public MUCMentionExtractor(Dictionaries dict, Properties props, Semantics semantics)
throws Exception {
super(dict, semantics);
String fileName = props.getProperty(Constants.MUC_PROP);
fileContents = IOUtils.slurpFile(fileName);
currentOffset = 0;
tokenizerFactory = PTBTokenizer.factory(new CoreLabelTokenFactory(false), "");
stanfordProcessor = loadStanfordProcessor(props);
}
public static void main(String[] args) throws IOException {
String serializedClassifier = "classifiers/english.all.3class.distsim.crf.ser.gz";
if (args.length > 0) {
serializedClassifier = args[0];
}
AbstractSequenceClassifier<CoreLabel> classifier =
CRFClassifier.getClassifierNoExceptions(serializedClassifier);
/* For either a file to annotate or for the hardcoded text example,
this demo file shows two ways to process the output, for teaching
purposes. For the file, it shows both how to run NER on a String
and how to run it on a whole file. For the hard-coded String,
it shows how to run it on a single sentence, and how to do this
and produce an inline XML output format.
*/
if (args.length > 1) {
String fileContents = IOUtils.slurpFile(args[1]);
List<List<CoreLabel>> out = classifier.classify(fileContents);
for (List<CoreLabel> sentence : out) {
for (CoreLabel word : sentence) {
System.out.print(word.word() + '/' + word.get(AnswerAnnotation.class) + ' ');
}
System.out.println();
}
out = classifier.classifyFile(args[1]);
for (List<CoreLabel> sentence : out) {
for (CoreLabel word : sentence) {
System.out.print(word.word() + '/' + word.get(AnswerAnnotation.class) + ' ');
}
System.out.println();
}
} else {
String s1 = "Good afternoon Rajat Raina, how are you today?";
String s2 = "I go to school at Stanford University, which is located in California.";
System.out.println(classifier.classifyToString(s1));
System.out.println(classifier.classifyWithInlineXML(s2));
System.out.println(classifier.classifyToString(s2, "xml", true));
int i = 0;
for (List<CoreLabel> lcl : classifier.classify(s2)) {
for (CoreLabel cl : lcl) {
System.out.println(i++ + ":");
System.out.println(cl);
}
}
}
}
/**
* Prints out all matches of a tree pattern on each tree in the path. Usage: <br>
* <br>
* <code>
* java edu.stanford.nlp.trees.tregex.TregexPattern [[-TCwfosnu] [-filter] [-h <node-name>]]* pattern
* filepath </code>
*
* <p>Arguments:<br>
*
* <ul>
* <li><code>pattern</code>: the tree pattern which optionally names some set of nodes (i.e.,
* gives it the "handle") <code>=name</code> (for some arbitrary string "name")
* <li><code>filepath</code>: the path to files with trees. If this is a directory, there will
* be recursive descent and the pattern will be run on all files beneath the specified
* directory.
* </ul>
*
* <p>Options:<br>
* <li><code>-C</code> suppresses printing of matches, so only the number of matches is printed.
* <li><code>-w</code> causes the whole of a tree that matches to be printed.
* <li><code>-f</code> causes the filename to be printed.
* <li><code>-i <filename></code> causes the pattern to be matched to be read from <code>
* <filename></code> rather than the command line. Don't specify a pattern when this
* option is used.
* <li><code>-o</code> Specifies that each tree node can be reported only once as the root of a
* match (by default a node will be printed once for every <em>way</em> the pattern matches).
* <li><code>-s</code> causes trees to be printed all on one line (by default they are pretty
* printed).
* <li><code>-n</code> causes the number of the tree in which the match was found to be printed
* before every match.
* <li><code>-u</code> causes only the label of each matching node to be printed, not complete
* subtrees.
* <li><code>-t</code> causes only the yield (terminal words) of the selected node to be printed
* (or the yield of the whole tree, if the <code>-w</code> option is used).
* <li><code>-encoding <charset_encoding></code> option allows specification of character
* encoding of trees..
