Commit Graph

350 Commits

Author SHA1 Message Date
hiyouga
a465785df0 fix #1349
Former-commit-id: f8703aac08
2023-11-02 17:02:44 +08:00
hiyouga
15cef791ba fix #1356
Former-commit-id: dff128c7e3
2023-11-02 16:51:52 +08:00
hiyouga
22b3c913e9 fix #1325
Former-commit-id: 083787dbfe
2023-11-01 23:38:49 +08:00
hiyouga
d57f70bf8a fix chat
Former-commit-id: 8b912690e3
2023-11-01 23:07:58 +08:00
hiyouga
e3d64e119e update gradio, support multiple resp in api
Former-commit-id: 84af10cec9
2023-11-01 23:02:16 +08:00
hiyouga
abd6ec174f fix SFT trainer
Former-commit-id: d8cf8cfdeb
2023-10-31 21:52:52 +08:00
hiyouga
6493f6d2e9 fix #1316
Former-commit-id: f4e4a04529
2023-10-31 11:32:08 +08:00
hiyouga
d48478ef88 update constants
Former-commit-id: f28a034a9b
2023-10-29 13:30:20 +08:00
hiyouga
bf0faf129d fix vicuna template
Former-commit-id: 52fc24d166
2023-10-27 22:15:25 +08:00
hiyouga
5705c82cd8 fix chatglm3 template
Former-commit-id: 4117f38827
2023-10-27 21:12:06 +08:00
hiyouga
8a76b1e499 support chatglm3
Former-commit-id: 1c0ab9a908
2023-10-27 19:16:28 +08:00
hiyouga
fcfcac4858 support dataset cache
Former-commit-id: 3fe7df628d
2023-10-26 21:48:45 +08:00
hiyouga
4906092811 fix #1287
Former-commit-id: 838ed9aa87
2023-10-26 17:49:41 +08:00
hiyouga
4d6606a72c fix #1285
Former-commit-id: aff9363ce3
2023-10-26 16:34:52 +08:00
hiyouga
84e27a1c0b remove filter in preprocess
Former-commit-id: 2caf91f824
2023-10-23 23:46:02 +08:00
hiyouga
daeff710eb update neftune logic
Former-commit-id: 7de7174ce3
2023-10-22 17:42:13 +08:00
hiyouga
2af83198c7 fix webui
Former-commit-id: 11b55a3270
2023-10-22 17:24:56 +08:00
hiyouga
59fd3155a1 add new options in webui
Former-commit-id: f793ca0a2c
2023-10-22 17:17:58 +08:00
hiyouga
ce112dc744 fix recursion error
Former-commit-id: b79ca8781e
2023-10-22 16:28:37 +08:00
hiyouga
d6c77d9196 reimplement neftune
Former-commit-id: 7b4acf7265
2023-10-22 16:15:08 +08:00
anvie
3635823fbe add NEFTune optimization
Former-commit-id: 57fb40aa04
2023-10-21 13:24:10 +07:00
hiyouga
d18c708f14 fix openchat template
Former-commit-id: 8fdff07e1f
2023-10-21 01:25:42 +08:00
hiyouga
77781d9516 fix tokenizer padding side in evaluate.py
Former-commit-id: 641ffa2f6e
2023-10-21 00:30:04 +08:00
hiyouga
95697652f1 fix #1232
Former-commit-id: b665e9e133
2023-10-20 23:28:52 +08:00
hiyouga
1712a59280 fix #1215
Former-commit-id: 0fcf66049d
2023-10-19 16:19:21 +08:00
hiyouga
4930118761 fix #1218
Former-commit-id: 7a11a42dfd
2023-10-19 16:17:41 +08:00
hiyouga
0622f8a3a3 fix #1228
Former-commit-id: cb0edd2302
2023-10-19 15:54:10 +08:00
hiyouga
f6a8069159 fix #1217
Former-commit-id: 6496a99b7d
2023-10-19 15:52:24 +08:00
hiyouga
6c8bdeb6e0 rename webui
Former-commit-id: 1611fad6bc
2023-10-16 15:16:24 +08:00
hiyouga
8febf5974d fix #1197
Former-commit-id: 85480f2e86
2023-10-16 15:13:46 +08:00
hiyouga
451a90fca1 release v0.2.0
Former-commit-id: 7a53188048
2023-10-15 20:49:43 +08:00
hiyouga
eafde5b73f fix config, #1191
Former-commit-id: a6a04be2e6
2023-10-15 18:28:45 +08:00
hiyouga
e5e740b54d disable tqdm in webui mode
Former-commit-id: 0d63584c03
2023-10-15 16:18:25 +08:00
hiyouga
f3fa47fa7d refactor export, fix #1190
Former-commit-id: ea82f8a82a
2023-10-15 16:01:48 +08:00
hiyouga
0503d45782 fix eval resuming in webui
Former-commit-id: 273745f9b9
2023-10-15 15:45:38 +08:00
hiyouga
99592478c9 tiny fix
Former-commit-id: 3ad8c92eca
2023-10-15 05:02:48 +08:00
hiyouga
4f9ca28e11 fix callback
Former-commit-id: 1e9401744c
2023-10-15 04:59:44 +08:00
hiyouga
3ae6229140 implement webui resuming training
Former-commit-id: accde3cd39
2023-10-15 04:52:19 +08:00
hiyouga
eadb7c61af fix bugs in webui
Former-commit-id: fde05cacfc
2023-10-15 03:41:58 +08:00
hiyouga
a902ce4dc7 refactor webui
Former-commit-id: 7ed1fa6fe9
2023-10-15 03:06:21 +08:00
hiyouga
089785c71b fix loading dtype
Former-commit-id: c874e764b8
2023-10-14 20:15:24 +08:00
hiyouga
1faba4b704 fix #1176 #1177
Former-commit-id: 01d8cb1ca7
2023-10-14 20:00:17 +08:00
hiyouga
e585c789ce fix #1184
Former-commit-id: af18b0dce7
2023-10-14 19:20:11 +08:00
hiyouga
9ef9cb316b fix webui
Former-commit-id: b240b6792f
2023-10-13 16:27:59 +08:00
hiyouga
2562376f84 fix ppo args
Former-commit-id: 11bd271364
2023-10-11 23:40:50 +08:00
hiyouga
c9d1cd108d refactor model_dtype, fix PPO trainer
Former-commit-id: 2818af0b09
2023-10-11 23:16:01 +08:00
hiyouga
c350ba0f05 add averaging in evaluation
Former-commit-id: 5310e4d182
2023-10-10 23:16:31 +08:00
hiyouga
141937ead6 fix aquila template, repair sft packing mechanism
Former-commit-id: be420e4179
2023-10-10 18:49:55 +08:00
hiyouga
7082526df5 tiny fix
Former-commit-id: e1dcb8e4dc
2023-10-10 17:41:13 +08:00
hiyouga
180fd06e61 fix flash shift short attention
Former-commit-id: 0a356bc897
2023-10-09 17:54:48 +08:00