How to use

PearsonDistance

in

cascading.assembly

@Test
public void testPearsonDistanceComposite() throws Exception
 {
 getPlatform().copyFromLocal( inputFileCritics );
 Tap source = getPlatform().getTextFile( inputFileCritics );
 Tap sink = getPlatform().getTextFile( new Fields( "line" ), getOutputPath( "pearson/composite" ), SinkMode.REPLACE );
 // unknown number of elements
 Pipe pipe = new Each( "pearson", new Fields( "line" ), new RegexSplitter( "\t" ) );
 // break not names and movies
 pipe = new Each( pipe, new UnGroup( new Fields( "name", "movie", "rate" ), Fields.FIRST, 2 ) );
 // name and rate against others of same movie
 pipe = new PearsonDistance( pipe, new Fields( "name", "movie", "rate" ), new Fields( "name1", "name2", "distance" ) );
 Flow flow = getPlatform().getFlowConnector().connect( source, sink, pipe );
 flow.complete();
 validateLength( flow, 21 );
 List<Tuple> results = getSinkAsList( flow );
 assertTrue( results.contains( new Tuple( "GeneSeymour\tLisaRose\t0.39605901719066977" ) ) );
 }
}

@Test
public void testPearsonDistanceComposite() throws Exception
 {
 getPlatform().copyFromLocal( inputFileCritics );
 Tap source = getPlatform().getTextFile( inputFileCritics );
 Tap sink = getPlatform().getTextFile( new Fields( "line" ), getOutputPath( "pearson/composite" ), SinkMode.REPLACE );
 // unknown number of elements
 Pipe pipe = new Each( "pearson", new Fields( "line" ), new RegexSplitter( "\t" ) );
 // break not names and movies
 pipe = new Each( pipe, new UnGroup( new Fields( "name", "movie", "rate" ), Fields.FIRST, 2 ) );
 // name and rate against others of same movie
 pipe = new PearsonDistance( pipe, new Fields( "name", "movie", "rate" ), new Fields( "name1", "name2", "distance" ) );
 Flow flow = getPlatform().getFlowConnector().connect( source, sink, pipe );
 flow.complete();
 validateLength( flow, 21 );
 List<Tuple> results = getSinkAsList( flow );
 assertTrue( results.contains( new Tuple( "GeneSeymour\tLisaRose\t0.39605901719066977" ) ) );
 }
}