Merge pull request h2oai#16003 from h2oai/sy-#15988-Isotonic

h2oaiGH-15988: Craft Isotonic Regression Examples

Author: shaunyogeshwaran

h2o-bindings/bin/custom/python/gen_isotonicregression.py

@@ -0,0 +1,19 @@
+examples = dict(
+    out_of_bounds="""
+>>> import h2o
+>>> from h2o import H2OFrame
+>>> from h2o.estimators.isotonicregression import H2OIsotonicRegressionEstimator
+>>> from sklearn.datasets import make_regression
+>>> import numpy as np
+>>> h2o.init()
+>>> X, y = make_regression(n_samples=10000, n_features=1, random_state=41, noise=0.8)
+>>> X = X.reshape(-1)
+>>> train = H2OFrame(np.column_stack((y, X)), column_names=["y", "X"])
+>>> w_values = np.random.rand(train.shape[0])
+>>> w_frame = H2OFrame(w_values.reshape(-1, 1), column_names=["w"])
+>>> train = train.cbind(w_frame)
+>>> h2o_iso_reg = H2OIsotonicRegressionEstimator(out_of_bounds="clip")
+>>> h2o_iso_reg.train(training_frame=train, x="X", y="y")
+>>> h2o_iso_reg.predict(train)
+"""
+)

h2o-py/h2o/estimators/isotonicregression.py

@@ -184,6 +184,24 @@ def out_of_bounds(self):
         Method of handling values of X predictor that are outside of the bounds seen in training.
 
         Type: ``Literal["na", "clip"]``, defaults to ``"na"``.
+
+        :examples:
+
+        >>> import h2o
+        >>> from h2o import H2OFrame
+        >>> from h2o.estimators.isotonicregression import H2OIsotonicRegressionEstimator
+        >>> from sklearn.datasets import make_regression
+        >>> import numpy as np
+        >>> h2o.init()
+        >>> X, y = make_regression(n_samples=10000, n_features=1, random_state=41, noise=0.8)
+        >>> X = X.reshape(-1)
+        >>> train = H2OFrame(np.column_stack((y, X)), column_names=["y", "X"])
+        >>> w_values = np.random.rand(train.shape[0])
+        >>> w_frame = H2OFrame(w_values.reshape(-1, 1), column_names=["w"])
+        >>> train = train.cbind(w_frame)
+        >>> h2o_iso_reg = H2OIsotonicRegressionEstimator(out_of_bounds="clip")
+        >>> h2o_iso_reg.train(training_frame=train, x="X", y="y")
+        >>> h2o_iso_reg.predict(train)
         """
         return self._parms.get("out_of_bounds")