sklearn.metrics
.make_scorer¶
- sklearn.metrics.make_scorer(score_func, *, response_method=None, greater_is_better=True, needs_proba='deprecated', needs_threshold='deprecated', **kwargs)[source]¶
Make a scorer from a performance metric or loss function.
A scorer is a wrapper around an arbitrary metric or loss function that is called with the signature
scorer(estimator, X, y_true, **kwargs)
.It is accepted in all scikit-learn estimators or functions allowing a
scoring
parameter.The parameter
response_method
allows to specify which method of the estimator should be used to feed the scoring/loss function.Read more in the User Guide.
- Parameters:
- score_funccallable
Score function (or loss function) with signature
score_func(y, y_pred, **kwargs)
.- response_method{“predict_proba”, “decision_function”, “predict”} or list/tuple of such str, default=None
Specifies the response method to use get prediction from an estimator (i.e. predict_proba, decision_function or predict). Possible choices are:
if
str
, it corresponds to the name to the method to return;if a list or tuple of
str
, it provides the method names in order of preference. The method returned corresponds to the first method in the list and which is implemented byestimator
.if
None
, it is equivalent to"predict"
.
New in version 1.4.
- greater_is_betterbool, default=True
Whether
score_func
is a score function (default), meaning high is good, or a loss function, meaning low is good. In the latter case, the scorer object will sign-flip the outcome of thescore_func
.- needs_probabool, default=False
Whether
score_func
requirespredict_proba
to get probability estimates out of a classifier.If True, for binary
y_true
, the score function is supposed to accept a 1Dy_pred
(i.e., probability of the positive class, shape(n_samples,)
).Deprecated since version 1.4:
needs_proba
is deprecated in version 1.4 and will be removed in 1.6. Useresponse_method="predict_proba"
instead.- needs_thresholdbool, default=False
Whether
score_func
takes a continuous decision certainty. This only works for binary classification using estimators that have either adecision_function
orpredict_proba
method.If True, for binary
y_true
, the score function is supposed to accept a 1Dy_pred
(i.e., probability of the positive class or the decision function, shape(n_samples,)
).For example
average_precision
or the area under the roc curve can not be computed using discrete predictions alone.Deprecated since version 1.4:
needs_threshold
is deprecated in version 1.4 and will be removed in 1.6. Useresponse_method=("decision_function", "predict_proba")
instead to preserve the same behaviour.- **kwargsadditional arguments
Additional parameters to be passed to
score_func
.
- Returns:
- scorercallable
Callable object that returns a scalar score; greater is better.
Examples
>>> from sklearn.metrics import fbeta_score, make_scorer >>> ftwo_scorer = make_scorer(fbeta_score, beta=2) >>> ftwo_scorer make_scorer(fbeta_score, response_method='predict', beta=2) >>> from sklearn.model_selection import GridSearchCV >>> from sklearn.svm import LinearSVC >>> grid = GridSearchCV(LinearSVC(), param_grid={'C': [1, 10]}, ... scoring=ftwo_scorer)
Examples using sklearn.metrics.make_scorer
¶
Features in Histogram Gradient Boosting Trees
Prediction Intervals for Gradient Boosting Regression
Lagged features for time series forecasting
Demonstration of multi-metric evaluation on cross_val_score and GridSearchCV