[MRG+1] Added sample_weight parameter to ransac.fit #6140
Conversation
|
Please make sure no-one else is working on it, before issuing a pull request next time. |
| @@ -243,6 +244,17 @@ def fit(self, X, y): | |||
| except ValueError: | |||
| pass | |||
|
|
|||
| fit_parameters = signature(base_estimator.fit).parameters | |||
| estimator_name = type(base_estimator).__name__ | |||
There was a problem hiding this comment.
you can reuse has_fit_parameter from sklearn.utils.validation
|
Can you add a test, testing this new feature? |
…d raise an exception when sample_weight is passed to base_estimators that dont support it.
|
Thanks for the feedback @MechCoder , |
| @@ -195,6 +195,10 @@ def fit(self, X, y): | |||
| `is_data_valid` and `is_model_valid` return False for all | |||
| `max_trials` randomly chosen sub-samples. | |||
|
|
|||
| TypeError | |||
There was a problem hiding this comment.
you should document sample_weight as well
There was a problem hiding this comment.
Also, can you remove this block and document it under sample_weight?, i.e
sample_weight: array-like,
blah, blah.. raises Error if balh blah
|
I have modified the tests as you suggested , they all passed. I'm still not sure about considering sample_weight when sampling ie subset_idxs |
| @@ -188,6 +188,10 @@ def fit(self, X, y): | |||
| y : array-like, shape = [n_samples] or [n_samples, n_targets] | |||
| Target values. | |||
|
|
|||
| sample_weight: numpy array of shape [n_samples] | |||
There was a problem hiding this comment.
numpy array -> array-like
Also indent the next line just 4 spaces to the first line.
Look at how all the other parameters are documented
|
@imaculate please run pep8 on both the files. (just the above cosmetic comments and that's it from me) |
|
Thanks .. Done! |
|
|
||
| assert_almost_equal(ransac_estimator.estimator_.coef_, ref_coef_) | ||
|
|
||
| """ check that if base_estimator.fit doesn't support |
There was a problem hiding this comment.
just make this a comment starting with #
|
Done. |
|
LGTM. Well Done. |
|
Merged as 41526cb (after updating whatsnew). Thanks 🍻 🍷 !! |
|
Thanks @MechCoder @ahojnnes .First contribution! 🍻 . |
|
🍻 :) |
Resolved issue #6113 and #5871.
For estimators that used sample_weight parameter in the fit method, it was requested that similar should be applied to ransacRegressor. This has been intergrated.