[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r830722471
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r830722471
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829014907
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829729279
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829706772
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829680935
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829680935
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829681672
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829681672
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829680935
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829680935
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829675971
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,201 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.iteration.operator.OperatorStateUtils;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream features =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxValues =
+features.transform(
+"reduceInEachPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxValues,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minVector = iter.next();
+DenseVector maxVector = iter.next();
+out.collect(new
MinMaxScalerModelData(minVector, maxVector));
+}
+});
+
+MinMaxScalerModel model =
+new
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54: URL: https://github.com/apache/flink-ml/pull/54#discussion_r829675465 ## File path: flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java ## @@ -0,0 +1,201 @@ +/* + * 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.minmaxscaler; + +import org.apache.flink.api.common.functions.MapFunction; +import org.apache.flink.api.common.functions.RichMapPartitionFunction; +import org.apache.flink.api.common.state.ListState; +import org.apache.flink.api.common.state.ListStateDescriptor; +import org.apache.flink.iteration.operator.OperatorStateUtils; +import org.apache.flink.ml.api.Estimator; +import org.apache.flink.ml.common.datastream.DataStreamUtils; +import org.apache.flink.ml.linalg.DenseVector; +import org.apache.flink.ml.param.Param; +import org.apache.flink.ml.util.ParamUtils; +import org.apache.flink.ml.util.ReadWriteUtils; +import org.apache.flink.runtime.state.StateInitializationContext; +import org.apache.flink.runtime.state.StateSnapshotContext; +import org.apache.flink.streaming.api.datastream.DataStream; +import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment; +import org.apache.flink.streaming.api.operators.AbstractStreamOperator; +import org.apache.flink.streaming.api.operators.BoundedOneInput; +import org.apache.flink.streaming.api.operators.OneInputStreamOperator; +import org.apache.flink.streaming.runtime.streamrecord.StreamRecord; +import org.apache.flink.table.api.Table; +import org.apache.flink.table.api.bridge.java.StreamTableEnvironment; +import org.apache.flink.table.api.internal.TableImpl; +import org.apache.flink.types.Row; +import org.apache.flink.util.Collector; +import org.apache.flink.util.Preconditions; + +import java.io.IOException; +import java.util.HashMap; +import java.util.Iterator; +import java.util.Map; + +/** + * An Estimator which implements the MinMaxScaler algorithm. + * + * See https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization). Review comment: OK -- 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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829675125
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainData);
+trainDataTable = tEnv.fromDataStream(dataStream,
schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictDataTable = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol,
DenseVector expected)
+throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(1, result.size());
+assertEquals(expected,
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829674990
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainData);
+trainDataTable = tEnv.fromDataStream(dataStream,
schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictDataTable = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol,
DenseVector expected)
+throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(1, result.size());
+assertEquals(expected,
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829674316
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829650219
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainData);
+trainDataTable = tEnv.fromDataStream(dataStream,
schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictDataTable = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol,
DenseVector expected)
+throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(1, result.size());
+assertEquals(expected,
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829015961
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainData);
+trainDataTable = tEnv.fromDataStream(dataStream,
schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictDataTable = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol,
DenseVector expected)
+throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(1, result.size());
+assertEquals(expected,
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829018999
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,205 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream features =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxValues =
+features.transform(
+"reduceInEachPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxValues,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minVector = iter.next();
+DenseVector maxVector = iter.next();
+out.collect(new
MinMaxScalerModelData(minVector, maxVector));
+}
+});
+
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829025886
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829018999
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,205 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream features =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxValues =
+features.transform(
+"reduceInEachPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxValues,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minVector = iter.next();
+DenseVector maxVector = iter.next();
+out.collect(new
MinMaxScalerModelData(minVector, maxVector));
+}
+});
+
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829015961
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,208 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainDataTable;
