make binning inplace optional

Blue-Yonder-OSS · Dec 9, 2023 · c969ff7 · c969ff7
1 parent 81c8dcf
commit c969ff7
Show file tree

Hide file tree

Showing 4 changed files with 77 additions and 69 deletions.
diff --git a/cyclic_boosting/binning/bin_number_transformer.py b/cyclic_boosting/binning/bin_number_transformer.py
@@ -124,13 +124,15 @@ def __init__(
         weight_column=None,
         epsilon=1e-9,
         tolerance=0.1,
+        inplace=False,
     ):
         self.n_bins = n_bins
         self.feature_properties = feature_properties
         self.weight_column = weight_column
         self.epsilon = epsilon
         self.tolerance = tolerance
         self.nan_representation = MISSING_VALUE_AS_BINNO
+        self.inplace = inplace
         ECdfTransformer.__init__(
             self,
             n_bins=self.n_bins,
@@ -172,8 +174,13 @@ def is_finite(x):
             xt = MISSING_VALUE_AS_BINNO
         return xt
 
-    def transform(self, X, y=None):
-        self._check_input_for_transform(X)
+    def transform(self, X_orig, y=None):
+        self._check_input_for_transform(X_orig)
+
+        if not self.inplace:
+            X = X_orig.copy()
+        else:
+            X = X_orig
 
         if check_frame_empty(X):
             if isinstance(X, pd.DataFrame):

diff --git a/cyclic_boosting/pipelines.py b/cyclic_boosting/pipelines.py
@@ -48,6 +48,7 @@ def pipeline_CB(
     regalpha=0.0,
     quantile=None,
     costs=None,
+    inplace=False,
 ):
     if estimator in [CBPoissonRegressor, CBLocPoissonRegressor, CBLocationRegressor, CBClassifier]:
         estimatorCB = estimator(
@@ -171,7 +172,7 @@ def pipeline_CB(
         )
     else:
         raise Exception("No valid CB estimator.")
-    binner = binning.BinNumberTransformer(n_bins=number_of_bins, feature_properties=feature_properties)
+    binner = binning.BinNumberTransformer(n_bins=number_of_bins, feature_properties=feature_properties, inplace=inplace)
 
     return Pipeline([("binning", binner), ("CB", estimatorCB)])
 

diff --git a/docs/source/tutorial.md b/docs/source/tutorial.md
@@ -155,16 +155,16 @@ from cyclic_boosting.pipelines import pipeline_CBMultiplicativeQuantileRegressor
 from cyclic_boosting.quantile_matching import J_QPD_S
 
 CB_est_qlow = pipeline_CBMultiplicativeQuantileRegressor(quantile=0.2)
-CB_est_qlow.fit(X_train.copy(), y)
-yhat_qlow = CB_est_qlow.predict(X_test.copy())
+CB_est_qlow.fit(X_train, y)
+yhat_qlow = CB_est_qlow.predict(X_test)
 
 CB_est_qmedian = pipeline_CBMultiplicativeQuantileRegressor(quantile=0.5)
-CB_est_qmedian.fit(X_train.copy(), y)
-yhat_qmedian = CB_est_qmedian.predict(X_test.copy())
+CB_est_qmedian.fit(X_train, y)
+yhat_qmedian = CB_est_qmedian.predict(X_test)
 
 CB_est_qhigh = pipeline_CBMultiplicativeQuantileRegressor(quantile=0.8)
-CB_est_qhigh.fit(X_train.copy(), y)
-yhat_qhigh = CB_est_qhigh.predict(X_test.copy())
+CB_est_qhigh.fit(X_train, y)
+yhat_qhigh = CB_est_qhigh.predict(X_test)
 
 j_qpd_s_42 = J_QPD_S(0.2, yhat_qlow[42], yhat_qmedian[42], yhat_qhigh[42])
 yhat_42_percentile95 = j_qpd_s_42.ppf(0.95)