zhipeng93 commented on code in PR #154:
URL: https://github.com/apache/flink-ml/pull/154#discussion_r963183034
##########
flink-ml-core/src/main/java/org/apache/flink/ml/linalg/BLAS.java:
##########
@@ -129,6 +129,33 @@ private static double norm2(SparseVector x) {
return JAVA_BLAS.dnrm2(x.values.length, x.values, 1);
}
+ /** Calculates the p-norm of the vector x. */
+ public static double norm(Vector x, double p) {
+ Preconditions.checkArgument(p >= 1.0, "p value must >= 1.0.");
Review Comment:
How about we update the error message to also print the value of `p`?
e.g., "To compute p-norm of a vector, p should be greater than or equal to
1.0. However the specified p is " + p.
##########
flink-ml-examples/src/main/java/org/apache/flink/ml/examples/feature/NormalizerExample.java:
##########
@@ -0,0 +1,62 @@
+/*
+ * 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.examples.feature;
+
+import org.apache.flink.ml.feature.normalizer.Normalizer;
+import org.apache.flink.ml.linalg.Vector;
+import org.apache.flink.ml.linalg.Vectors;
+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.types.Row;
+import org.apache.flink.util.CloseableIterator;
+
+/** Simple program that creates a Normalizer instance and uses it for feature
engineering. */
Review Comment:
It should be `an` Normalizer.
Same for all other cases in this PR.
##########
flink-ml-core/src/main/java/org/apache/flink/ml/linalg/BLAS.java:
##########
@@ -129,6 +129,33 @@ private static double norm2(SparseVector x) {
return JAVA_BLAS.dnrm2(x.values.length, x.values, 1);
}
+ /** Calculates the p-norm of the vector x. */
+ public static double norm(Vector x, double p) {
Review Comment:
Could you please also add a unit test for function `norm`?
##########
flink-ml-core/src/main/java/org/apache/flink/ml/linalg/BLAS.java:
##########
@@ -129,6 +129,33 @@ private static double norm2(SparseVector x) {
return JAVA_BLAS.dnrm2(x.values.length, x.values, 1);
}
+ /** Calculates the p-norm of the vector x. */
+ public static double norm(Vector x, double p) {
+ Preconditions.checkArgument(p >= 1.0, "p value must >= 1.0.");
+ double norm = 0.0;
+ double[] data =
+ (x instanceof DenseVector) ? ((DenseVector) x).values :
((SparseVector) x).values;
+
+ if (p == 1.0) {
+ for (double datum : data) {
+ norm += Math.abs(datum);
+ }
+ } else if (p == 2.0) {
+ norm = norm2(x);
+ } else if (Double.isInfinite(p)) {
Review Comment:
It should be `p == Double.POSITIVE_INFINITY`
##########
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/normalizer/Normalizer.java:
##########
@@ -0,0 +1,115 @@
+/*
+ * 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.normalizer;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Transformer;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.BLAS;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.SparseVector;
+import org.apache.flink.ml.linalg.Vector;
+import org.apache.flink.ml.linalg.typeinfo.VectorTypeInfo;
+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.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.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** A Transformer that normalizes a vector to have unit norm using the given
p-norm. */
+public class Normalizer implements Transformer<Normalizer>,
NormalizerParams<Normalizer> {
+ private final Map<Param<?>, Object> paramMap = new HashMap<>();
+
+ public Normalizer() {
+ ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+ }
+
+ @Override
+ public Table[] transform(Table... inputs) {
+ Preconditions.checkArgument(inputs.length == 1);
+ StreamTableEnvironment tEnv =
+ (StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+ RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+
+ RowTypeInfo outputTypeInfo =
+ new RowTypeInfo(
+ ArrayUtils.addAll(inputTypeInfo.getFieldTypes(),
VectorTypeInfo.INSTANCE),
+ ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getOutputCol()));
+
+ DataStream<Row> output =
+ tEnv.toDataStream(inputs[0])
+ .map(new NormalizeFunction(getP(), getInputCol()),
outputTypeInfo);
+
+ Table outputTable = tEnv.fromDataStream(output);
+ return new Table[] {outputTable};
+ }
+
+ @Override
+ public void save(String path) throws IOException {
+ ReadWriteUtils.saveMetadata(this, path);
+ }
+
+ public static Normalizer load(StreamTableEnvironment env, String path)
throws IOException {
+ return ReadWriteUtils.loadStageParam(path);
+ }
+
+ @Override
+ public Map<Param<?>, Object> getParamMap() {
+ return paramMap;
+ }
+
+ /** Normalize function which normalizes a vector to have unit norm. */
Review Comment:
How about updating the class name to `NormalizationFunction` and update the
comment as:
"Normalization function that transforms a vector to have unit norm."
