Github user yanboliang commented on a diff in the pull request:
https://github.com/apache/spark/pull/19172#discussion_r137929197
--- Diff: python/pyspark/ml/tests.py ---
@@ -1655,6 +1655,25 @@ def
test_multinomial_logistic_regression_with_bound(self):
np.allclose(model.interceptVector.toArray(), [-0.9057,
-1.1392, -0.0033], atol=1E-4))
+class MultilayerPerceptronClassifierTest(SparkSessionTestCase):
+
+ def test_multilayer_perceptron_classifier(self):
+
+ data_path = "data/mllib/sample_multiclass_classification_data.txt"
--- End diff --
For other algorithms, I agree with you. However, too simple dataset is not
ideal to test MLPC. This dataset is a tiny dataset with 150 rows and 4 columns,
so I think we can use it. Thanks.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
