lindong28 commented on a change in pull request #52: URL: https://github.com/apache/flink-ml/pull/52#discussion_r832739713
########## File path: flink-ml-lib/src/test/java/org/apache/flink/ml/feature/stringindexer/IndexToStringModelTest.java ########## @@ -0,0 +1,165 @@ +/* + * 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.feature.stringindexer; + +import org.apache.flink.api.common.restartstrategy.RestartStrategies; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.ml.util.StageTestUtils; +import org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions; +import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment; +import org.apache.flink.table.api.Table; +import org.apache.flink.table.api.bridge.java.StreamTableEnvironment; +import org.apache.flink.test.util.AbstractTestBase; +import org.apache.flink.types.Row; + +import org.apache.commons.collections.IteratorUtils; +import org.junit.Before; +import org.junit.Rule; +import org.junit.Test; +import org.junit.rules.TemporaryFolder; + +import java.util.Arrays; +import java.util.Collections; +import java.util.List; + +import static org.junit.Assert.assertArrayEquals; +import static org.junit.Assert.assertEquals; +import static org.junit.Assert.fail; + +/** Tests the {@link IndexToStringModel}. */ +public class IndexToStringModelTest extends AbstractTestBase { + @Rule public final TemporaryFolder tempFolder = new TemporaryFolder(); + + private StreamExecutionEnvironment env; + private StreamTableEnvironment tEnv; + private Table predictTable; + private Table modelTable; + private Table predictTableWithUnseenValues; + + private final List<Row> expectedPrediction = + Arrays.asList(Row.of(0, 3, "a", "2.0"), Row.of(1, 2, "b", "1.0")); + private final String[][] stringArrays = + new String[][] {{"a", "b", "c", "d"}, {"-1.0", "0.0", "1.0", "2.0"}}; + + @Before + public void before() { + Configuration config = new Configuration(); + config.set(ExecutionCheckpointingOptions.ENABLE_CHECKPOINTS_AFTER_TASKS_FINISH, true); + env = StreamExecutionEnvironment.getExecutionEnvironment(config); + env.setParallelism(4); + env.enableCheckpointing(100); + env.setRestartStrategy(RestartStrategies.noRestart()); + tEnv = StreamTableEnvironment.create(env); + + modelTable = + tEnv.fromDataStream(env.fromElements(new StringIndexerModelData(stringArrays))) + .as("stringArrays"); + predictTable = + tEnv.fromDataStream(env.fromCollection(Arrays.asList(Row.of(0, 3), Row.of(1, 2)))) + .as("inputCol1", "inputCol2"); + predictTableWithUnseenValues = + tEnv.fromDataStream( + env.fromCollection( + Arrays.asList(Row.of(0, 3), Row.of(1, 2), Row.of(4, 1)))) + .as("inputCol1", "inputCol2"); + } + + @Test + public void testPredictParam() { + IndexToStringModel indexToStringModel = + new IndexToStringModel() + .setInputCols("inputCol1", "inputCol2") + .setOutputCols("outputCol1", "outputCol2") + .setModelData(modelTable); + Table output = indexToStringModel.transform(predictTable)[0]; + assertEquals( + Arrays.asList("inputCol1", "inputCol2", "outputCol1", "outputCol2"), + output.getResolvedSchema().getColumnNames()); + } + + @Test + public void testInputWithUnseenValues() { + IndexToStringModel indexToStringModel = + new IndexToStringModel() + .setInputCols("inputCol1", "inputCol2") + .setOutputCols("outputCol1", "outputCol2") + .setModelData(modelTable); + Table output = indexToStringModel.transform(predictTableWithUnseenValues)[0]; + try { + IteratorUtils.toList(tEnv.toDataStream(output).executeAndCollect()); + fail(); + } catch (Exception e) { + assertEquals( + "The input contains unseen index: 4.", + e.getCause().getCause().getCause().getCause().getCause().getMessage()); + } + } + + @Test + @SuppressWarnings("unchecked") + public void testPredict() throws Exception { + IndexToStringModel indexToStringModel = + new IndexToStringModel() + .setInputCols("inputCol1", "inputCol2") + .setOutputCols("outputCol1", "outputCol2") + .setModelData(modelTable); + Table output = indexToStringModel.transform(predictTable)[0]; + List<Row> predictedResult = + IteratorUtils.toList(tEnv.toDataStream(output).executeAndCollect()); + StringIndexerTest.verifyPredictionResult(expectedPrediction, predictedResult); + } + + @Test + @SuppressWarnings("unchecked") + public void testSaveLoadAndPredict() throws Exception { Review comment: nits: The code seems a bit too crowded as there is no line break in this method body. Any chance we could make it less crowded? Maybe see `KMeansTest` or Spark's `KMeansSuite` for example? Same for other tests. ########## File path: flink-ml-lib/src/main/java/org/apache/flink/ml/feature/stringindexer/StringIndexerModelParams.java ########## @@ -0,0 +1,60 @@ +/* + * 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.feature.stringindexer; + +import org.apache.flink.ml.common.param.HasInputCols; +import org.apache.flink.ml.common.param.HasOutputCols; +import org.apache.flink.ml.param.Param; +import org.apache.flink.ml.param.ParamValidators; +import org.apache.flink.ml.param.StringParam; + +/** + * Params of {@link StringIndexerModel}. + * + * @param <T> The class type of this instance. + */ +public interface StringIndexerModelParams<T> extends HasInputCols<T>, HasOutputCols<T> { + String ERROR_INVALID = "error"; + String SKIP_INVALID = "skip"; + String KEEP_INVALID = "keep"; + /** + * Supported options and the corresponding behavior to handle invalid entries in + * StringIndexerModel are listed as follows. + * + * <ul> + * <li>error: raise an exception. + * <li>skip: filter out rows with bad values. + * <li>keep: put the invalid entries in a special bucket. Review comment: Would it be useful to mention what this `special bucket` is? For example, we can say the index of these invalid entries is the expected number of distinct values in this column. -- 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]
