rowSimilarityJob.run(new String[] { "--input", inputFile.getAbsolutePath(), "--output", outputDir.getAbsolutePath(),
"--numberOfColumns", String.valueOf(5), "--similarityClassname", TanimotoCoefficientSimilarity.class.getName(),
"--tempDir", tmpDir.getAbsolutePath() });
OpenIntIntHashMap observationsPerColumn =
Vectors.readAsIntMap(new Path(tmpDir.getAbsolutePath(), "observationsPerColumn.bin"), conf);
assertEquals(4, observationsPerColumn.size());
assertEquals(1, observationsPerColumn.get(0));
assertEquals(2, observationsPerColumn.get(2));
assertEquals(2, observationsPerColumn.get(3));
assertEquals(1, observationsPerColumn.get(4));
Matrix similarityMatrix = MathHelper.readMatrix(conf, new Path(outputDir.getAbsolutePath(), "part-r-00000"), 3, 3);
assertNotNull(similarityMatrix);
assertEquals(3, similarityMatrix.numCols());