* <li><code>-h <node-handle></code> If a <code>-h</code> option is given, the root tree
* node will not be printed. Instead, for each <code>node-handle</code> specified, the node
* matched and given that handle will be printed. Multiple nodes can be printed by using the
* <code>-h</code> option multiple times on a single command line.
* <li><code>-hf <headfinder-class-name></code> use the specified {@link HeadFinder} class
* to determine headship relations.
* <li><code>-hfArg <string></code> pass a string argument in to the {@link HeadFinder}
* class's constructor. <code>-hfArg</code> can be used multiple times to pass in multiple
* arguments.
* <li><code>-trf <TreeReaderFactory-class-name></code> use the specified {@link
* TreeReaderFactory} class to read trees from files.
* <li><code>-v</code> print every tree that contains no matches of the specified pattern, but
* print no matches to the pattern.
* <li><code>-x</code> Instead of the matched subtree, print the matched subtree's identifying
* number as defined in <tt>tgrep2</tt>:a unique identifier for the subtree and is in the form
* s:n, where s is an integer specifying the sentence number in the corpus (starting with 1),
* and n is an integer giving the order in which the node is encountered in a depth-first
* search starting with 1 at top node in the sentence tree.
* <li><code>-extract <code> <tree-file></code> extracts the subtree s:n specified by
* <tt>code</tt> from the specified <tt>tree-file</tt>. Overrides all other behavior of
* tregex. Can't specify multiple encodings etc. yet.
* <li><code>-extractFile <code-file> <tree-file></code> extracts every subtree
* specified by the subtree codes in <tt>code-file</tt>, which must appear exactly one per
* line, from the specified <tt>tree-file</tt>. Overrides all other behavior of tregex. Can't
* specify multiple encodings etc. yet.
* <li><code>-filter</code> causes this to act as a filter, reading tree input from stdin
* <li><code>-T</code> causes all trees to be printed as processed (for debugging purposes).
* Otherwise only matching nodes are printed.
* <li><code>-macros <filename></code> filename with macro substitutions to use. file with
* tab separated lines original-tab-replacement
* </ul>
*/
public static void main(String[] args) throws IOException {
Timing.startTime();
StringBuilder treePrintFormats = new StringBuilder();
String printNonMatchingTreesOption = "-v";
String subtreeCodeOption = "-x";
String extractSubtreesOption = "-extract";
String extractSubtreesFileOption = "-extractFile";
String inputFileOption = "-i";
String headFinderOption = "-hf";
String headFinderArgOption = "-hfArg";
String trfOption = "-trf";
String headFinderClassName = null;
String[] headFinderArgs = StringUtils.EMPTY_STRING_ARRAY;
String treeReaderFactoryClassName = null;
String printHandleOption = "-h";
String markHandleOption = "-k";
String encodingOption = "-encoding";
String encoding = "UTF-8";
String macroOption = "-macros";
String macroFilename = "";
String yieldOnly = "-t";
String printAllTrees = "-T";
String quietMode = "-C";
String wholeTreeMode = "-w";
String filenameOption = "-f";
String oneMatchPerRootNodeMode = "-o";
String reportTreeNumbers = "-n";
String rootLabelOnly = "-u";
String oneLine = "-s";
Map<String, Integer> flagMap = Generics.newHashMap();
flagMap.put(extractSubtreesOption, 2);
flagMap.put(extractSubtreesFileOption, 2);
flagMap.put(subtreeCodeOption, 0);
flagMap.put(printNonMatchingTreesOption, 0);
flagMap.put(encodingOption, 1);
flagMap.put(inputFileOption, 1);
flagMap.put(printHandleOption, 1);
flagMap.put(markHandleOption, 2);
flagMap.put(headFinderOption, 1);
flagMap.put(headFinderArgOption, 1);
flagMap.put(trfOption, 1);
flagMap.put(macroOption, 1);
flagMap.put(yieldOnly, 0);
flagMap.put(quietMode, 0);
flagMap.put(wholeTreeMode, 0);
flagMap.put(printAllTrees, 0);
flagMap.put(filenameOption, 0);
flagMap.put(oneMatchPerRootNodeMode, 0);
flagMap.put(reportTreeNumbers, 0);
flagMap.put(rootLabelOnly, 0);
flagMap.put(oneLine, 0);
Map<String, String[]> argsMap = StringUtils.argsToMap(args, flagMap);
args = argsMap.get(null);
if (argsMap.containsKey(encodingOption)) {
encoding = argsMap.get(encodingOption)[0];
System.err.println("Encoding set to " + encoding);
}
PrintWriter errPW = new PrintWriter(new OutputStreamWriter(System.err, encoding), true);
if (argsMap.containsKey(extractSubtreesOption)) {
List<String> subTreeStrings =
Collections.singletonList(argsMap.get(extractSubtreesOption)[0]);
extractSubtrees(subTreeStrings, argsMap.get(extractSubtreesOption)[1]);
return;
}
if (argsMap.containsKey(extractSubtreesFileOption)) {
List<String> subTreeStrings =
Arrays.asList(
IOUtils.slurpFile(argsMap.get(extractSubtreesFileOption)[0]).split("\n|\r|\n\r"));
extractSubtrees(subTreeStrings, argsMap.get(extractSubtreesFileOption)[0]);
return;
}
if (args.length < 1) {
errPW.println(
"Usage: java edu.stanford.nlp.trees.tregex.TregexPattern [-T] [-C] [-w] [-f] [-o] [-n] [-s] [-filter] [-hf class] [-trf class] [-h handle]* pattern [filepath]");
return;
}
String matchString = args[0];
if (argsMap.containsKey(macroOption)) {
macroFilename = argsMap.get(macroOption)[0];
}
if (argsMap.containsKey(headFinderOption)) {
headFinderClassName = argsMap.get(headFinderOption)[0];
errPW.println("Using head finder " + headFinderClassName + "...");
}
if (argsMap.containsKey(headFinderArgOption)) {
headFinderArgs = argsMap.get(headFinderArgOption);
}
if (argsMap.containsKey(trfOption)) {
treeReaderFactoryClassName = argsMap.get(trfOption)[0];
errPW.println("Using tree reader factory " + treeReaderFactoryClassName + "...");
}
if (argsMap.containsKey(printAllTrees)) {
TRegexTreeVisitor.printTree = true;
}
if (argsMap.containsKey(inputFileOption)) {
String inputFile = argsMap.get(inputFileOption)[0];
matchString = IOUtils.slurpFile(inputFile, encoding);
String[] newArgs = new String[args.length + 1];
System.arraycopy(args, 0, newArgs, 1, args.length);
args = newArgs;
}
if (argsMap.containsKey(quietMode)) {
TRegexTreeVisitor.printMatches = false;
TRegexTreeVisitor.printNumMatchesToStdOut = true;
}
if (argsMap.containsKey(printNonMatchingTreesOption)) {
TRegexTreeVisitor.printNonMatchingTrees = true;
}
if (argsMap.containsKey(subtreeCodeOption)) {
TRegexTreeVisitor.printSubtreeCode = true;
TRegexTreeVisitor.printMatches = false;
}
if (argsMap.containsKey(wholeTreeMode)) {
TRegexTreeVisitor.printWholeTree = true;
}
if (argsMap.containsKey(filenameOption)) {
TRegexTreeVisitor.printFilename = true;
}
if (argsMap.containsKey(oneMatchPerRootNodeMode)) TRegexTreeVisitor.oneMatchPerRootNode = true;
if (argsMap.containsKey(reportTreeNumbers)) TRegexTreeVisitor.reportTreeNumbers = true;
if (argsMap.containsKey(rootLabelOnly)) {
treePrintFormats.append(TreePrint.rootLabelOnlyFormat).append(',');
} else if (argsMap.containsKey(oneLine)) { // display short form
treePrintFormats.append("oneline,");
} else if (argsMap.containsKey(yieldOnly)) {
treePrintFormats.append("words,");
} else {
treePrintFormats.append("penn,");
}
HeadFinder hf = new CollinsHeadFinder();
if (headFinderClassName != null) {
Class[] hfArgClasses = new Class[headFinderArgs.length];
for (int i = 0; i < hfArgClasses.length; i++) hfArgClasses[i] = String.class;
try {
hf =
(HeadFinder)
Class.forName(headFinderClassName)
.getConstructor(hfArgClasses)
.newInstance(
(Object[])
headFinderArgs); // cast to Object[] necessary to avoid varargs-related
// warning.
