lindong28 commented on a change in pull request #28: URL: https://github.com/apache/flink-ml/pull/28#discussion_r754839066
########## File path: flink-ml-lib/src/main/java/org/apache/flink/ml/common/param/HasL2.java ########## @@ -0,0 +1,38 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.ml.common.param; + +import org.apache.flink.ml.param.DoubleParam; +import org.apache.flink.ml.param.Param; +import org.apache.flink.ml.param.ParamValidators; +import org.apache.flink.ml.param.WithParams; + +/** Interface for the shared L2 regularization param. */ +public interface HasL2<T> extends WithParams<T> { Review comment: Ideally we can make the parameter future proof. According to the scikit-learn doc [1], it looks like the regularization can be one of `l1`, `l2` and `elasticnet`. And scikit-learn supports all choices. Though Spark provides only the `HasElasticNetParam` without explicit `l1` or `l2` choices, the parameter doc suggests that `l1` or `l2` regularization is effectively used if user sets the parameter value to be `1` or `2`. So both scikit-learn and Spark support all three modes. I guess we also want to be able to support these three modes in Flink ML, even if we support only one for now. If we add `HasL2` here, how do we expect users to specify `L1` and `elasticnet` mode in the future? Should we use a double-valued `HasElasticNetParam` like Spark, or use a string-valued `HasPenalty` similar to Scikit-learn? [1] https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html [1] https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
