mmarch/text-generation-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Bump peft and transformers

Browse source
This commit is contained in:
oobabooga 2023-08-22 13:14:59 -07:00
parent 727fd229f4
commit 335c49cc7e
2 changed files with 6 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

4
modules/training.py
View file

@ -21,7 +21,7 @@ from datasets import Dataset, load_dataset
from peft import (
LoraConfig,
get_peft_model,
prepare_model_for_int8_training,
prepare_model_for_kbit_training,
set_peft_model_state_dict
)
from peft.utils.other import \
@ -483,7 +483,7 @@ def do_train(lora_name: str, always_override: bool, save_steps: int, micro_batch
# == Start prepping the model itself ==
if not hasattr(shared.model, 'lm_head') or hasattr(shared.model.lm_head, 'weight'):
logger.info("Getting model ready...")
prepare_model_for_int8_training(shared.model)
prepare_model_for_kbit_training(shared.model)
# base model is now frozen and should not be reused for any other LoRA training than this one
shared.model_dirty_from_training = True