blob: 997e43dd41f94eca28a23ac8a8ca72f713cbf6fa [file] [log] [blame]
/*
* 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.sysds.test.functions.binary.matrix_full_other;
import java.util.HashMap;
import org.junit.AfterClass;
import org.junit.BeforeClass;
import org.junit.Test;
import org.apache.sysds.common.Types.ExecMode;
import org.apache.sysds.lops.LopProperties.ExecType;
import org.apache.sysds.runtime.matrix.data.MatrixValue.CellIndex;
import org.apache.sysds.test.AutomatedTestBase;
import org.apache.sysds.test.TestConfiguration;
import org.apache.sysds.test.TestUtils;
public class FullMatrixMultiplicationTest extends AutomatedTestBase
{
private final static String TEST_NAME = "FullMatrixMultiplication";
private final static String TEST_DIR = "functions/binary/matrix_full_other/";
private final static String TEST_CLASS_DIR = TEST_DIR + FullMatrixMultiplicationTest.class.getSimpleName() + "/";
private final static double eps = 1e-10;
private final static int rowsA = 1101;
private final static int colsA = 1203;
private final static int rowsB = 1203;
private final static int colsB = 1007;
private final static double sparsity1 = 0.7;
private final static double sparsity2 = 0.1;
@Override
public void setUp()
{
TestUtils.clearAssertionInformation();
addTestConfiguration(
TEST_NAME,
new TestConfiguration(TEST_CLASS_DIR, TEST_NAME,
new String[] { "C" }) );
if (TEST_CACHE_ENABLED) {
setOutAndExpectedDeletionDisabled(true);
}
}
@BeforeClass
public static void init()
{
TestUtils.clearDirectory(TEST_DATA_DIR + TEST_CLASS_DIR);
}
@AfterClass
public static void cleanUp()
{
if (TEST_CACHE_ENABLED) {
TestUtils.clearDirectory(TEST_DATA_DIR + TEST_CLASS_DIR);
}
}
@Test
public void testMMDenseDenseCP()
{
runMatrixMatrixMultiplicationTest(false, false, ExecType.CP);
}
@Test
public void testMMDenseSparseCP()
{
runMatrixMatrixMultiplicationTest(false, true, ExecType.CP);
}
@Test
public void testMMSparseDenseCP()
{
runMatrixMatrixMultiplicationTest(true, false, ExecType.CP);
}
@Test
public void testMMSparseSparseCP()
{
runMatrixMatrixMultiplicationTest(true, true, ExecType.CP);
}
@Test
public void testMVDenseDenseCP()
{
runMatrixVectorMultiplicationTest(false, ExecType.CP);
}
@Test
public void testMVSparseDenseCP()
{
runMatrixVectorMultiplicationTest(true, ExecType.CP);
}
@Test
public void testVMDenseDenseCP()
{
runVectorMatrixMultiplicationTest(false, false, ExecType.CP);
}
@Test
public void testVMDenseSparseCP()
{
runVectorMatrixMultiplicationTest(false, true, ExecType.CP);
}
@Test
public void testVMSparseDenseCP()
{
runVectorMatrixMultiplicationTest(true, false, ExecType.CP);
}
@Test
public void testVMSparseSparseCP()
{
runVectorMatrixMultiplicationTest(true, true, ExecType.CP);
}
@Test
public void testVVDenseDenseCP()
{
runVectorVectorMultiplicationTest(false, false, ExecType.CP);
}
@Test
public void testVVSparseDenseCP()
{
runVectorVectorMultiplicationTest(true, false, ExecType.CP);
}
@Test
public void testVtVtDenseDenseCP()
{
runVectorVectorMultiplicationTest(false, true, ExecType.CP);
}
@Test
public void testVtVtSparseDenseCP()
{
runVectorVectorMultiplicationTest(true, true, ExecType.CP);
}
private void runMatrixMatrixMultiplicationTest( boolean sparseM1, boolean sparseM2, ExecType instType)
{
//setup exec type, rows, cols
//rtplatform for MR
ExecMode platformOld = rtplatform;
rtplatform = ExecMode.HYBRID;
try
{
TestConfiguration config = getTestConfiguration(TEST_NAME);
double sparsityA = sparseM1?sparsity2:sparsity1;
double sparsityB = sparseM2?sparsity2:sparsity1;
String TEST_CACHE_DIR = "";
if (TEST_CACHE_ENABLED) {
TEST_CACHE_DIR = "mm" + String.valueOf(sparsityA) + "_" + String.valueOf(sparsityB) + "/";
}
loadTestConfiguration(config, TEST_CACHE_DIR);
/* This is for running the junit test the new way, i.e., construct the arguments directly */
String HOME = SCRIPT_DIR + TEST_DIR;
fullDMLScriptName = HOME + TEST_NAME + ".dml";
programArgs = new String[]{"-stats","-args",
input("A"), Integer.toString(rowsA), Integer.toString(colsA),
input("B"), Integer.toString(rowsB), Integer.toString(colsB), output("C") };
fullRScriptName = HOME + TEST_NAME + ".R";
rCmd = "Rscript" + " " + fullRScriptName + " " + inputDir() + " " + expectedDir();
//generate actual dataset
double[][] A = getRandomMatrix(rowsA, colsA, 0, 1, sparsityA, 7);
writeInputMatrix("A", A, true);
double[][] B = getRandomMatrix(rowsB, colsB, 0, 1, sparsityB, 3);
writeInputMatrix("B", B, true);
boolean exceptionExpected = false;
runTest(true, exceptionExpected, null, -1);
runRScript(true);
//compare matrices
HashMap<CellIndex, Double> dmlfile = readDMLMatrixFromHDFS("C");
HashMap<CellIndex, Double> rfile = readRMatrixFromFS("C");
TestUtils.compareMatrices(dmlfile, rfile, eps, "Stat-DML", "Stat-R");
}
finally
{
rtplatform = platformOld;
}
}
/**
* Note: second matrix is always dense if vector.
*
* @param sparseM1
* @param instType
*/
private void runMatrixVectorMultiplicationTest( boolean sparseM1, ExecType instType)
{
//setup exec type, rows, cols
//rtplatform for MR
ExecMode platformOld = rtplatform;
rtplatform = ExecMode.HYBRID;
try
{
TestConfiguration config = getTestConfiguration(TEST_NAME);
double sparsityA = sparseM1?sparsity2:sparsity1;
String TEST_CACHE_DIR = "";
if (TEST_CACHE_ENABLED) {
TEST_CACHE_DIR = "mv" + String.valueOf(sparsityA) + "/";
}
loadTestConfiguration(config, TEST_CACHE_DIR);
/* This is for running the junit test the new way, i.e., construct the arguments directly */
String HOME = SCRIPT_DIR + TEST_DIR;
fullDMLScriptName = HOME + TEST_NAME + ".dml";
programArgs = new String[]{"-args",
input("A"), Integer.toString(rowsA), Integer.toString(colsA),
input("B"), Integer.toString(rowsB), Integer.toString(1), output("C") };
fullRScriptName = HOME + TEST_NAME + ".R";
rCmd = "Rscript" + " " + fullRScriptName + " " + inputDir() + " " + expectedDir();
//generate actual dataset
double[][] A = getRandomMatrix(rowsA, colsA, 0, 1, sparsityA, 7);
writeInputMatrix("A", A, true);
double[][] B = getRandomMatrix(rowsB, 1, 0, 1, sparsity1, 3);
writeInputMatrix("B", B, true);
boolean exceptionExpected = false;
runTest(true, exceptionExpected, null, -1);
runRScript(true);
//compare matrices
HashMap<CellIndex, Double> dmlfile = readDMLMatrixFromHDFS("C");
HashMap<CellIndex, Double> rfile = readRMatrixFromFS("C");
TestUtils.compareMatrices(dmlfile, rfile, eps, "Stat-DML", "Stat-R");
}
finally
{
rtplatform = platformOld;
}
}
private void runVectorMatrixMultiplicationTest( boolean sparseM1, boolean sparseM2, ExecType instType)
{
//setup exec type, rows, cols
//rtplatform for MR
ExecMode platformOld = rtplatform;
rtplatform = ExecMode.HYBRID;
try
{
TestConfiguration config = getTestConfiguration(TEST_NAME);
double sparsityA = sparseM1?sparsity2:sparsity1;
double sparsityB = sparseM2?sparsity2:sparsity1;
String TEST_CACHE_DIR = "";
if (TEST_CACHE_ENABLED) {
TEST_CACHE_DIR = "vm" + String.valueOf(sparsityA) + "_" + String.valueOf(sparsityB) + "/";
}
loadTestConfiguration(config, TEST_CACHE_DIR);
/* This is for running the junit test the new way, i.e., construct the arguments directly */
String HOME = SCRIPT_DIR + TEST_DIR;
fullDMLScriptName = HOME + TEST_NAME + ".dml";
programArgs = new String[]{"-args", input("A"), Integer.toString(1), Integer.toString(colsA),
input("B"), Integer.toString(rowsB), Integer.toString(colsB), output("C") };
fullRScriptName = HOME + TEST_NAME + ".R";
rCmd = "Rscript" + " " + fullRScriptName + " " + inputDir() + " " + expectedDir();
//generate actual dataset
double[][] A = getRandomMatrix(1, colsA, 0, 1, sparsityA, 7);
writeInputMatrix("A", A, true);
double[][] B = getRandomMatrix(rowsB, colsB, 0, 1, sparsityB, 3);
writeInputMatrix("B", B, true);
setOutputBuffering(false);
runTest(true, false, null, -1);
runRScript(true);
//compare matrices
HashMap<CellIndex, Double> dmlfile = readDMLMatrixFromHDFS("C");
HashMap<CellIndex, Double> rfile = readRMatrixFromFS("C");
TestUtils.compareMatrices(dmlfile, rfile, eps, "Stat-DML", "Stat-R");
}
finally
{
rtplatform = platformOld;
}
}
private void runVectorVectorMultiplicationTest( boolean sparseM1, boolean outer, ExecType instType)
{
//setup exec type, rows, cols
//rtplatform for MR
ExecMode platformOld = rtplatform;
rtplatform = ExecMode.HYBRID;
try
{
TestConfiguration config = getTestConfiguration(TEST_NAME);
int rows1 = outer?colsA:1;
int rows2 = outer?1:rowsB;
int cols1 = outer?1:colsA;
int cols2 = outer?rowsB:1;
double sparsityA = sparseM1?sparsity2:sparsity1;
String TEST_CACHE_DIR = "";
if (TEST_CACHE_ENABLED) {
TEST_CACHE_DIR = "vv" + rows1 + "_" + cols1 + "_" + rows2 + "_" + cols2 + "_" + String.valueOf(sparsityA) + "/";
}
loadTestConfiguration(config, TEST_CACHE_DIR);
/* This is for running the junit test the new way, i.e., construct the arguments directly */
String HOME = SCRIPT_DIR + TEST_DIR;
fullDMLScriptName = HOME + TEST_NAME + ".dml";
programArgs = new String[]{"-args", input("A"), Integer.toString(rows1), Integer.toString(cols1),
input("B"), Integer.toString(rows2), Integer.toString(cols2), output("C") };
fullRScriptName = HOME + TEST_NAME + ".R";
rCmd = "Rscript" + " " + fullRScriptName + " " + inputDir() + " " + expectedDir();
//generate actual dataset
double[][] A = getRandomMatrix(rows1, cols1, 0, 1, sparsityA, 7);
writeInputMatrix("A", A, true);
double[][] B = getRandomMatrix(rows2, cols2, 0, 1, sparsity1, 3);
writeInputMatrix("B", B, true);
setOutputBuffering(false);
runTest(true, false, null, -1);
runRScript(true);
//compare matrices
HashMap<CellIndex, Double> dmlfile = readDMLMatrixFromHDFS("C");
HashMap<CellIndex, Double> rfile = readRMatrixFromFS("C");
TestUtils.compareMatrices(dmlfile, rfile, eps, "Stat-DML", "Stat-R");
}
finally
{
rtplatform = platformOld;
}
}
}