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

Low-priority-but-easy: Log progress during the final evaluation. #145

Closed
sleepinyourhat opened this issue Jul 12, 2018 · 1 comment
Closed
Labels
jiant-v1-legacy Relevant to versions <= v1.3.2 low-priority Only if you're bored. Ask Sam/Ian/Alex before starting.

Comments

@sleepinyourhat
Copy link
Contributor

Evaluating on the GLUE test set on a K80 can take over an hour, and there's no logging during that time. At least log when each task is started/finished?

@sleepinyourhat sleepinyourhat added the low-priority Only if you're bored. Ask Sam/Ian/Alex before starting. label Jul 12, 2018
@W4ngatang
Copy link
Collaborator

Closed with #234

@jeswan jeswan added the jiant-v1-legacy Relevant to versions <= v1.3.2 label Sep 17, 2020
jeswan added a commit that referenced this issue Sep 23, 2020
* update _create_examples to store all example identifiers in guid

* add method to format preds for SuperGLUE leaderboard submission

* add script to translate SuperGLUE task raw preds to leaderboard submission format

* improve description of script

* use diagnostic task names specified in tasks/constants.py

* print absolute target path for outputs

* add prediction formatter functionality to RTE task class

* add SuperGLUE prediction formatter to CB task class

* add SuperGLUE prediction formatter to COPA task class

* add SuperGLUE prediction formatter to BoolQ task class

* add SuperGLUE prediction formatter to WiC task class

* add SuperGLUE prediction formatter to WSC task class

* added SuperGlueMixin for prediction formatting

Co-authored-by: Phil Yeres <6176602+pyeres@users.noreply.github.com>
leo-liuzy pushed a commit to leo-liuzy/dynamic_jiant that referenced this issue Nov 11, 2020
)

* update _create_examples to store all example identifiers in guid

* add method to format preds for SuperGLUE leaderboard submission

* add script to translate SuperGLUE task raw preds to leaderboard submission format

* improve description of script

* use diagnostic task names specified in tasks/constants.py

* print absolute target path for outputs

* add prediction formatter functionality to RTE task class

* add SuperGLUE prediction formatter to CB task class

* add SuperGLUE prediction formatter to COPA task class

* add SuperGLUE prediction formatter to BoolQ task class

* add SuperGLUE prediction formatter to WiC task class

* add SuperGLUE prediction formatter to WSC task class

* added SuperGlueMixin for prediction formatting

Co-authored-by: Phil Yeres <6176602+pyeres@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
jiant-v1-legacy Relevant to versions <= v1.3.2 low-priority Only if you're bored. Ask Sam/Ian/Alex before starting.
Projects
None yet
Development

No branches or pull requests

3 participants