public static IList <Tree> GetTopParsesForOneTree(LexicalizedParser parser, int dvKBest, Tree tree, ITreeTransformer transformer)
{
IParserQuery pq = parser.ParserQuery();
IList <Word> sentence = tree.YieldWords();
// Since the trees are binarized and otherwise manipulated, we
// need to chop off the last word in order to remove the end of
// sentence symbol
if (sentence.Count <= 1)
{
return(null);
}
sentence = sentence.SubList(0, sentence.Count - 1);
if (!pq.Parse(sentence))
{
log.Info("Failed to use the given parser to reparse sentence \"" + sentence + "\"");
return(null);
}
IList <Tree> parses = new List <Tree>();
IList <ScoredObject <Tree> > bestKParses = pq.GetKBestPCFGParses(dvKBest);
foreach (ScoredObject <Tree> so in bestKParses)
{
Tree result = so.Object();
if (transformer != null)
{
result = transformer.TransformTree(result);
}
parses.Add(result);
}
return(parses);
}
/// <exception cref="System.IO.IOException"/>
public static void Main(string[] args)
{
string modelPath = null;
string outputPath = null;
string inputPath = null;
string testTreebankPath = null;
IFileFilter testTreebankFilter = null;
IList <string> unusedArgs = Generics.NewArrayList();
for (int argIndex = 0; argIndex < args.Length;)
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-model"))
{
modelPath = args[argIndex + 1];
argIndex += 2;
}
else
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-output"))
{
outputPath = args[argIndex + 1];
argIndex += 2;
}
else
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-input"))
{
inputPath = args[argIndex + 1];
argIndex += 2;
}
else
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-testTreebank"))
{
Pair <string, IFileFilter> treebankDescription = ArgUtils.GetTreebankDescription(args, argIndex, "-testTreebank");
argIndex = argIndex + ArgUtils.NumSubArgs(args, argIndex) + 1;
testTreebankPath = treebankDescription.First();
testTreebankFilter = treebankDescription.Second();
}
else
{
unusedArgs.Add(args[argIndex++]);
}
}
}
}
}
string[] newArgs = Sharpen.Collections.ToArray(unusedArgs, new string[unusedArgs.Count]);
LexicalizedParser parser = ((LexicalizedParser)LexicalizedParser.LoadModel(modelPath, newArgs));
DVModel model = DVParser.GetModelFromLexicalizedParser(parser);
File outputFile = new File(outputPath);
FileSystem.CheckNotExistsOrFail(outputFile);
FileSystem.MkdirOrFail(outputFile);
int count = 0;
if (inputPath != null)
{
Reader input = new BufferedReader(new FileReader(inputPath));
DocumentPreprocessor processor = new DocumentPreprocessor(input);
foreach (IList <IHasWord> sentence in processor)
{
count++;
// index from 1
IParserQuery pq = parser.ParserQuery();
if (!(pq is RerankingParserQuery))
{
throw new ArgumentException("Expected a RerankingParserQuery");
}
RerankingParserQuery rpq = (RerankingParserQuery)pq;
if (!rpq.Parse(sentence))
{
throw new Exception("Unparsable sentence: " + sentence);
}
IRerankerQuery reranker = rpq.RerankerQuery();
if (!(reranker is DVModelReranker.Query))
{
throw new ArgumentException("Expected a DVModelReranker");
}
DeepTree deepTree = ((DVModelReranker.Query)reranker).GetDeepTrees()[0];
IdentityHashMap <Tree, SimpleMatrix> vectors = deepTree.GetVectors();
foreach (KeyValuePair <Tree, SimpleMatrix> entry in vectors)
{
log.Info(entry.Key + " " + entry.Value);
}
FileWriter fout = new FileWriter(outputPath + File.separator + "sentence" + count + ".txt");
BufferedWriter bout = new BufferedWriter(fout);
bout.Write(SentenceUtils.ListToString(sentence));
bout.NewLine();
bout.Write(deepTree.GetTree().ToString());
bout.NewLine();
foreach (IHasWord word in sentence)
{
OutputMatrix(bout, model.GetWordVector(word.Word()));
}
Tree rootTree = FindRootTree(vectors);
OutputTreeMatrices(bout, rootTree, vectors);
bout.Flush();
fout.Close();
}
}
}
public virtual void ParseFiles <_T0>(string[] args, int argIndex, bool tokenized, ITokenizerFactory <_T0> tokenizerFactory, string elementDelimiter, string sentenceDelimiter, IFunction <IList <IHasWord>, IList <IHasWord> > escaper, string tagDelimiter
)
where _T0 : IHasWord
{
DocumentPreprocessor.DocType docType = (elementDelimiter == null) ? DocumentPreprocessor.DocType.Plain : DocumentPreprocessor.DocType.Xml;
if (op.testOptions.verbose)
{
if (tokenizerFactory != null)
{
pwErr.Println("parseFiles: Tokenizer factory is: " + tokenizerFactory);
}
}
Timing timer = new Timing();
// timer.start(); // constructor already starts it.
//Loop over the files
for (int i = argIndex; i < args.Length; i++)
{
string filename = args[i];
DocumentPreprocessor documentPreprocessor;
if (filename.Equals("-"))
{
try
{
documentPreprocessor = new DocumentPreprocessor(IOUtils.ReaderFromStdin(op.tlpParams.GetInputEncoding()), docType);
}
catch (IOException e)
{
throw new RuntimeIOException(e);
}
}
else
{
documentPreprocessor = new DocumentPreprocessor(filename, docType, op.tlpParams.GetInputEncoding());
}
//Unused values are null per the main() method invocation below
//null is the default for these properties
documentPreprocessor.SetSentenceFinalPuncWords(tlp.SentenceFinalPunctuationWords());
documentPreprocessor.SetEscaper(escaper);
documentPreprocessor.SetSentenceDelimiter(sentenceDelimiter);
documentPreprocessor.SetTagDelimiter(tagDelimiter);
documentPreprocessor.SetElementDelimiter(elementDelimiter);
if (tokenizerFactory == null)
{
documentPreprocessor.SetTokenizerFactory((tokenized) ? null : tlp.GetTokenizerFactory());
}
else
{
documentPreprocessor.SetTokenizerFactory(tokenizerFactory);
}
//Setup the output
PrintWriter pwo = pwOut;
if (op.testOptions.writeOutputFiles)
{
string normalizedName = filename;
try
{
new URL(normalizedName);
// this will exception if not a URL
normalizedName = normalizedName.ReplaceAll("/", "_");
}
catch (MalformedURLException)
{
}
//It isn't a URL, so silently ignore
string ext = (op.testOptions.outputFilesExtension == null) ? "stp" : op.testOptions.outputFilesExtension;
string fname = normalizedName + '.' + ext;
if (op.testOptions.outputFilesDirectory != null && !op.testOptions.outputFilesDirectory.IsEmpty())
{
string fseparator = Runtime.GetProperty("file.separator");
if (fseparator == null || fseparator.IsEmpty())
{
fseparator = "/";
}
File fnameFile = new File(fname);
fname = op.testOptions.outputFilesDirectory + fseparator + fnameFile.GetName();
}
try
{
pwo = op.tlpParams.Pw(new FileOutputStream(fname));
}
catch (IOException ioe)
{
throw new RuntimeIOException(ioe);
}
}
treePrint.PrintHeader(pwo, op.tlpParams.GetOutputEncoding());
pwErr.Println("Parsing file: " + filename);
int num = 0;
int numProcessed = 0;
if (op.testOptions.testingThreads != 1)
{
MulticoreWrapper <IList <IHasWord>, IParserQuery> wrapper = new MulticoreWrapper <IList <IHasWord>, IParserQuery>(op.testOptions.testingThreads, new ParsingThreadsafeProcessor(pqFactory, pwErr));
foreach (IList <IHasWord> sentence in documentPreprocessor)
{
num++;
numSents++;
int len = sentence.Count;
numWords += len;
pwErr.Println("Parsing [sent. " + num + " len. " + len + "]: " + SentenceUtils.ListToString(sentence, true));
wrapper.Put(sentence);
while (wrapper.Peek())
{
IParserQuery pq = wrapper.Poll();
ProcessResults(pq, numProcessed++, pwo);
}
}
wrapper.Join();
while (wrapper.Peek())
{
IParserQuery pq = wrapper.Poll();
ProcessResults(pq, numProcessed++, pwo);
}
}
else
{
IParserQuery pq = pqFactory.ParserQuery();
foreach (IList <IHasWord> sentence in documentPreprocessor)
{
num++;
numSents++;
int len = sentence.Count;
numWords += len;
pwErr.Println("Parsing [sent. " + num + " len. " + len + "]: " + SentenceUtils.ListToString(sentence, true));
pq.ParseAndReport(sentence, pwErr);
ProcessResults(pq, numProcessed++, pwo);
}
}
treePrint.PrintFooter(pwo);
if (op.testOptions.writeOutputFiles)
{
pwo.Close();
}
pwErr.Println("Parsed file: " + filename + " [" + num + " sentences].");
}
long millis = timer.Stop();
if (summary)
{
if (pcfgLL != null)
{
pcfgLL.Display(false, pwErr);
}
if (depLL != null)
{
depLL.Display(false, pwErr);
}
if (factLL != null)
{
factLL.Display(false, pwErr);
}
}
if (saidMemMessage)
{
ParserUtils.PrintOutOfMemory(pwErr);
}
double wordspersec = numWords / (((double)millis) / 1000);
double sentspersec = numSents / (((double)millis) / 1000);
NumberFormat nf = new DecimalFormat("0.00");
// easier way!
pwErr.Println("Parsed " + numWords + " words in " + numSents + " sentences (" + nf.Format(wordspersec) + " wds/sec; " + nf.Format(sentspersec) + " sents/sec).");
if (numFallback > 0)
{
pwErr.Println(" " + numFallback + " sentences were parsed by fallback to PCFG.");
}
if (numUnparsable > 0 || numNoMemory > 0 || numSkipped > 0)
{
pwErr.Println(" " + (numUnparsable + numNoMemory + numSkipped) + " sentences were not parsed:");
if (numUnparsable > 0)
{
pwErr.Println(" " + numUnparsable + " were not parsable with non-zero probability.");
}
if (numNoMemory > 0)
{
pwErr.Println(" " + numNoMemory + " were skipped because of insufficient memory.");
}
if (numSkipped > 0)
{
pwErr.Println(" " + numSkipped + " were skipped as length 0 or greater than " + op.testOptions.maxLength);
}
}
}
/// <exception cref="System.Exception"/>
public static void Main(string[] args)
{
string modelPath = null;
string outputPath = null;
string testTreebankPath = null;
IFileFilter testTreebankFilter = null;
IList <string> unusedArgs = new List <string>();
for (int argIndex = 0; argIndex < args.Length;)
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-model"))
{
modelPath = args[argIndex + 1];
argIndex += 2;
}
else
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-testTreebank"))
{
Pair <string, IFileFilter> treebankDescription = ArgUtils.GetTreebankDescription(args, argIndex, "-testTreebank");
argIndex = argIndex + ArgUtils.NumSubArgs(args, argIndex) + 1;
testTreebankPath = treebankDescription.First();
testTreebankFilter = treebankDescription.Second();
}
else
{
if (Sharpen.Runtime.EqualsIgnoreCase(args[argIndex], "-output"))
{
outputPath = args[argIndex + 1];
argIndex += 2;
}
else
{
unusedArgs.Add(args[argIndex++]);
}
}
}
}
if (modelPath == null)
{
throw new ArgumentException("Need to specify -model");
}
if (testTreebankPath == null)
{
throw new ArgumentException("Need to specify -testTreebank");
}
if (outputPath == null)
{
throw new ArgumentException("Need to specify -output");
}
string[] newArgs = Sharpen.Collections.ToArray(unusedArgs, new string[unusedArgs.Count]);
LexicalizedParser lexparser = ((LexicalizedParser)LexicalizedParser.LoadModel(modelPath, newArgs));
Treebank testTreebank = null;
if (testTreebankPath != null)
{
log.Info("Reading in trees from " + testTreebankPath);
if (testTreebankFilter != null)
{
log.Info("Filtering on " + testTreebankFilter);
}
testTreebank = lexparser.GetOp().tlpParams.MemoryTreebank();
testTreebank.LoadPath(testTreebankPath, testTreebankFilter);
log.Info("Read in " + testTreebank.Count + " trees for testing");
}
FileWriter @out = new FileWriter(outputPath);
BufferedWriter bout = new BufferedWriter(@out);
log.Info("Parsing " + testTreebank.Count + " trees");
int count = 0;
IList <FindNearestNeighbors.ParseRecord> records = Generics.NewArrayList();
foreach (Tree goldTree in testTreebank)
{
IList <Word> tokens = goldTree.YieldWords();
IParserQuery parserQuery = lexparser.ParserQuery();
if (!parserQuery.Parse(tokens))
{
throw new AssertionError("Could not parse: " + tokens);
}
if (!(parserQuery is RerankingParserQuery))
{
throw new ArgumentException("Expected a LexicalizedParser with a Reranker attached");
}
RerankingParserQuery rpq = (RerankingParserQuery)parserQuery;
if (!(rpq.RerankerQuery() is DVModelReranker.Query))
{
throw new ArgumentException("Expected a LexicalizedParser with a DVModel attached");
}
DeepTree tree = ((DVModelReranker.Query)rpq.RerankerQuery()).GetDeepTrees()[0];
SimpleMatrix rootVector = null;
foreach (KeyValuePair <Tree, SimpleMatrix> entry in tree.GetVectors())
{
if (entry.Key.Label().Value().Equals("ROOT"))
{
rootVector = entry.Value;
break;
}
}
if (rootVector == null)
{
throw new AssertionError("Could not find root nodevector");
}
@out.Write(tokens + "\n");
@out.Write(tree.GetTree() + "\n");
for (int i = 0; i < rootVector.GetNumElements(); ++i)
{
@out.Write(" " + rootVector.Get(i));
}
@out.Write("\n\n\n");
count++;
if (count % 10 == 0)
{
log.Info(" " + count);
}
records.Add(new FindNearestNeighbors.ParseRecord(tokens, goldTree, tree.GetTree(), rootVector, tree.GetVectors()));
}
log.Info(" done parsing");
IList <Pair <Tree, SimpleMatrix> > subtrees = Generics.NewArrayList();
foreach (FindNearestNeighbors.ParseRecord record in records)
{
foreach (KeyValuePair <Tree, SimpleMatrix> entry in record.nodeVectors)
{
if (entry.Key.GetLeaves().Count <= maxLength)
{
subtrees.Add(Pair.MakePair(entry.Key, entry.Value));
}
}
}
log.Info("There are " + subtrees.Count + " subtrees in the set of trees");
PriorityQueue <ScoredObject <Pair <Tree, Tree> > > bestmatches = new PriorityQueue <ScoredObject <Pair <Tree, Tree> > >(101, ScoredComparator.DescendingComparator);
for (int i_1 = 0; i_1 < subtrees.Count; ++i_1)
{
log.Info(subtrees[i_1].First().YieldWords());
log.Info(subtrees[i_1].First());
for (int j = 0; j < subtrees.Count; ++j)
{
if (i_1 == j)
{
continue;
}
// TODO: look at basic category?
double normF = subtrees[i_1].Second().Minus(subtrees[j].Second()).NormF();
bestmatches.Add(new ScoredObject <Pair <Tree, Tree> >(Pair.MakePair(subtrees[i_1].First(), subtrees[j].First()), normF));
if (bestmatches.Count > 100)
{
bestmatches.Poll();
}
}
IList <ScoredObject <Pair <Tree, Tree> > > ordered = Generics.NewArrayList();
while (bestmatches.Count > 0)
{
ordered.Add(bestmatches.Poll());
}
Java.Util.Collections.Reverse(ordered);
foreach (ScoredObject <Pair <Tree, Tree> > pair in ordered)
{
log.Info(" MATCHED " + pair.Object().second.YieldWords() + " ... " + pair.Object().Second() + " with a score of " + pair.Score());
}
log.Info();
log.Info();
bestmatches.Clear();
}
/*
* for (int i = 0; i < records.size(); ++i) {
* if (i % 10 == 0) {
* log.info(" " + i);
* }
* List<ScoredObject<ParseRecord>> scored = Generics.newArrayList();
* for (int j = 0; j < records.size(); ++j) {
* if (i == j) continue;
*
* double score = 0.0;
* int matches = 0;
* for (Map.Entry<Tree, SimpleMatrix> first : records.get(i).nodeVectors.entrySet()) {
* for (Map.Entry<Tree, SimpleMatrix> second : records.get(j).nodeVectors.entrySet()) {
* String firstBasic = dvparser.dvModel.basicCategory(first.getKey().label().value());
* String secondBasic = dvparser.dvModel.basicCategory(second.getKey().label().value());
* if (firstBasic.equals(secondBasic)) {
++matches;
* double normF = first.getValue().minus(second.getValue()).normF();
* score += normF * normF;
* }
* }
* }
* if (matches == 0) {
* score = Double.POSITIVE_INFINITY;
* } else {
* score = score / matches;
* }
* //double score = records.get(i).vector.minus(records.get(j).vector).normF();
* scored.add(new ScoredObject<ParseRecord>(records.get(j), score));
* }
* Collections.sort(scored, ScoredComparator.ASCENDING_COMPARATOR);
*
* out.write(records.get(i).sentence.toString() + "\n");
* for (int j = 0; j < numNeighbors; ++j) {
* out.write(" " + scored.get(j).score() + ": " + scored.get(j).object().sentence + "\n");
* }
* out.write("\n\n");
* }
* log.info();
*/
bout.Flush();
@out.Flush();
@out.Close();
}