+private Table predictDataTable;
+private static final List trainData =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainData);
+trainDataTable = tEnv.fromDataStream(dataStream,
schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictDataTable = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol,
DenseVector expected)
+throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(1, result.size());
+assertEquals(expected,
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829014907
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r829000929
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,205 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream features =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxValues =
+features.transform(
+"reduceInEachPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxValues,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minVector = iter.next();
+DenseVector maxVector = iter.next();
+out.collect(new
MinMaxScalerModelData(minVector, maxVector));
+}
+});
+
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r828997214
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,181 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getPredictionCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictOutputFunction(
+broadcastModelKey,
+getMax(),
+getMin(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826538436
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,198 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream vectors =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxVectors =
+vectors.transform(
+"reduceInEachPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minDenseVector = iter.next();
+DenseVector maxDenseVector = iter.next();
+out.collect(
+new
MinMaxScalerModelData(minDenseVector, maxDenseVector));
+}
+});
+
+MinMaxScalerModel model =
+new
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r828829587
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,197 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream features =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxValues =
+features.transform(
+"reduceInEachPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+features.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxValues,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minVector = iter.next();
+DenseVector maxVector = iter.next();
+out.collect(new
MinMaxScalerModelData(minVector, maxVector));
+}
+});
+
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826542180
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,204 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainData;
+private Table predictData;
+private static final List trainRows =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainRows);
+trainData = tEnv.fromDataStream(dataStream, schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictData = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol)
throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+assertEquals(Vectors.dense(0.75, 0.3), result.get(0));
+}
+
+@Test
+public void testParam() {
+MinMaxScaler
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826541447
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,204 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainData;
+private Table predictData;
+private static final List trainRows =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainRows);
+trainData = tEnv.fromDataStream(dataStream, schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictData = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol)
throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
Review comment:
OK
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826540199
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,204 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainData;
+private Table predictData;
+private static final List trainRows =
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826539650
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,198 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream vectors =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxVectors =
+vectors.transform(
+"reduceInEachPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minDenseVector = iter.next();
+DenseVector maxDenseVector = iter.next();
+out.collect(
+new
MinMaxScalerModelData(minDenseVector, maxDenseVector));
+}
+});
+
+MinMaxScalerModel model =
+new
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826538436
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,198 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream vectors =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxVectors =
+vectors.transform(
+"reduceInEachPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.transform(
+"reduceInFinalPartition",
+vectors.getType(),
+new MinMaxReduceFunctionOperator())
+.setParallelism(1);
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction() {
+@Override
+public void mapPartition(
+Iterable values,
+Collector out) {
+Iterator iter = values.iterator();
+DenseVector minDenseVector = iter.next();
+DenseVector maxDenseVector = iter.next();
+out.collect(
+new
MinMaxScalerModelData(minDenseVector, maxDenseVector));
+}
+});
+
+MinMaxScalerModel model =
+new
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826537120
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,198 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream vectors =
+tEnv.toDataStream(inputs[0])
+.map(
+(MapFunction)
+value -> (DenseVector)
value.getField(featureCol));
+DataStream minMaxVectors =
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826536751
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.java
##
@@ -0,0 +1,198 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.common.state.ListState;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.runtime.state.StateInitializationContext;
+import org.apache.flink.runtime.state.StateSnapshotContext;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.streaming.api.operators.AbstractStreamOperator;
+import org.apache.flink.streaming.api.operators.BoundedOneInput;
+import org.apache.flink.streaming.api.operators.OneInputStreamOperator;
+import org.apache.flink.streaming.runtime.streamrecord.StreamRecord;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.Map;
+
+/**
+ * An Estimator which implements the MinMaxScaler algorithm.
+ *
+ * See
https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization).
