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

关于导出v5lite-e-mnnd_fp16.mnn模型 #273

Open
moyans opened this issue Sep 25, 2024 · 0 comments
Open

关于导出v5lite-e-mnnd_fp16.mnn模型 #273

moyans opened this issue Sep 25, 2024 · 0 comments

Comments

@moyans
Copy link

moyans commented Sep 25, 2024

利用export出来onnx转mnn模型出现一堆胶水算子,但是下载您提供的是没有的?想请教下是我哪里的步骤问题。

我是下载了官方的v5lite-e.pt模型, 利用export 转v5lite-e.onnx, 通过onnxsim优化下, 在利用mnnconvert将v5lite-e-sim.onnx转为v5lite-e-sim.mnn

netron可视化对比下载好的v5lite-e-mnnd_fp16.mnn多了很多胶水算子。

屏幕截图 2024-09-25 151310 屏幕截图 2024-09-25 151448

转onnx时候试过加或者不加mnnd/concat貌似都有很多Const算子
python export.py --weights v5lite-e.pt --mnnd --batch-size 1
onnxsim 优化
python -m onnxsim v5lite-e.onnx v5lite-e-sim.onnx
mnnconvert(版本2.9.3)导出
mnnconvert -f ONNX --modelFile v5lite-e-sim.onnx --MNNModel v5lite-e-sim.mnn --bizCode MNN

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant