11605c fix 2026-07-20 21:25:49 +09:00
tuna2134
0c2be00f0a ok 2026-07-20 21:08:44 +09:00
litagin02
c51c80fe1a Ver 2.7.0 maybe 2025-08-24 11:27:27 +09:00
litagin02
b6d09bad44 Fix: use sampler instead of batch_sampler for non-custom 2024-06-28 09:34:12 +09:00
litagin02
e78d2a6f7e Feat: use DistributedLengthGroupedSampler (experimental, to be checked) 2024-06-22 17:21:58 +09:00
litagin02
f4f94d4c20 change custom sampler to default 2024-05-29 05:46:31 +09:00
litagin02
1b656b37fa Feat: koharune-ami 2024-05-26 17:17:32 +09:00
litagin02
012f159ab5 Fix generating style bug 2024-05-26 11:07:43 +09:00
litagin02
f1e022a08c Fix tb logging 2024-05-26 08:27:39 +09:00
litagin02
5fa6210176 Batch sampler for backward compatibility, reduce tb log 2024-05-26 08:23:38 +09:00
litagin02
2274087da4 Fmt only (maybe) 2024-05-25 18:15:36 +09:00
litagin02
175aa62a26 Feat: make style vectors w.r.t. subdirs structure 2024-05-25 17:14:28 +09:00
litagin02
8976a3ed2f Refactor: Use PathConfig and pathlib instead of paths.yml loading 2024-05-25 10:25:22 +09:00
litagin02
4afe92d75f Fix: gen style in beginning of train for nested folders 2024-03-22 08:27:43 +09:00
litagin02
28fd715cb8 Restore gc empty_cache in train_ms_jp 2024-03-13 10:25:22 +09:00
litagin02
dada0af2b6 Format import using isort --profile black --gitignore --lai 2 . 2024-03-11 12:37:51 +09:00
litagin02
c776c08235 Apply black formatter 2024-03-11 09:47:47 +09:00
tsukumi
98ab8e7978 Refactor: separate module for utilities related to loading/saving checkpoints and safetensors 2024-03-09 15:58:57 +00:00
tsukumi
a84783a6cc Refactor: replace utils.HParams with HyperParameters Pydantic model
HyperParameters is largely a drop-in replacement for utils.HParams, which ensures type safety for hyper-parameters.
2024-03-08 15:52:37 +00:00
tsukumi
67ff3105c1 Refactor: moved utils.py to style_bert_vits2/models/ 2024-03-08 09:40:27 +00:00
tsukumi
fac4f9a8ab Refactor: rename text_processing to nlp
"text_processing" is clearer, but the import statement is longer.
"nlp" is shorter and makes it clear that it is natural language processing.
2024-03-08 06:20:44 +00:00
tsukumi
4d5c537f95 Refactor: introducing Ruff 2024-03-08 05:02:02 +00:00
tsukumi
89825e68d8 Refactor: moved model, attentions definitions and inference code to style_bert_vits2/models/
The code has not yet been cleaned up, just moved.
2024-03-06 23:43:25 +00:00
tsukumi
1936344c0c Refactor: remove old code that can be deleted and update where modules are imported 2024-03-06 22:51:25 +00:00
tsukumi
f880641eb5 Remove: modules under common/ that have been rewritten 2024-03-06 22:29:12 +00:00
tsukumi
9546495434 Refactor: moved commons.py to style_bert_vits2/models/ and added type definitions and comments 2024-03-06 22:17:03 +00:00
litagin02
ffdbb8e0d1 Feat: freezing decoder option 2024-02-23 20:57:54 +09:00
litagin02
a6a7a30fc6 First check repo-id is valid 2024-02-22 09:42:42 +09:00
litagin02
a04e569c86 Fix and upload all Data/ to hub 2024-02-20 00:42:00 +09:00
litagin02
d26336bf0a Upload to hug as future 2024-02-19 15:42:08 +09:00
litagin02
e0a601737a Error handling for hug upload 2024-02-17 20:26:58 +09:00
litagin02
8b76109b75 Fix hf push 2024-02-17 11:31:35 +09:00
litagin02
ff730476e7 Feat: huggingface hub push 2024-02-17 11:01:16 +09:00
litagin02
0a49d1ee00 Fix: wavlm shape bug when batch_size 1 2024-02-09 17:31:27 +09:00
litagin02
78689f6c2a Try to find best setting 2024-02-07 13:42:20 +09:00
litagin02
b2cde01d77 Fix incompatible setting 2024-02-07 13:34:45 +09:00
litagin02
0d7de644cd Update comment 2024-02-07 10:29:49 +09:00
litagin02
a0d4444db7 Try to reduce memory consumption (experimental) 2024-02-07 00:12:38 +09:00
litagin02
c81cc7b06f Add experimental speedup option for training (disable log and eval) 2024-02-06 10:21:06 +09:00
litagin02
b0bb147d44 Restore gc and emptycache when training 2024-02-05 16:02:16 +09:00
litagin02
99e1af4b58 Fix 2024-02-02 21:42:34 +09:00
litagin02
e566ac62f3 Update 2024-02-02 21:11:23 +09:00