+ */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+final String featureCol = getFeaturesCol();
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream vectors =
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r826536318
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,179 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getOutputCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictLabelFunction(
+broadcastModelKey,
+getMax(),
+getMIN(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823423537
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** An Estimator which implements the MinMaxScaler algorithm. */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream> minMaxVectors =
+computeMinMaxVectors(tEnv.toDataStream(inputs[0]),
getFeaturesCol());
+DataStream modelData =
genModelData(minMaxVectors);
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+ReadWriteUtils.updateExistingParams(model, getParamMap());
+return model;
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+}
+
+public static MinMaxScaler load(StreamExecutionEnvironment env, String
path)
+throws IOException {
+return ReadWriteUtils.loadStageParam(path);
+}
+
+/**
+ * Generates minMax scaler model data.
+ *
+ * @param minMaxVectors Input distributed minMaxVectors.
+ * @return MinMax scaler model data.
+ */
+private static DataStream genModelData(
+DataStream> minMaxVectors) {
+DataStream modelData =
+DataStreamUtils.mapPartition(
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823401050
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** An Estimator which implements the MinMaxScaler algorithm. */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream> minMaxVectors =
+computeMinMaxVectors(tEnv.toDataStream(inputs[0]),
getFeaturesCol());
+DataStream modelData =
genModelData(minMaxVectors);
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+ReadWriteUtils.updateExistingParams(model, getParamMap());
+return model;
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+}
+
+public static MinMaxScaler load(StreamExecutionEnvironment env, String
path)
+throws IOException {
+return ReadWriteUtils.loadStageParam(path);
+}
+
+/**
+ * Generates minMax scaler model data.
+ *
+ * @param minMaxVectors Input distributed minMaxVectors.
+ * @return MinMax scaler model data.
+ */
+private static DataStream genModelData(
+DataStream> minMaxVectors) {
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction<
+Tuple2,
MinMaxScalerModelData>() {
+@Override
+public void mapPartition(
+Iterable>
dataPoints,
+Collector out) {
+DenseVector minVector = null;
+DenseVector maxVector = null;
+int vecSize = 0;
+for (Tuple2
dataPoint : dataPoints) {
+if (maxVector == null) {
+vecSize = dataPoint.f0.size();
+maxVector = dataPoint.f1;
+minVector = dataPoint.f0;
+}
+for (int i = 0; i < vecSize; ++i) {
+minVector.values[i] =
+Math.min(
+dataPoint.f0.values[i],
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823425316
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerParams.java
##
@@ -0,0 +1,56 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.ml.common.param.HasFeaturesCol;
+import org.apache.flink.ml.common.param.HasOutputCol;
+import org.apache.flink.ml.param.DoubleParam;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.param.ParamValidators;
+
+/**
+ * Params for {@link MinMaxScaler}.
+ *
+ * @param The class type of this instance.
+ */
+public interface MinMaxScalerParams extends HasFeaturesCol,
HasOutputCol {
+Param MAX =
+new DoubleParam(
+"max", "Upper bound after transformation.", 1.0,
ParamValidators.notNull());
+
+default Double getMax() {
+return get(MAX);
+}
+
+default T setMax(Double value) {
+return set(MAX, value);
+}
+
+Param MIN =
+new DoubleParam(
+"min", "Lower bound after transformation.", 0.0,
ParamValidators.notNull());
+
+default Double getMIN() {
+return get(MIN);
+}
+
+default T setMIN(Double value) {
Review comment:
ditto.
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823414741
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerParams.java
##
@@ -0,0 +1,56 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.ml.common.param.HasFeaturesCol;
+import org.apache.flink.ml.common.param.HasOutputCol;
+import org.apache.flink.ml.param.DoubleParam;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.param.ParamValidators;
+
+/**
+ * Params for {@link MinMaxScaler}.
+ *
+ * @param The class type of this instance.
+ */
+public interface MinMaxScalerParams extends HasFeaturesCol,
HasOutputCol {
+Param MAX =
+new DoubleParam(
+"max", "Upper bound after transformation.", 1.0,
ParamValidators.notNull());
+
+default Double getMax() {
+return get(MAX);
+}
+
+default T setMax(Double value) {
+return set(MAX, value);
+}
+
+Param MIN =
+new DoubleParam(
+"min", "Lower bound after transformation.", 0.0,
ParamValidators.notNull());
+
+default Double getMIN() {
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r825551716
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerParams.java
##
@@ -0,0 +1,56 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.ml.common.param.HasFeaturesCol;
+import org.apache.flink.ml.common.param.HasOutputCol;
+import org.apache.flink.ml.param.DoubleParam;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.param.ParamValidators;
+
+/**
+ * Params for {@link MinMaxScaler}.
+ *
+ * @param The class type of this instance.
+ */
+public interface MinMaxScalerParams extends HasFeaturesCol,
HasOutputCol {
+Param MAX =
+new DoubleParam(
+"max", "Upper bound after transformation.", 1.0,
ParamValidators.notNull());
+
+default Double getMax() {
+return get(MAX);
+}
+
+default T setMax(Double value) {
+return set(MAX, value);
+}
+
+Param MIN =
+new DoubleParam(
+"min", "Lower bound after transformation.", 0.0,
ParamValidators.notNull());
+
+default Double getMIN() {
+return get(MIN);
+}
+
+default T setMIN(Double value) {
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823412958
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/common/param/HasOutputCol.java
##
@@ -0,0 +1,39 @@
+/*
+ * 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.Param;
+import org.apache.flink.ml.param.ParamValidators;
+import org.apache.flink.ml.param.StringParam;
+import org.apache.flink.ml.param.WithParams;
+
+/** Interface for the shared outputCol param. */
+public interface HasOutputCol extends WithParams {
Review comment:
PredictionCol may be better for ml algo. For feature engineer and data
proc, outputCol may be better.
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/common/param/HasOutputCol.java
##
@@ -0,0 +1,39 @@
+/*
+ * 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.Param;
+import org.apache.flink.ml.param.ParamValidators;
+import org.apache.flink.ml.param.StringParam;
+import org.apache.flink.ml.param.WithParams;
+
+/** Interface for the shared outputCol param. */
+public interface HasOutputCol extends WithParams {
Review comment:
OK
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823425316
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerParams.java
##
@@ -0,0 +1,56 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.ml.common.param.HasFeaturesCol;
+import org.apache.flink.ml.common.param.HasOutputCol;
+import org.apache.flink.ml.param.DoubleParam;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.param.ParamValidators;
+
+/**
+ * Params for {@link MinMaxScaler}.
+ *
+ * @param The class type of this instance.
+ */
+public interface MinMaxScalerParams extends HasFeaturesCol,
HasOutputCol {
+Param MAX =
+new DoubleParam(
+"max", "Upper bound after transformation.", 1.0,
ParamValidators.notNull());
+
+default Double getMax() {
+return get(MAX);
+}
+
+default T setMax(Double value) {
+return set(MAX, value);
+}
+
+Param MIN =
+new DoubleParam(
+"min", "Lower bound after transformation.", 0.0,
ParamValidators.notNull());
+
+default Double getMIN() {
+return get(MIN);
+}
+
+default T setMIN(Double value) {
Review comment:
ditto.
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823423537
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** An Estimator which implements the MinMaxScaler algorithm. */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream> minMaxVectors =
+computeMinMaxVectors(tEnv.toDataStream(inputs[0]),
getFeaturesCol());
+DataStream modelData =
genModelData(minMaxVectors);
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+ReadWriteUtils.updateExistingParams(model, getParamMap());
+return model;
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+}
+
+public static MinMaxScaler load(StreamExecutionEnvironment env, String
path)
+throws IOException {
+return ReadWriteUtils.loadStageParam(path);
+}
+
+/**
+ * Generates minMax scaler model data.
+ *
+ * @param minMaxVectors Input distributed minMaxVectors.
+ * @return MinMax scaler model data.
+ */
+private static DataStream genModelData(
+DataStream> minMaxVectors) {
+DataStream modelData =
+DataStreamUtils.mapPartition(
Review comment:
This mapPartition only has fewer record. For every worker only has one
record. I think it's no efficient problem.
