docs

docs/api/preprocessing.md

@@ -69,3 +69,8 @@
     options:
         show_root_full_path: true
         show_root_heading: true
+
+:::sklego.preprocessing.monotonicspline.MonotonicSplineTransformer
+    options:
+        show_root_full_path: true
+        show_root_heading: true

readme.md

@@ -125,6 +125,7 @@ Here's a list of features that this library currently offers:
 - `sklego.preprocessing.RepeatingBasisFunction` repeating feature engineering, useful for timeseries
 - `sklego.preprocessing.DictMapper` assign numeric values on categorical columns
 - `sklego.preprocessing.OutlierRemover` experimental method to remove outliers during training
+- `sklego.preprocessing.MonotonicSplineTransformer` re-uses `SplineTransformer` in an attempt to make monotonic features
 - `sklego.model_selection.GroupTimeSeriesSplit` timeseries Kfold for groups with different amount of observations per group
 - `sklego.model_selection.KlusterFoldValidation` experimental feature that does K folds based on clustering
 - `sklego.model_selection.TimeGapSplit` timeseries Kfold with a gap between train/test

tests/test_preprocessing/test_monospline.py

@@ -0,0 +1,20 @@
+import numpy as np
+import pytest
+
+from sklego.preprocessing import MonotonicSplineTransformer
+
+
+@pytest.mark.parametrize("n_knots", [3, 5])
+@pytest.mark.parametrize("degree", [3, 5])
+@pytest.mark.parametrize("knots", ["uniform", "quantile"])
+def test_monotonic_spline_transformer(n_knots, degree, knots):
+    X = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
+    transformer = MonotonicSplineTransformer(n_knots=n_knots, degree=degree, knots=knots)
+    transformer.fit(X)
+    out = transformer.transform(X)
+    # Check each column is monotonically increasing
+    for col in range(out.shape[1]):
+        col_values = out[:, col]
+        # numpy diff returns positive values if array is increasing
+        differences = np.diff(col_values)
+        assert np.all(differences >= 0), f"Column {col} is not monotonically increasing"