Skip to content

[FEATURE] Update Qwen2.5VL to v4.54.1 #1421

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

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

[FEATURE] Update Qwen2.5VL to v4.54.1 #1421

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
6136c44
update scripts
zhtmike Nov 6, 2025
f4cd807
update scripts
zhtmike Nov 6, 2025
6df3a76
fix several bugs
zhtmike Nov 6, 2025
3c4f46f
Merge branch 'master' into qwen25vl_4_54_1
zhtmike Nov 27, 2025
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 mindone/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3057,7 +3057,7 @@ def _find_mismatched_keys(
if state_dict is not None:
# Whole checkpoint
# checkpoint mapping from pt to hf
matching = [s for s in key_renaming_mapping.keys() if "LayerNorm.gamma" in s]
matching = key_renaming_mapping.keys()
Copy link
Collaborator

@zhanghuiyao zhanghuiyao Nov 7, 2025

Choose a reason for hiding this comment

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

这里是为什么去掉了 layernorm.gamma 的判断，是之前写多了吗？

Copy link
Collaborator Author

@zhtmike zhtmike Nov 7, 2025

Choose a reason for hiding this comment

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

目前这个判断会导致qwen25vl 权重读取失败。去掉这个判断可以正确读取权重，具体修复待 @wcrzlh 分析

Copy link
Contributor

@wcrzlh wcrzlh Nov 27, 2025

Choose a reason for hiding this comment

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

此修改会造成其他模型加载权重有问题，需进一步验证才能合入。

if matching:
# Fix the key names when model weight names contain LayerNorm.gamma/LayerNorm.beta
state_dict = {key_renaming_mapping[k]: v for k, v in state_dict.items() if k in key_renaming_mapping}
Expand Down Expand Up @@ -3090,7 +3090,7 @@ def _find_mismatched_keys(
for shard_file in resolved_archive_file:
state_dict = load_state_dict(shard_file)
# checkpoint mapping from pt to hf
matching = [s for s in key_renaming_mapping.keys() if "LayerNorm.gamma" in s]
matching = key_renaming_mapping.keys()
if matching:
# Fix the key names when model weight names contain LayerNorm.gamma/LayerNorm.beta
state_dict = {
Expand Down
Loading