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823414741
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerParams.java
##
@@ -0,0 +1,56 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.ml.common.param.HasFeaturesCol;
+import org.apache.flink.ml.common.param.HasOutputCol;
+import org.apache.flink.ml.param.DoubleParam;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.param.ParamValidators;
+
+/**
+ * Params for {@link MinMaxScaler}.
+ *
+ * @param The class type of this instance.
+ */
+public interface MinMaxScalerParams extends HasFeaturesCol,
HasOutputCol {
+Param MAX =
+new DoubleParam(
+"max", "Upper bound after transformation.", 1.0,
ParamValidators.notNull());
+
+default Double getMax() {
+return get(MAX);
+}
+
+default T setMax(Double value) {
+return set(MAX, value);
+}
+
+Param MIN =
+new DoubleParam(
+"min", "Lower bound after transformation.", 0.0,
ParamValidators.notNull());
+
+default Double getMIN() {
Review comment:
I define api as alink and spark.
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54: URL: https://github.com/apache/flink-ml/pull/54#discussion_r823414091 ## File path: flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler; + +import org.apache.flink.api.common.functions.RichMapPartitionFunction; +import org.apache.flink.api.java.tuple.Tuple2; +import org.apache.flink.ml.api.Estimator; +import org.apache.flink.ml.common.datastream.DataStreamUtils; +import org.apache.flink.ml.linalg.DenseVector; +import org.apache.flink.ml.param.Param; +import org.apache.flink.ml.util.ParamUtils; +import org.apache.flink.ml.util.ReadWriteUtils; +import org.apache.flink.streaming.api.datastream.DataStream; +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.table.api.internal.TableImpl; +import org.apache.flink.types.Row; +import org.apache.flink.util.Collector; +import org.apache.flink.util.Preconditions; + +import java.io.IOException; +import java.util.HashMap; +import java.util.Map; + +/** An Estimator which implements the MinMaxScaler algorithm. */ Review comment: OK -- 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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823413985
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** An Estimator which implements the MinMaxScaler algorithm. */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream> minMaxVectors =
+computeMinMaxVectors(tEnv.toDataStream(inputs[0]),
getFeaturesCol());
+DataStream modelData =
genModelData(minMaxVectors);
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+ReadWriteUtils.updateExistingParams(model, getParamMap());
+return model;
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+}
+
+public static MinMaxScaler load(StreamExecutionEnvironment env, String
path)
+throws IOException {
+return ReadWriteUtils.loadStageParam(path);
+}
+
+/**
+ * Generates minMax scaler model data.
+ *
+ * @param minMaxVectors Input distributed minMaxVectors.
+ * @return MinMax scaler model data.
+ */
+private static DataStream genModelData(
+DataStream> minMaxVectors) {
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction<
+Tuple2,
MinMaxScalerModelData>() {
+@Override
+public void mapPartition(
+Iterable>
dataPoints,
+Collector out) {
+DenseVector minVector = null;
+DenseVector maxVector = null;
+int vecSize = 0;
+for (Tuple2
dataPoint : dataPoints) {
+if (maxVector == null) {
+vecSize = dataPoint.f0.size();
+maxVector = dataPoint.f1;
+minVector = dataPoint.f0;
+}
+for (int i = 0; i < vecSize; ++i) {
Review comment:
ok
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823413341
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,206 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainData;
+private Table predictData;
+private static final List trainRows =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainRows);
+trainData = tEnv.fromDataStream(dataStream, schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictData = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol)
throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+for (DenseVector t2 : result) {
+assertEquals(Vectors.dense(0.75, 0.3), t2);
+}
+}
+
+@Test
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823412958
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/common/param/HasOutputCol.java
##
@@ -0,0 +1,39 @@
+/*
+ * 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.Param;
+import org.apache.flink.ml.param.ParamValidators;
+import org.apache.flink.ml.param.StringParam;
+import org.apache.flink.ml.param.WithParams;
+
+/** Interface for the shared outputCol param. */
+public interface HasOutputCol extends WithParams {
Review comment:
PredictionCol may be better for ml algo. For feature engineer and data
proc, outputCol may be better.
