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.

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

Fix passing torch_dtype and device_map via model_init_kwargs #12

Merged
merged 2 commits into from
Jun 12, 2024
Merged
Changes from 1 commit
Commits
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
Next Next commit
Fix passing torch_dtype and device_map via model_init_kwargs
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
  • Loading branch information
tdoublep committed Jun 11, 2024
commit dd10acfad1a1718381440ab478168c5da16fcbdc
4 changes: 2 additions & 2 deletions auto_fp8/modeling.py
Original file line number Diff line number Diff line change
@@ -71,10 +71,10 @@ def skip(*args, **kwargs):
torch.cuda.empty_cache()

# Important defaults
if not hasattr(model_init_kwargs, "torch_dtype"):
if not "torch_dtype" in model_init_kwargs:
model_init_kwargs["torch_dtype"] = "auto"

if not hasattr(model_init_kwargs, "device_map"):
if not "device_map" in model_init_kwargs:
model_init_kwargs["device_map"] = "auto"

merged_kwargs = {**model_init_kwargs, **cached_file_kwargs}