import pandas as pd
from evalml.pipelines import PipelineBase
[docs]class ClassificationPipeline(PipelineBase):
"""Pipeline subclass for all classification pipelines."""
[docs] def predict_proba(self, X):
"""Make probability estimates for labels.
Arguments:
X (pd.DataFrame or np.array) : data of shape [n_samples, n_features]
Returns:
pd.DataFrame : probability estimates
"""
if not isinstance(X, pd.DataFrame):
X = pd.DataFrame(X)
X = self._transform(X)
proba = self.estimator.predict_proba(X)
return proba