-
Notifications
You must be signed in to change notification settings - Fork 19.5k
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
Layers in nested list in custom layers can not be saved/loaded correctly. #20598
Comments
Thank you for your response. |
Hi @yubori - Oh! You are error is regarding Attached gist for the reference as well. |
Hi @mehtamansi29, What's the point of relaxing the error bounds so that no error occurs? |
Hi @yubori- Here there is an assertion error.
Because of tolerance variation in both prediction it is giving an error. Here instead of And Here when we comparing model weights both are same. You can find the same in gist. |
Hi @mehtamansi29, As you can see in your last gist, there is the message of "Predictions differ." The prediction results are between the base model and the reloaded model. Please dump the NG model's weights. There are no weights of NGLayer. Keras3.7 fails to handle the layer weights saved in the nested list. OKLayer's weights (via print(model.weights)):
NGLayer's weights (via print(model.weights)):
|
Dear all,
I faced a bug.
Layers saved in a nested list are not saved/loaded correctly.
The following code can reproduce the bug.
If any other information is necessary, please let me know.
Best Regards,
Environments
Code
Result
The text was updated successfully, but these errors were encountered: