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

How to embed the weights into the library? #137

Open
monkbai opened this issue Jan 4, 2021 · 3 comments
Open

How to embed the weights into the library? #137

monkbai opened this issue Jan 4, 2021 · 3 comments

Comments

@monkbai
Copy link

monkbai commented Jan 4, 2021

So far the compiled library seems to do not contain weights, is it possible to embed the weights into the binary?

@github-actions
Copy link

github-actions bot commented Jan 4, 2021

Thank you so much for filing the issue. We will look at it and take appropriate action as soon as possible.' first issue

@srohit0
Copy link
Member

srohit0 commented Jan 6, 2021

weights are part of library and header file.

Can you be more specific on how do you know if the compiled library does not contain weights?

Also, let us know what you find in apache/tvm#7170 ?

@monkbai
Copy link
Author

monkbai commented Jan 6, 2021

If I follow the steps described in the video you provided and compile the onnx model under the directory with command compile-onnx, it will generate a cpp file and *.weight files, it looks like the weights are read at runtime.

As for tvm, I haven't succeeded to embed weights in binary so far, but they provided some advice, maybe you can follow the discussion here.

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

2 participants