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

compute_ate.py has problems #5

Open
yang840808206 opened this issue Dec 12, 2019 · 3 comments
Open

compute_ate.py has problems #5

yang840808206 opened this issue Dec 12, 2019 · 3 comments

Comments

@yang840808206
Copy link

Dear :
I am doing a peerread experiment now.But BERT only outputs a test_result.tsv,and use this file for compute_ate, I think the code is buggy and cannot run successfully.
For example,ate.psi_very_naive don't have a return value.
could you update the correct code?
thanks

@vveitch
Copy link
Contributor

vveitch commented Dec 12, 2019

Can you elaborate on exactly what you're doing and what problem you're seeing?

@yang840808206
Copy link
Author

I don't know how the .tsv file needed for compute_ate.py or compute_att.py is obtained, and I want to know how the effect is calculated from the tsv file, but without the .tsv file, I can't understand.

And your psi_very_naive.py function in ate.py has no return value.

@vveitch
Copy link
Contributor

vveitch commented Apr 5, 2020

First, sorry for the very late reply---I'd completely forgotten there was an open issue here.

Second, the required tsv is output by the training code (it's the predictions of the trained model). I'm still confused what the confusion is. But,

Third, we made a substantially streamlined version of the code in tf2, which will perhaps resolve your problem: https://github.com/vveitch/causal-text-embeddings-tf2

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