} catch (Exception e) {
throw new RuntimeException("Error occurred while constructing HeadFinder: " + e);
}
}
TRegexTreeVisitor.tp =
new TreePrint(treePrintFormats.toString(), new PennTreebankLanguagePack());
try {
// TreePattern p = TreePattern.compile("/^S/ > S=dt $++ '' $-- ``");
TregexPatternCompiler tpc = new TregexPatternCompiler(hf);
Macros.addAllMacros(tpc, macroFilename, encoding);
TregexPattern p = tpc.compile(matchString);
errPW.println("Pattern string:\n" + p.pattern());
errPW.println("Parsed representation:");
p.prettyPrint(errPW);
String[] handles = argsMap.get(printHandleOption);
if (argsMap.containsKey("-filter")) {
TreeReaderFactory trf = getTreeReaderFactory(treeReaderFactoryClassName);
treebank =
new MemoryTreebank(
trf, encoding); // has to be in memory since we're not storing it on disk
// read from stdin
Reader reader = new BufferedReader(new InputStreamReader(System.in, encoding));
((MemoryTreebank) treebank).load(reader);
reader.close();
} else if (args.length == 1) {
errPW.println("using default tree");
TreeReader r =
new PennTreeReader(
new StringReader(
"(VP (VP (VBZ Try) (NP (NP (DT this) (NN wine)) (CC and) (NP (DT these) (NNS snails)))) (PUNCT .))"),
new LabeledScoredTreeFactory(new StringLabelFactory()));
Tree t = r.readTree();
treebank = new MemoryTreebank();
treebank.add(t);
} else {
int last = args.length - 1;
errPW.println("Reading trees from file(s) " + args[last]);
TreeReaderFactory trf = getTreeReaderFactory(treeReaderFactoryClassName);
treebank = new DiskTreebank(trf, encoding);
treebank.loadPath(args[last], null, true);
}
TRegexTreeVisitor vis = new TRegexTreeVisitor(p, handles, encoding);
treebank.apply(vis);
Timing.endTime();
if (TRegexTreeVisitor.printMatches) {
errPW.println("There were " + vis.numMatches() + " matches in total.");
}
if (TRegexTreeVisitor.printNumMatchesToStdOut) {
System.out.println(vis.numMatches());
}
} catch (IOException e) {
e.printStackTrace();
} catch (TregexParseException e) {
errPW.println("Error parsing expression: " + args[0]);
errPW.println("Parse exception: " + e.toString());
}
}
public static void main(String[] args) throws Exception {
// String serializedClassifier = "classifiers/english.all.3class.distsim.crf.ser.gz";
String serializedClassifier = "classifiers/english.muc.7class.distsim.crf.ser.gz";
if (args.length > 0) {
serializedClassifier = args[0];
}
AbstractSequenceClassifier<CoreLabel> classifier =
CRFClassifier.getClassifier(serializedClassifier);
/* For either a file to annotate or for the hardcoded text example, this
demo file shows several ways to process the input, for teaching purposes.
*/
if (args.length > 1) {
/* For the file, it shows (1) how to run NER on a String, (2) how
to get the entities in the String with character offsets, and
(3) how to run NER on a whole file (without loading it into a String).
