lindong28 commented on a change in pull request #29: URL: https://github.com/apache/flink-ml/pull/29#discussion_r747186614
########## File path: flink-ml-lib/src/main/java/org/apache/flink/ml/operator/BaseAlgoImpl.java ########## @@ -0,0 +1,98 @@ +/* + * 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.operator; + +import org.apache.flink.ml.api.core.AlgoOperator; +import org.apache.flink.ml.param.Param; +import org.apache.flink.ml.param.WithParams; +import org.apache.flink.ml.util.ParamUtils; +import org.apache.flink.table.api.Table; + +import java.util.HashMap; +import java.util.Map; + +/** + * Base class for algorithm operators. + * + * <p>Base class for the algorithm operators. It hosts the parameters and output tables of an + * algorithm operator. Each BaseAlgoImpl may have one or more output tables. One of the output table + * is the primary output table which can be obtained by calling {@link #getOutput}. The other output + * tables are side output tables that can be obtained by calling {@link #getSideOutputs()}. + * + * <p>The input of an BaseAlgoImpl is defined in the subclasses of the BaseAlgoImpl. + * + * @param <T> The class type of the {@link BaseAlgoImpl} implementation itself + */ +public abstract class BaseAlgoImpl<T extends BaseAlgoImpl<T>> + implements AlgoOperator<T>, WithParams<T> { + + /** Params for algorithms. */ + private Map<Param<?>, Object> params; + + /** The table held by operator. */ + private transient Table output = null; + + /** The side outputs of operator that be similar to the stream's side outputs. */ + private transient Table[] sideOutputs = null; + + /** Construct the operator with the initial Params. */ + protected BaseAlgoImpl(Map<Param<?>, Object> params) { + this.params = new HashMap<>(); + if (null != params) { + for (Map.Entry<Param<?>, Object> entry : params.entrySet()) { + this.params.put(entry.getKey(), entry.getValue()); + } + } + ParamUtils.initializeMapWithDefaultValues(this.params, this); + } + + @Override + public Map<Param<?>, Object> getParamMap() { + return this.params; + } + + /** Returns the table held by operator. */ + public Table getOutput() { + return this.output; + } + + /** Returns the side outputs. */ + public Table[] getSideOutputs() { + return this.sideOutputs; + } + + /** + * Set the table held by operator. + * + * @param output the output table. + */ + protected void setOutput(Table output) { + this.output = output; + } + + /** + * Set the side outputs. + * + * @param sideOutputs the side outputs set the operator. + */ + protected void setSideOutputs(Table[] sideOutputs) { Review comment: Thanks for the explanation :) I agree there is case where differentiating between main output and side outputs could make the caller code simpler. On the other hand, there is also case this could make caller code more complex -- otherwise we would already different between main output and side outputs in FLIP-173, right? The pros/cons between these two design options apply to both FLIP-173 and this PR. So it is not clear why this PR uses side outputs but FLIP-173 does not. And in general, it is probably preferred to use same API flavor in a given project. What do you think? -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
