Skip to content

Support loading hugging face checkpoint #286

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
Yantom1 wants to merge 2 commits into from
Closed

Support loading hugging face checkpoint #286

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
21282c9
Support loading checkpoint with INC
Yantom1 Jul 8, 2024
0e5254e
load_cp explanation
Yantom1 Jul 9, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions examples/text-generation/run_generation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -306,6 +306,12 @@ def __call__(self, parser, namespace, values, option_string=None):
action="store_true",
help="Whether to enable device map auto. In case no space left on cpu, weights will be offloaded to disk.",
)
parser.add_argument(
"--load_cp",
action="store_true",
help="Whether to load model from hugging face checkpoint.",
)

args = parser.parse_args()

if args.torch_compile:
Expand Down
9 changes: 9 additions & 0 deletions examples/text-generation/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -222,6 +222,14 @@ def setup_model(args, model_dtype, model_kwargs, logger):
model = AutoModelForCausalLM.from_pretrained(
args.model_name_or_path, torch_dtype=model_dtype, quantization_config=quantization_config, **model_kwargs
)
elif args.load_cp:
from neural_compressor.torch.quantization import load
model = load(
model_name_or_path=args.model_name_or_path,
format="huggingface",
device="hpu",
**model_kwargs
)
else:
if args.peft_model is not None:
model = peft_model(args, model_dtype, logger, **model_kwargs)
Expand Down Expand Up @@ -457,6 +465,7 @@ def initialize_model(args, logger):
model_kwargs = {
"revision": args.model_revision,
"token": args.token,
"torch_dtype": torch.bfloat16,

@MrGeva MrGeva Jul 10, 2024

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please remove if not needed

}

model = (
Expand Down