Skip to content

Issues during encoder training with wikisection data #14

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

Closed
zcsheng95 opened this issue Oct 12, 2022 · 3 comments
Closed

Issues during encoder training with wikisection data #14

zcsheng95 opened this issue Oct 12, 2022 · 3 comments

Comments

@zcsheng95
Copy link

zcsheng95 commented Oct 12, 2022

Hi Rose,

The issue occurred during the testing phase. The error message reads like:

ValueError: 'self.log(test_loss, [3.5329943])' was called, but 'ndarray' values cannot be logged

And it was raised from pytorch_lighting/core/lighting.py

Do you have any ideas?

Thanks so much!

Zhecheng

@rosewang2008
Copy link
Owner

Hey! Thanks for raising this to my attention. It seems like the current version logs a single-element list when it should just log the element.

FWIW, the current repository contains a lot of refactoring bugs which I've been diligently working through. I expect to release a cleaner, reproducible version in due time. Apologies for the wait!

@zcsheng95
Copy link
Author

Thanks Rose. Another one quick question: Is it feasible to get the parameters that characterizes the learned latent space from the current code?

@rosewang2008
Copy link
Owner

If you mean training the encoder, then technically yes---however there are a lot of bugs that I introduced in refactoring, eg if-conditions being incorrect mentioned here: #10

There are some other issues I'm resolving through the process of reproducing this work (eg. fixing the seed for data loading). I plan to release the correct, reproducible version of this work once I ensure that all the results can be perfectly replicated. Again, apologies for the wait!

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