Adding inference support for next logit probability. #78
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
xrdaukar
reviewed
Jun 13, 2024
oelachqar
reviewed
Jun 13, 2024
src/lema/infer_prob.py
Outdated
| for acceptable_logit in acceptable_logits_enc: | ||
| acceptable_logit_prob = next_logit_probs[acceptable_logit] | ||
| acceptable_logit_probs.append( | ||
| acceptable_logit_prob.detach().cpu().numpy().item() |
There was a problem hiding this comment.
I'd recommend using the inference_mode or no_grad decorator. That'll avoid the need to manually detach + inference will be faster
I also just noticed that we do not set the model.eval() in our inference scripts, we should add that as well
There was a problem hiding this comment.
no_grad : Done.
model.eval() in our inference scripts --> can you elaborate?
wizeng23
reviewed
Jun 13, 2024
OPE-35 Add in-training evaluation options
HuggingFace trainers already support evaluation every N training steps. The goal of this task is to enable configuring a list of datasets to evaluate intermediate checkpoints. |
xrdaukar
reviewed
Jun 14, 2024
xrdaukar
reviewed
Jun 14, 2024
xrdaukar
reviewed
Jun 16, 2024
xrdaukar
reviewed
Jun 16, 2024
xrdaukar
reviewed
Jun 16, 2024
xrdaukar
reviewed
Jun 16, 2024
xrdaukar
approved these changes
Jun 16, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Contributes to OPE-35