How to use the mleap.sklearn.preprocessing.data.Binarizer function in mleap

def binarizer_test(self):

        extract_features = ['a']
        feature_extractor = FeatureExtractor(input_scalars=['a'],
                                         output_vector='extracted_a_output',
                                         output_vector_items=["{}_out".format(x) for x in extract_features])

        binarizer = Binarizer(threshold=0.0)
        binarizer.mlinit(prior_tf=feature_extractor,
                         output_features='a_binary')

        Xres = binarizer.fit_transform(self.df[['a']])

        # Test that the binarizer functions as expected
        self.assertEqual(float(len(self.df[self.df.a >= 0]))/10.0, Xres.mean())

        binarizer.serialize_to_bundle(self.tmp_dir, binarizer.name)

        expected_model = {
          "op": "binarizer",
          "attributes": {
            "threshold": {
              "double": 0.0
            }

output_vector_items=["{}_out".format(x) for x in extract_features])

        binarizer = Binarizer(threshold=0.0)
        binarizer.mlinit(prior_tf=feature_extractor,
                         output_features='a_binary')

        Xres = binarizer.fit_transform(self.df[['a']])

        # Test that the binarizer functions as expected
        self.assertEqual(float(len(self.df[self.df.a >= 0]))/10.0, Xres.mean())

        binarizer.serialize_to_bundle(self.tmp_dir, binarizer.name)

        # Deserialize the Binarizer
        node_name = "{}.node".format(binarizer.name)
        binarizer_tf_ds = Binarizer()
        binarizer_tf_ds.deserialize_from_bundle(self.tmp_dir, node_name)

        # Transform some sample data
        res_a = binarizer.transform(self.df[['a']])
        res_b = binarizer_tf_ds.transform(self.df[['a']])

        self.assertEqual(res_a[0][0], res_b[0][0])
        self.assertEqual(res_a[1][0], res_b[1][0])
        self.assertEqual(res_a[2][0], res_b[2][0])
        self.assertEqual(res_a[3][0], res_b[3][0])

def binarizer_deserializer_test(self):

        extract_features = ['a']
        feature_extractor = FeatureExtractor(input_scalars=['a'],
                                         output_vector='extracted_a_output',
                                         output_vector_items=["{}_out".format(x) for x in extract_features])

        binarizer = Binarizer(threshold=0.0)
        binarizer.mlinit(prior_tf=feature_extractor,
                         output_features='a_binary')

        Xres = binarizer.fit_transform(self.df[['a']])

        # Test that the binarizer functions as expected
        self.assertEqual(float(len(self.df[self.df.a >= 0]))/10.0, Xres.mean())

        binarizer.serialize_to_bundle(self.tmp_dir, binarizer.name)

        # Deserialize the Binarizer
        node_name = "{}.node".format(binarizer.name)
        binarizer_tf_ds = Binarizer()
        binarizer_tf_ds.deserialize_from_bundle(self.tmp_dir, node_name)

        # Transform some sample data