##########
flink-ml-lib/src/main/java/org/apache/flink/ml/feature/normalizer/Normalizer.java:
##########
@@ -0,0 +1,115 @@
+/*
+ * 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.normalizer;
+
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.java.typeutils.RowTypeInfo;
+import org.apache.flink.ml.api.Transformer;
+import org.apache.flink.ml.common.datastream.TableUtils;
+import org.apache.flink.ml.linalg.BLAS;
+import org.apache.flink.ml.linalg.DenseVector;
+import org.apache.flink.ml.linalg.SparseVector;
+import org.apache.flink.ml.linalg.Vector;
+import org.apache.flink.ml.linalg.typeinfo.VectorTypeInfo;
+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.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.Preconditions;
+
+import org.apache.commons.lang3.ArrayUtils;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+
+/** A Transformer that normalizes a vector to have unit norm using the given
p-norm. */
+public class Normalizer implements Transformer<Normalizer>,
NormalizerParams<Normalizer> {
+ private final Map<Param<?>, Object> paramMap = new HashMap<>();
+
+ public Normalizer() {
+ ParamUtils.initializeMapWithDefaultValues(paramMap, this);
+ }
+
+ @Override
+ public Table[] transform(Table... inputs) {
+ Preconditions.checkArgument(inputs.length == 1);
+ StreamTableEnvironment tEnv =
+ (StreamTableEnvironment) ((TableImpl)
inputs[0]).getTableEnvironment();
+ RowTypeInfo inputTypeInfo =
TableUtils.getRowTypeInfo(inputs[0].getResolvedSchema());
+
+ RowTypeInfo outputTypeInfo =
+ new RowTypeInfo(
+ ArrayUtils.addAll(inputTypeInfo.getFieldTypes(),
VectorTypeInfo.INSTANCE),
+ ArrayUtils.addAll(inputTypeInfo.getFieldNames(),
getOutputCol()));
+
+ DataStream<Row> output =
+ tEnv.toDataStream(inputs[0])
+ .map(new NormalizeFunction(getP(), getInputCol()),
outputTypeInfo);
+
+ Table outputTable = tEnv.fromDataStream(output);
+ return new Table[] {outputTable};
+ }
+
+ @Override
+ public void save(String path) throws IOException {
+ ReadWriteUtils.saveMetadata(this, path);
+ }
+
+ public static Normalizer load(StreamTableEnvironment env, String path)
throws IOException {
+ return ReadWriteUtils.loadStageParam(path);
+ }
+
+ @Override
+ public Map<Param<?>, Object> getParamMap() {
+ return paramMap;
+ }
+
+ /** Normalize function which normalizes a vector to have unit norm. */
+ private static class NormalizeFunction implements MapFunction<Row, Row> {
+ private final double p;
+ private final String inputCol;
+
+ public NormalizeFunction(double p, String inputCol) {
+ this.p = p;
+ this.inputCol = inputCol;
+ }
+
+ @Override
+ public Row map(Row row) throws Exception {
+ Vector inputVec = row.getFieldAs(inputCol);
+ Vector outputVec = inputVec.clone();
+ double norm = BLAS.norm(inputVec, p);
+
+ double[] data =
Review Comment:
This logic could be simplified with `BLAS.scal`.
--
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
