12 Commits

Author	SHA1	Message	Date
woctordho	2f6046b072	Set scheduler interval to step and fewer warmup steps	2025-05-17 22:58:41 +08:00
woctordho	3d6fe064f5	Enable gradient checkpointing	2025-05-17 20:40:21 +08:00
Andy	db5a9fea4e	Update trainer.py ... fix lora trainer bug	2025-05-15 23:08:57 +08:00
chuxij	a5be43bbbe	fix_train_bug:adapter_name	2025-05-15 14:40:21 +00:00
chuxij	2a283c7e78	fix_train_bug:adapter_name	2025-05-15 14:35:57 +00:00
chuxij	6ca5bd880e	update ui and add examples	2025-05-13 07:00:35 +00:00
chuxij	ce2caec957	fix train bugs and add train details	2025-05-12 18:50:54 +00:00
Gong Junmin	2cc21c9a91	Update trainer.py	2025-05-08 17:59:32 +08:00
mrfakename	afa0332e8c	switch -> git+ url	2025-05-06 20:19:51 -07:00
mrfakename	a5746eaab6	work on pip package	2025-05-06 18:59:32 -07:00
Sayo	ac9a34a03f	[fix] encoding	2025-05-06 20:24:16 +08:00
chuxij	1fb1d20567	release training code of base model and lora	2025-05-05 07:20:50 +00:00