--
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
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823410761
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,179 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getOutputCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictLabelFunction(
+broadcastModelKey,
+getMax(),
+getMIN(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823407289
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScalerModel.java
##
@@ -0,0 +1,179 @@
+/*
+ * 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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Model;
+import org.apache.flink.ml.common.broadcast.BroadcastUtils;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.table.runtime.typeutils.ExternalTypeInfo;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * A Model which do a minMax scaler operation using the model data computed by
{@link MinMaxScaler}.
+ */
+public class MinMaxScalerModel
+implements Model,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+private Table modelDataTable;
+
+public MinMaxScalerModel() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel setModelData(Table... inputs) {
+modelDataTable = inputs[0];
+return this;
+}
+
+@Override
+public Table[] getModelData() {
+return new Table[] {modelDataTable};
+}
+
+@Override
+@SuppressWarnings("unchecked")
+public Table[] transform(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream data = tEnv.toDataStream(inputs[0]);
+DataStream minMaxScalerModel =
+MinMaxScalerModelData.getModelDataStream(modelDataTable);
+final String broadcastModelKey = "broadcastModelKey";
+RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+RowTypeInfo outputTypeInfo =
+new RowTypeInfo(
+ArrayUtils.addAll(
+inputTypeInfo.getFieldTypes(),
+ExternalTypeInfo.of(DenseVector.class)),
+ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getOutputCol()));
+DataStream output =
+BroadcastUtils.withBroadcastStream(
+Collections.singletonList(data),
+Collections.singletonMap(broadcastModelKey,
minMaxScalerModel),
+inputList -> {
+DataStream input = inputList.get(0);
+return input.map(
+new PredictLabelFunction(
+broadcastModelKey,
+getMax(),
+getMIN(),
+getFeaturesCol()),
+outputTypeInfo);
+});
+return new Table[] {tEnv.fromDataStream(output)};
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+ReadWriteUtils.saveModelData(
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823402845
##
File path:
flink-ml-lib/src/test/java/org/apache/flink/ml/feature/MinMaxScalerTest.java
##
@@ -0,0 +1,206 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.restartstrategy.RestartStrategies;
+import org.apache.flink.configuration.Configuration;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScaler;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModel;
+import org.apache.flink.ml.feature.minmaxscaler.MinMaxScalerModelData;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.Vectors;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.ml.util.StageTestUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+import
org.apache.flink.streaming.api.environment.ExecutionCheckpointingOptions;
+import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
+import org.apache.flink.table.api.DataTypes;
+import org.apache.flink.table.api.Schema;
+import org.apache.flink.table.api.Table;
+import org.apache.flink.table.api.bridge.java.StreamTableEnvironment;
+import org.apache.flink.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+
+import org.apache.commons.collections.IteratorUtils;
+import org.junit.Assert;
+import org.junit.Before;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.rules.TemporaryFolder;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.List;
+
+import static org.junit.Assert.assertEquals;
+
+/** Tests {@link MinMaxScaler} and {@link MinMaxScalerModel}. */
+public class MinMaxScalerTest {
+@Rule public final TemporaryFolder tempFolder = new TemporaryFolder();
+private StreamExecutionEnvironment env;
+private StreamTableEnvironment tEnv;
+private Table trainData;
+private Table predictData;
+private static final List trainRows =
+new ArrayList<>(
