423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-15 19:30:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix #242

Browse Source
This commit is contained in:
hiyouga
2023-07-25 17:04:02 +08:00
parent 2c31e05e63
commit 00efa8a07f
4 changed files with 22 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/llmtuner/tuner/core/loader.py
View File

@@ -93,9 +93,6 @@ def load_model_and_tokenizer(
config_kwargs["device_map"] = {"": int(os.environ.get("LOCAL_RANK", "0"))}
logger.info("Quantizing model to {} bit.".format(model_args.quantization_bit))
if not is_trainable: # `device_map=auto` should be used for inference only
config_kwargs["device_map"] = "auto"
if model_args.checkpoint_dir is not None and finetuning_args.finetuning_type == "full":
model_to_load = model_args.checkpoint_dir[0]
else: