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.

Sign up for GitHub

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

Jump to bottom

Fixed non-scalar (text, vector, set) output feature explanations #3269

Merged
merged 2 commits into from
Mar 18, 2023
Merged

Fixed non-scalar (text, vector, set) output feature explanations #3269

merged 2 commits into from
Mar 18, 2023

Conversation

tgaddair
Copy link
Collaborator

No description provided.

@tgaddair tgaddair added the bug Something isn't working label Mar 18, 2023
arnavgarg1
arnavgarg1 reviewed Mar 18, 2023
View reviewed changes
ludwig/explain/captum.py
Comment on lines +121 to +122
# If the target feature is a non-scalar type (vector, set, etc.), sum it to get a scalar value.
# https://github.com/pytorch/captum/issues/377
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

QQ: Does this change the interpretation/explanation for these kinds of features?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be different from looking at each element of the output individually, yes, but it's difficult to reason about what that alternative means for things like vectors.

justinxzhao
justinxzhao approved these changes Mar 18, 2023
View reviewed changes
Copy link
Collaborator

@justinxzhao justinxzhao left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seems reasonable to me! Thanks for finding and linking to a relevant discussion on the captum project.

@github-actions
Copy link

github-actions bot commented Mar 18, 2023 •
edited

Unit Test Results

  5 files   -     1    5 suites   - 1   17m 33s ⏱️ - 1h 32m 10s
12 tests  - 146  10 ✔️  - 131    2 💤  - 15  0 ❌ ±0 
52 runs   - 146  42 ✔️  - 131  10 💤  - 15  0 ❌ ±0 

Results for commit 29da637. ± Comparison against base commit 6fc4713.

This pull request removes 146 tests. 
tests.integration_tests.test_automl ‑ test_auto_train
tests.integration_tests.test_automl ‑ test_autoconfig_preprocessing_balanced
tests.integration_tests.test_automl ‑ test_autoconfig_preprocessing_imbalanced
tests.integration_tests.test_automl ‑ test_autoconfig_preprocessing_text_image
tests.integration_tests.test_automl ‑ test_create_auto_config[image]
tests.integration_tests.test_automl ‑ test_create_auto_config[multimodal]
tests.integration_tests.test_automl ‑ test_create_auto_config[tabular_large]
tests.integration_tests.test_automl ‑ test_create_auto_config[tabular_small]
tests.integration_tests.test_automl ‑ test_create_auto_config[text]
tests.integration_tests.test_automl ‑ test_create_auto_config_with_dataset_profile
…

♻️ This comment has been updated with latest results.

@tgaddair tgaddair changed the title Fixed set and vector output feature explanations Fixed non-scalar (text, vector, set) output feature explanations Mar 18, 2023
@tgaddair tgaddair merged commit a265fc2 into master Mar 18, 2023
@tgaddair tgaddair deleted the set-explain branch March 18, 2023 13:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@arnavgarg1 arnavgarg1 arnavgarg1 left review comments

@justinxzhao justinxzhao justinxzhao approved these changes

@jppgks jppgks jppgks approved these changes

Assignees
No one assigned
Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
4 participants
@tgaddair @justinxzhao @jppgks @arnavgarg1