feat: ZeroInflatedRegressor.score_samples(...)
#680
Conversation
| with pytest.raises(AttributeError, match="This 'ZeroInflatedRegressor' has no attribute 'score_samples'"): | ||
| zir.score_samples(X) |
There was a problem hiding this comment.
Wondering if this is the best error message to give to the user - automatically generated from available_if decorator
docs/user-guide/meta-models.md
Outdated
| If the underlying classifier is able to predict the _probability_ of a sample to be zero (i.e. it implements a `predict_proba` method), then the `ZeroInflatedRegressor` can be used to predict the probability of a sample being non-zero times the expected value of such sample. | ||
|
|
||
| This quantity is sometimes called _risk estimate_ or _expected impact_, however, to adhere to scikit-learn convention, we made it accessible via the `score_samples` method. |
There was a problem hiding this comment.
cc: @edgBR
Is this description reasonable to you? Any input is more than welcomed
There was a problem hiding this comment.
Hi guys, the description looks clear to me. However I will add that it can be called risk only if the classifier is calibrated.
There was a problem hiding this comment.
I will add a warning to mention that classifier should be calibrated
There was a problem hiding this comment.
Did not spot anything out of the ordinary, feels cool to merge to me.
@FBruzzesi let me know if you merge this because it feels like it would be ready for a new release as well.
|
I was out on vacation for the weekend. Merging now 😁 |
Description
Implements
score_samplesforZeroInflatedRegressoras discusses in #585Fixes #585
Type of change
Checklist: