public void testOne () { String input = "abcdefghijklmnopqrstuvwxyz"; Pipe p = new SerialPipes (new Pipe[] { new CharSequence2TokenSequence ("."), //new PrintInput("1:"), new TokenSequenceLowercase (), //new PrintInput("2:"), new TokenText (), //new PrintInput("3:"), new RegexMatches ("V", Pattern.compile("[aeiou]")), //new PrintInput("4:"), new OffsetConjunctions (new int[][] {{0,0}, {0,1}, {-1,0,1}, {-1}, {-2}}), new PrintInput("5:"), }); Instance carrier = p.instanceFrom(new Instance (input, null, null, null)); TokenSequence ts = (TokenSequence) carrier.getData(); assertTrue (ts.size() == 26); assertTrue (ts.get(0).getFeatureValue("a_&_b@1") == 1.0); assertTrue (ts.get(0).getFeatureValue("V_&_a") == 1.0); assertTrue (ts.get(2).getFeatureValue("b@-1_&_c_&_d@1") == 1.0); }
public void testOne () { String input = "abcdefghijklmnopqrstuvwxyz"; Pipe p = new SerialPipes (new Pipe[] { new CharSequence2TokenSequence ("."), //new PrintInput("1:"), new TokenSequenceLowercase (), //new PrintInput("2:"), new TokenText (), //new PrintInput("3:"), new RegexMatches ("V", Pattern.compile("[aeiou]")), //new PrintInput("4:"), new OffsetConjunctions (new int[][] {{0,0}, {0,1}, {-1,0,1}, {-1}, {-2}}), new PrintInput("5:"), }); Instance carrier = p.instanceFrom(new Instance (input, null, null, null)); TokenSequence ts = (TokenSequence) carrier.getData(); assertTrue (ts.size() == 26); assertTrue (ts.get(0).getFeatureValue("a_&_b@1") == 1.0); assertTrue (ts.get(0).getFeatureValue("V_&_a") == 1.0); assertTrue (ts.get(2).getFeatureValue("b@-1_&_c_&_d@1") == 1.0); }
public static void testMultiTagSerialization () throws IOException, ClassNotFoundException { Pipe origPipe = new SerialPipes (new Pipe[] { new SimpleTaggerSentence2TokenSequence (), new TokenText (), new RegexMatches ("digits", Pattern.compile ("[0-9]+")), new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")), new OffsetFeatureConjunction ("time", new String[] { "digits", "ampm" }, new int[] { 0, 1 }, true), new PrintInputAndTarget (), }); Pipe mtPipe = (Pipe) TestSerializable.cloneViaSerialization (origPipe); InstanceList mtLst = new InstanceList (mtPipe); mtLst.addThruPipe (new ArrayIterator (doc1)); Instance mtInst = mtLst.get (0); TokenSequence mtTs = (TokenSequence) mtInst.getData (); assertEquals (6, mtTs.size ()); assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15); }
public static void testMultiTagSerialization () throws IOException, ClassNotFoundException { Pipe origPipe = new SerialPipes (new Pipe[] { new SimpleTaggerSentence2TokenSequence (), new TokenText (), new RegexMatches ("digits", Pattern.compile ("[0-9]+")), new RegexMatches ("ampm", Pattern.compile ("[aApP][mM]")), new OffsetFeatureConjunction ("time", new String[] { "digits", "ampm" }, new int[] { 0, 1 }, true), new PrintInputAndTarget (), }); Pipe mtPipe = (Pipe) TestSerializable.cloneViaSerialization (origPipe); InstanceList mtLst = new InstanceList (mtPipe); mtLst.addThruPipe (new ArrayIterator (doc1)); Instance mtInst = mtLst.get (0); TokenSequence mtTs = (TokenSequence) mtInst.getData (); assertEquals (6, mtTs.size ()); assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15); }
assertEquals (6, noMtTs.size ()); assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, noMtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15); assertEquals (0.0, noMtTs.get (4).getFeatureValue ("time"), 1e-15);
assertEquals (6, noMtTs.size ()); assertEquals (1.0, mtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, noMtTs.get (3).getFeatureValue ("time"), 1e-15); assertEquals (1.0, mtTs.get (4).getFeatureValue ("time"), 1e-15); assertEquals (0.0, noMtTs.get (4).getFeatureValue ("time"), 1e-15);