def test_get_model_repr_pipeline_instance_model_in_pipeline(
pipeline_instance_model_in_pipeline, ):
model_repr = get_estimator_repr(pipeline_instance_model_in_pipeline)
assert model_repr.find("...") == -1
assert model_repr == pipeline_instance_model_in_pipeline.__repr__(
N_CHAR_MAX=10000).replace("\n", "").replace(" ", "")
def _score(self, method_caller, estimator, X, y_true, sample_weight=None):
"""Evaluate predicted target values for X relative to y_true.
Parameters
----------
method_caller : callable
Returns predictions given an estimator, method name, and other
arguments, potentially caching results.
estimator : object
Trained estimator to use for scoring. Must have a predict_proba
method; the output of that is used to compute the score.
X : array-like or sparse matrix
Test data that will be fed to estimator.predict.
y_true : array-like
Gold standard target values for X.
sample_weight : array-like
Sample weights.
Returns
-------
score : float
Score function applied to prediction of estimator on X.
"""
y_pred = estimator.predict(X)
estimator_repr = get_estimator_repr(estimator)
estimator_hash = generate_estimator_hash(estimator)
self._upsert_estimator_hash(estimator_repr, estimator_hash)
self._save_prediction(y_pred=y_pred,
estimator_label=estimator_hash,
y_true=y_true)
if y_pred.isna().any().any() or np.isinf(y_pred).any().any():
return np.nan
else:
if sample_weight is not None:
return self._sign * self._score_func(
y_true,
y_pred,
sample_weight=sample_weight,
**self._kwargs)
else:
return self._sign * self._score_func(y_true, y_pred, **
self._kwargs)
def test_get_model_repr_single_model(wrapper_instance):
model_repr = get_estimator_repr(wrapper_instance)
assert model_repr.find("...") == -1
assert model_repr == wrapper_instance.__repr__(N_CHAR_MAX=10000).replace(
"\n", "").replace(" ", "")
