xuyang1706 commented on a change in pull request #8631: [FLINK-12745][ml] add sparse and dense vector class, and dense matrix class with basic operations. URL: https://github.com/apache/flink/pull/8631#discussion_r311994360
########## File path: flink-ml-parent/flink-ml-lib/src/main/java/org/apache/flink/ml/common/matrix/Vector.java ########## @@ -0,0 +1,340 @@ +/* + * 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.matrix; + +import org.apache.commons.lang3.StringUtils; + +import java.io.Serializable; + +/** + * The Vector class defines some common methods for both DenseVector and + * SparseVector. + */ +public abstract class Vector implements Serializable { + + /** + * Parse a DenseVector from a formatted string. + */ + public static DenseVector dense(String str) { + return DenseVector.deserialize(str); + } + + /** + * Parse a SparseVector from a formatted string. + */ + public static SparseVector sparse(String str) { + return SparseVector.deserialize(str); + } + + /** + * To check whether the formatted string represents a SparseVector. + */ + public static boolean isSparse(String str) { + if (org.apache.flink.util.StringUtils.isNullOrWhitespaceOnly(str)) { + return true; + } + return StringUtils.indexOf(str, ':') != -1 || StringUtils.indexOf(str, "$") != -1; + } + + /** + * Parse the tensor from a formatted string. + */ + public static Vector deserialize(String str) { + Vector vec; + if (isSparse(str)) { + vec = Vector.sparse(str); + } else { + vec = Vector.dense(str); + } + return vec; + } + + /** + * Plus two vectors and create a new vector to store the result. + */ + public static Vector plus(Vector vec1, Vector vec2) { + return vec1.plus(vec2); + } + + /** + * Minus two vectors and create a new vector to store the result. + */ + public static Vector minus(Vector vec1, Vector vec2) { + return vec1.minus(vec2); + } + + /** + * Compute the dot product of two vectors. + */ + public static double dot(Vector vec1, Vector vec2) { + return vec1.dot(vec2); + } + + /** + * Compute || vec1 - vec2 ||_1 . + */ + public static double sumAbsDiff(Vector vec1, Vector vec2) { + if (vec1 instanceof DenseVector) { + if (vec2 instanceof DenseVector) { + return applySum((DenseVector) vec1, (DenseVector) vec2, (a, b) -> Math.abs(a - b)); + } else { + return applySum((DenseVector) vec1, (SparseVector) vec2, (a, b) -> Math.abs(a - b)); + } + } else { + if (vec2 instanceof DenseVector) { + return applySum((SparseVector) vec1, (DenseVector) vec2, (a, b) -> Math.abs(a - b)); + } else { + return applySum((SparseVector) vec1, (SparseVector) vec2, (a, b) -> Math.abs(a - b)); + } + } + } + + /** + * Compute || vec1 - vec2 ||_2^2 . + */ + public static double sumSquaredDiff(Vector vec1, Vector vec2) { + if (vec1 instanceof DenseVector) { + if (vec2 instanceof DenseVector) { + return applySum((DenseVector) vec1, (DenseVector) vec2, (a, b) -> (a - b) * (a - b)); + } else { + return applySum((DenseVector) vec1, (SparseVector) vec2, (a, b) -> (a - b) * (a - b)); + } + } else { + if (vec2 instanceof DenseVector) { + return applySum((SparseVector) vec1, (DenseVector) vec2, (a, b) -> (a - b) * (a - b)); + } else { + return applySum((SparseVector) vec1, (SparseVector) vec2, (a, b) -> (a - b) * (a - b)); + } + } + } + + /** + * Compute element wise sum between two DenseVector. + * \sum_i func(x1_i, x2_i) + */ + public static double applySum(DenseVector x1, DenseVector x2, BinaryOp func) { + assert x1.size() == x2.size(); + double[] x1data = x1.getData(); + double[] x2data = x2.getData(); + double s = 0.; + for (int i = 0; i < x1data.length; i++) { + s += func.f(x1data[i], x2data[i]); + } + return s; + } + + /** + * Compute element wise sum between two SparseVector. + * \sum_i func(x1_i, x2_i) + */ + public static double applySum(SparseVector x1, SparseVector x2, BinaryOp func) { + double s = 0.; + int p1 = 0; + int p2 = 0; + int[] x1Indices = x1.getIndices(); + double[] x1Values = x1.getValues(); + int[] x2Indices = x2.getIndices(); + double[] x2Values = x2.getValues(); + int nnz1 = x1Indices.length; + int nnz2 = x2Indices.length; + while (p1 < nnz1 || p2 < nnz2) { + if (p1 < nnz1 && p2 < nnz2) { + if (x1Indices[p1] == x2Indices[p2]) { + s += func.f(x1Values[p1], x2Values[p2]); + p1++; + p2++; + } else if (x1Indices[p1] < x2Indices[p2]) { + s += func.f(x1Values[p1], 0.); + p1++; + } else { + s += func.f(0., x2Values[p2]); + p2++; + } + } else { + if (p1 < nnz1) { + s += func.f(x1Values[p1], 0.); + p1++; + } else { // p2 < nnz2 + s += func.f(0., x2Values[p2]); + p2++; + } + } + } + return s; + } + + /** + * Compute element wise sum between a DenseVector and a SparseVector. + * \sum_i func(x1_i, x2_i) + */ + public static double applySum(DenseVector x1, SparseVector x2, BinaryOp func) { + assert x1.size() == x2.size(); + double s = 0.; + int p2 = 0; + int[] x2Indices = x2.getIndices(); + double[] x2Values = x2.getValues(); + int nnz2 = x2Indices.length; + double[] x1data = x1.getData(); + for (int i = 0; i < x1data.length; i++) { + if (p2 < nnz2 && x2Indices[p2] == i) { + s += func.f(x1data[i], x2Values[p2]); + p2++; + } else { + s += func.f(x1data[i], 0.); + } + } + return s; + } + + /** + * Compute element wise sum between a SparseVector and a DenseVector. + * \sum_i func(x1_i, x2_i) + */ + public static double applySum(SparseVector x1, DenseVector x2, BinaryOp func) { + assert x1.size() == x2.size(); + double s = 0.; + int p1 = 0; + int[] x1Indices = x1.getIndices(); + double[] x1Values = x1.getValues(); + int nnz1 = x1Indices.length; + double[] x2data = x2.getData(); + for (int i = 0; i < x2data.length; i++) { + if (p1 < nnz1 && x1Indices[p1] == i) { + s += func.f(x1Values[p1], x2data[i]); + p1++; + } else { + s += func.f(0., x2data[i]); + } + } + return s; + } + + /** + * Get the size of the vector. + */ + public abstract int size(); + + /** + * Get the i-th element of the vector. + */ + public abstract double get(int i); + + /** + * Set the i-th element of the vector to value "val". + */ + public abstract void set(int i, double val); + + /** + * Add the i-th element of the vector by value "val". + */ + public abstract void add(int i, double val); + + /** + * Return the L1 norm of the vector. + */ + public abstract double normL1(); + + /** + * Return the Inf norm of the vector. + */ + public abstract double normInf(); + + /** + * Return the L2 norm of the vector. + */ + public abstract double normL2(); + + /** + * Return the square of L2 norm of the vector. + */ + public abstract double normL2Square(); + + /** + * Scale the vector by value "v" and create a new vector to store the result. + */ + public abstract Vector scale(double v); + + /** + * Scale the vector by value "v". + */ + public abstract void scaleEqual(double v); + + /** + * Normalize the vector. + */ + public abstract void normalizeEqual(double p); + + /** + * Standardize the vector. + */ + public abstract void standardizeEqual(double mean, double stdvar); + + /** + * Create a new vector by adding an element to the head of the vector. + */ + public abstract Vector prefix(double v); + + /** + * Create a new vector by adding an element to the end of the vector. + */ + public abstract Vector append(double v); + + /** + * Create a new vector by plussing another vector. + */ + public abstract Vector plus(Vector vec); + + /** + * Create a new vector by subtracting another vector. + */ + public abstract Vector minus(Vector vec); + + /** + * Compute the dot product with another vector. + */ + public abstract double dot(Vector vec); + + /** + * Get the iterator of the vector. + */ + public abstract VectorIterator iterator(); + + /** + * Serialize the vector to a string. + */ + public abstract String serialize(); + + /** + * Slice the vector. + */ + public abstract Vector slice(int[] indexes); + + /** + * Convert the vector to DenseVector. + */ + public DenseVector toDenseVector() { Review comment: I have removed this method in this class. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
