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

Unable to train Word2Box: the script gives error messages #4

Open
ChiaraBi opened this issue Feb 4, 2024 · 0 comments
Open

Unable to train Word2Box: the script gives error messages #4

ChiaraBi opened this issue Feb 4, 2024 · 0 comments

Comments

@ChiaraBi
Copy link

ChiaraBi commented Feb 4, 2024

Hi, I am trying to train the model Word2Box using the following command:

language-modeling-with-boxes train \
 --model_type Word2Box \
 --int_temp 0.0 \
 --loss_fn nce \
 --box_type BoxTensor \
 --dataset ptb \
 --num_epochs 2 --log_frequency 10 
  • If int_temp != 0.0 (I am using the default value 0.01), it gives the following error message: "Gumbel intersection is not possible".
  • If int_temp == 0.0, it gives the following error message: "The size of tensor a (681) must match the size of tensor b (8) at non-singleton dimension 1".

What value should I use for int_temp to make the script work? Is there any other parameter I need to use?

Thank you in advance!

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