423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-16 11:50:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

improve model export

Browse Source
This commit is contained in:
hiyouga
2024-01-05 18:51:49 +08:00
parent f6fdd83f8a
commit d2a676c8ba
3 changed files with 61 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/llmtuner/train/tuner.py
View File

@@ -1,3 +1,4 @@
import torch
from typing import TYPE_CHECKING, Any, Dict, List, Optional
from llmtuner.extras.callbacks import LogCallback
@@ -46,7 +47,12 @@ def export_model(args: Optional[Dict[str, Any]] = None):
logger.warning("Cannot merge adapters to a quantized model.")
model.config.use_cache = True
model = model.to("cpu")
if getattr(model.config, "torch_dtype", None) == "bfloat16":
model = model.to(torch.bfloat16).to("cpu")
else:
model = model.to(torch.float16).to("cpu")
setattr(model.config, "torch_dtype", "float16")
model.save_pretrained(
save_directory=model_args.export_dir,
max_shard_size="{}GB".format(model_args.export_size),