Skip to content

fix qwen_vl and load_pretrained patch #2190

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

Merged
merged 1 commit into from
Oct 13, 2025
Merged

fix qwen_vl and load_pretrained patch #2190

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions mindnlp/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,8 +184,8 @@ def wrapper(
pretrained_model_name_or_path,
**kwargs,
):
device_map = kwargs.pop("device_map", None)
sharded_metadata = kwargs.pop("sharded_metadata", None)
device_map = kwargs.get("device_map", None)
sharded_metadata = kwargs.get("sharded_metadata", None)

# if device_map is not None and not initialize distribute module, raise Error.
if device_map is not None:
Expand Down
2 changes: 1 addition & 1 deletion mindtorch/nn/functional.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1220,7 +1220,7 @@ def scaled_dot_product_attention(query, key, value, attn_mask=None, dropout_p=0.

attn_weight = query @ key.transpose(-2, -1) * scale_factor
attn_weight += attn_bias
attn_weight = softmax(attn_weight, dim=-1, dtype=mindtorch.float32).to(query.dtype)
attn_weight = softmax(attn_weight, dim=-1)
attn_weight = dropout(attn_weight, dropout_p, training=True)
return attn_weight @ value

Expand Down
Loading