Skip to content
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

Intel ARC/XPU Improvements #2052

Merged
merged 10 commits into from
Jul 23, 2023
Merged

Intel ARC/XPU Improvements #2052

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
2217d46
Empty XPU Cache
Steve-Tech Jul 21, 2023
aaaab51
A little cleanup around XPU code
Steve-Tech Jul 21, 2023
55691ae
Set `XPU_VISIBLE_DEVICES` from `gpus` arg
Steve-Tech Jul 21, 2023
3a97720
IPEX breaks if `import torch` isn't before `fastapi` in model_workers
Steve-Tech Jul 22, 2023
c34cc80
Empty XPU Cache in other spots
Steve-Tech Jul 22, 2023
2b352d2
Re-add dtype to `torch.xpu.optimize`
Steve-Tech Jul 22, 2023
1cffaca
Update fastchat/serve/model_worker.py
merrymercy Jul 23, 2023
061f32a
Update fastchat/serve/multi_model_worker.py
merrymercy Jul 23, 2023
fc1ce38
Update fastchat/serve/multi_model_worker.py
merrymercy Jul 23, 2023
aa015b6
Update fastchat/serve/model_worker.py
merrymercy Jul 23, 2023
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
2 changes: 2 additions & 0 deletions fastchat/model/compression.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -165,6 +165,8 @@ def load_compress_model(model_path, device, torch_dtype, use_fast, revision="mai
tensor = None
gc.collect()
torch.cuda.empty_cache()
if device == "xpu":
torch.xpu.empty_cache()

for name in model.state_dict():
if name not in linear_weights:
Expand Down
11 changes: 6 additions & 5 deletions fastchat/model/model_adapter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -239,13 +239,14 @@ def load_model(
adapter = get_model_adapter(model_path)
model, tokenizer = adapter.load_model(model_path, kwargs)

if (device == "cuda" and num_gpus == 1 and not cpu_offloading) or device == "mps":
if (device == "cuda" and num_gpus == 1 and not cpu_offloading) or device in (
"mps",
"xpu",
):
model.to(device)

elif device == "xpu":
model.eval()
model = model.to("xpu")
model = torch.xpu.optimize(model, dtype=torch.bfloat16, inplace=True)
if device == "xpu":
model = torch.xpu.optimize(model, dtype=kwargs["torch_dtype"], inplace=True)

if debug:
print(model)
Expand Down
2 changes: 2 additions & 0 deletions fastchat/model/model_codet5p.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,3 +104,5 @@ def __call__(
# clean
gc.collect()
torch.cuda.empty_cache()
if device == "xpu":
torch.xpu.empty_cache()
2 changes: 2 additions & 0 deletions fastchat/model/model_falcon.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -136,3 +136,5 @@ def generate_stream_falcon(
# clean
gc.collect()
torch.cuda.empty_cache()
if device == "xpu":
torch.xpu.empty_cache()
1 change: 1 addition & 0 deletions fastchat/serve/cli.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,6 +177,7 @@ def main(args):
f"Larger --num-gpus ({args.num_gpus}) than --gpus {args.gpus}!"
)
os.environ["CUDA_VISIBLE_DEVICES"] = args.gpus
os.environ["XPU_VISIBLE_DEVICES"] = args.gpus

if args.style == "simple":
chatio = SimpleChatIO(args.multiline)
Expand Down
2 changes: 2 additions & 0 deletions fastchat/serve/inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -253,6 +253,8 @@ def generate_stream(
del past_key_values, out
gc.collect()
torch.cuda.empty_cache()
if device == "xpu":
torch.xpu.empty_cache()


class ChatIO(abc.ABC):
Expand Down