Skip to content
This repository has been archived by the owner on Oct 25, 2024. It is now read-only.

Fixed issue of loading woq model for intel GPU #1614

Merged
merged 2 commits into from
Jun 21, 2024
Merged

Fixed issue of loading woq model for intel GPU #1614

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c973693
Fixed issue of loading woq model for intel GPU
PenghuiCheng Jun 15, 2024
042e086
Merge branch 'main' into penghuic/woq_loading_issue
PenghuiCheng Jun 20, 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
2 changes: 1 addition & 1 deletion intel_extension_for_transformers/transformers/modeling/modeling_auto.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,7 +180,7 @@ def build_woq_model(model, quantization_config):

def convert_model_to_public(model):
# reorder weight and scales if they have been transposed
if model.device == "xpu":
if model.device == "xpu" or (isinstance(model.device, torch.device) and model.device.type == "xpu"):
for name, module in model.named_modules():
if isinstance(module, WeightOnlyQuantizedLinear):
if module.weight_transposed:
Expand Down
Loading