Skip to content
This repository has been archived by the owner on Jul 24, 2024. It is now read-only.

BUG Satisfy scikit-learn Estimator checks #51

Merged
merged 5 commits into from
Mar 8, 2019

Conversation

stephen-hoover
Copy link
Contributor

@stephen-hoover stephen-hoover commented Mar 8, 2019

The scikit-learn check_estimator function appears to have become more comprehensive in the past year. Add fixes for the things that have changed. Includes

Note that this PR does hide the previously public cv attribute, but that attribute was introduced in PR #24 which has not yet been released.

Closes #25 , #38

Stephen Hoover added 5 commits March 8, 2019 12:00
It looks like the calling parameters for `sklearn.utils.testing.ignore_warnings` have changed. Call it with a keyword argument instead of relying on a positional argument.
For full scikit-learn compatibility, we shouldn't set anything in the __init__ except for parameters passed into the `__init__` function. The `cv` attribute is a private internal attribute, so mark it as such.
Satisfy the scikit-learn estimator checks: Correctly handle list inputs for `sample_weights`.
This was failing on the scikit-learn check_estimator. Bring in the fix from civisanalytics#26.
Copy link

@elsander elsander left a comment

Choose a reason for hiding this comment

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

LGTM!

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

model.fit fails due to reshaping error in predict_proba()
2 participants