+Arrays.asList(
+Row.of(Vectors.dense(0.0, 3.0)),
+Row.of(Vectors.dense(2.1, 0.0)),
+Row.of(Vectors.dense(4.1, 5.1)),
+Row.of(Vectors.dense(6.1, 8.1)),
+Row.of(Vectors.dense(200, 300;
+private static final List predictRows =
+new
ArrayList<>(Collections.singletonList(Row.of(Vectors.dense(150.0, 90.0;
+
+@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);
+Schema schema = Schema.newBuilder().column("f0",
DataTypes.of(DenseVector.class)).build();
+DataStream dataStream = env.fromCollection(trainRows);
+trainData = tEnv.fromDataStream(dataStream, schema).as("features");
+DataStream predDataStream = env.fromCollection(predictRows);
+predictData = tEnv.fromDataStream(predDataStream,
schema).as("features");
+}
+
+private static void verifyPredictionResult(Table output, String outputCol)
throws Exception {
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
output).getTableEnvironment();
+DataStream stream =
+tEnv.toDataStream(output)
+.map(
+(MapFunction)
+row -> (DenseVector)
row.getField(outputCol));
+List result =
IteratorUtils.toList(stream.executeAndCollect());
+for (DenseVector t2 : result) {
+assertEquals(Vectors.dense(0.75, 0.3), t2);
+}
+}
+
+@Test
+
[GitHub] [flink-ml] weibozhao commented on a change in pull request #54: [FLINK-25552] Add Estimator and Transformer for MinMaxScaler in FlinkML
weibozhao commented on a change in pull request #54:
URL: https://github.com/apache/flink-ml/pull/54#discussion_r823401050
##
File path:
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/minmaxscaler/MinMaxScaler.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.minmaxscaler;
+
+import org.apache.flink.api.common.functions.RichMapPartitionFunction;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.ml.api.Estimator;
+import org.apache.flink.ml.common.datastream.DataStreamUtils;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.param.Param;
+import org.apache.flink.ml.util.ParamUtils;
+import org.apache.flink.ml.util.ReadWriteUtils;
+import org.apache.flink.streaming.api.datastream.DataStream;
+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.table.api.internal.TableImpl;
+import org.apache.flink.types.Row;
+import org.apache.flink.util.Collector;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** An Estimator which implements the MinMaxScaler algorithm. */
+public class MinMaxScaler
+implements Estimator,
MinMaxScalerParams {
+private final Map, Object> paramMap = new HashMap<>();
+
+public MinMaxScaler() {
+ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+}
+
+@Override
+public MinMaxScalerModel fit(Table... inputs) {
+Preconditions.checkArgument(inputs.length == 1);
+StreamTableEnvironment tEnv =
+(StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+DataStream> minMaxVectors =
+computeMinMaxVectors(tEnv.toDataStream(inputs[0]),
getFeaturesCol());
+DataStream modelData =
genModelData(minMaxVectors);
+MinMaxScalerModel model =
+new
MinMaxScalerModel().setModelData(tEnv.fromDataStream(modelData));
+ReadWriteUtils.updateExistingParams(model, getParamMap());
+return model;
+}
+
+@Override
+public Map, Object> getParamMap() {
+return paramMap;
+}
+
+@Override
+public void save(String path) throws IOException {
+ReadWriteUtils.saveMetadata(this, path);
+}
+
+public static MinMaxScaler load(StreamExecutionEnvironment env, String
path)
+throws IOException {
+return ReadWriteUtils.loadStageParam(path);
+}
+
+/**
+ * Generates minMax scaler model data.
+ *
+ * @param minMaxVectors Input distributed minMaxVectors.
+ * @return MinMax scaler model data.
+ */
+private static DataStream genModelData(
+DataStream> minMaxVectors) {
+DataStream modelData =
+DataStreamUtils.mapPartition(
+minMaxVectors,
+new RichMapPartitionFunction<
+Tuple2,
MinMaxScalerModelData>() {
+@Override
+public void mapPartition(
+Iterable>
dataPoints,
+Collector out) {
+DenseVector minVector = null;
+DenseVector maxVector = null;
+int vecSize = 0;
+for (Tuple2
dataPoint : dataPoints) {
+if (maxVector == null) {
+vecSize = dataPoint.f0.size();
+maxVector = dataPoint.f1;
+minVector = dataPoint.f0;
+}
+for (int i = 0; i < vecSize; ++i) {
+minVector.values[i] =
+Math.min(
+dataPoint.f0.values[i],
+
