init

mllib/src/main/resources/META-INF/services/org.apache.spark.ml.Transformer

@@ -18,6 +18,7 @@
 # Spark Connect ML uses ServiceLoader to find out the supported Spark Ml non-model transformer.
 # So register the supported transformer here if you're trying to add a new one.
 ########### Transformers
+org.apache.spark.ml.feature.DCT
 org.apache.spark.ml.feature.VectorAssembler
 
 ########### Model for loading

python/pyspark/ml/tests/test_feature.py

@@ -23,6 +23,7 @@
 import numpy as np
 
 from pyspark.ml.feature import (
+    DCT,
     Binarizer,
     CountVectorizer,
     CountVectorizerModel,
@@ -56,6 +57,34 @@
 
 
 class FeatureTestsMixin:
+    def test_dct(self):
+        df = self.spark.createDataFrame([(Vectors.dense([5.0, 8.0, 6.0]),)], ["vec"])
+        dct = DCT()
+        dct.setInverse(False)
+        dct.setInputCol("vec")
+        dct.setOutputCol("resultVec")
+
+        self.assertFalse(dct.getInverse())
+        self.assertEqual(dct.getInputCol(), "vec")
+        self.assertEqual(dct.getOutputCol(), "resultVec")
+
+        output = dct.transform(df)
+        self.assertEqual(output.columns, ["vec", "resultVec"])
+        self.assertEqual(output.count(), 1)
+        self.assertTrue(
+            np.allclose(
+                output.head().resultVec.toArray(),
+                [10.96965511, -0.70710678, -2.04124145],
+                atol=1e-4,
+            )
+        )
+
+        # save & load
+        with tempfile.TemporaryDirectory(prefix="dct") as d:
+            dct.write().overwrite().save(d)
+            dct2 = DCT.load(d)
+            self.assertEqual(str(dct), str(dct2))
+
     def test_string_indexer(self):
         df = (
             self.spark.createDataFrame(