*/
String fileContents = IOUtils.slurpFile(args[1]);
List<List<CoreLabel>> out = classifier.classify(fileContents);
for (List<CoreLabel> sentence : out) {
for (CoreLabel word : sentence) {
System.out.print(
word.word() + '/' + word.get(CoreAnnotations.AnswerAnnotation.class) + ' ');
}
System.out.println();
}
System.out.println("---");
out = classifier.classifyFile(args[1]);
for (List<CoreLabel> sentence : out) {
for (CoreLabel word : sentence) {
System.out.print(
word.word() + '/' + word.get(CoreAnnotations.AnswerAnnotation.class) + ' ');
}
System.out.println();
}
System.out.println("---");
List<Triple<String, Integer, Integer>> list =
classifier.classifyToCharacterOffsets(fileContents);
for (Triple<String, Integer, Integer> item : list) {
// print entity/or non-entity - their nearby tokens
System.out.println(
item.first() + ": " + fileContents.substring(item.second(), item.third()));
}
System.out.println("---");
System.out.println("Ten best entity labelings");
DocumentReaderAndWriter<CoreLabel> readerAndWriter =
classifier.makePlainTextReaderAndWriter();
classifier.classifyAndWriteAnswersKBest(args[1], 10, readerAndWriter);
System.out.println("---");
System.out.println("Per-token marginalized probabilities");
classifier.printProbs(args[1], readerAndWriter);
// -- This code prints out the first order (token pair) clique probabilities.
// -- But that output is a bit overwhelming, so we leave it commented out by default.
// System.out.println("---");
// System.out.println("First Order Clique Probabilities");
// ((CRFClassifier) classifier).printFirstOrderProbs(args[1], readerAndWriter);
} else {
/* For the hard-coded String, it shows how to run it on a single
sentence, and how to do this and produce several formats, including
slash tags and an inline XML output format. It also shows the full
contents of the {@code CoreLabel}s that are constructed by the
classifier. And it shows getting out the probabilities of different
assignments and an n-best list of classifications with probabilities.
*/
String[] example = {
"Good afternoon Rajat Raina, how are you today? I go to Washington DC on September 19. And Tomorrow.",
"I go to school at Stanford University, which is located in California."
};
for (String str : example) {
System.out.println(classifier.classifyToString(str));
}
System.out.println("---");
// ***sentence-by-sentence
for (String str : example) {
// This one puts in spaces and newlines between tokens, so just print not println.
System.out.print(classifier.classifyToString(str, "slashTags", false));
}
System.out.println("---");
// ***print: entities + Classes + remaining text in the text
for (String str : example) {
// This one is best for dealing with the output as a TSV (tab-separated column) file.
// The first column gives entities, the second their classes, and the third the remaining
// text in a document
System.out.print(classifier.classifyToString(str, "tabbedEntities", false));
}
System.out.println("---");
for (String str : example) {
System.out.println(classifier.classifyWithInlineXML(str));
}
System.out.println("---");
for (String str : example) {
System.out.println(classifier.classifyToString(str, "xml", true));
}
System.out.println("---");
for (String str : example) {
System.out.print(classifier.classifyToString(str, "tsv", false));
}
System.out.println("---");
// This gets out entities with character offsets
System.out.print("character offsets");
int j = 0;
for (String str : example) {
j++;
List<Triple<String, Integer, Integer>> triples = classifier.classifyToCharacterOffsets(str);
for (Triple<String, Integer, Integer> trip : triples) {
System.out.printf(
"%s over character offsets [%d, %d) in sentence %d.%n",
trip.first(), trip.second(), trip.third, j);
}
}
System.out.println("---");
// This prints out all the details of what is stored for each token
int i = 0;
for (String str : example) {
for (List<CoreLabel> lcl : classifier.classify(str)) {
for (CoreLabel cl : lcl) {
System.out.print(i++ + ": ");
System.out.println(cl.toShorterString());
}
}
}
System.out.println("---");
}
}
public void annotate(CoreMap document) throws IOException {
// write input file in GUTime format
Element inputXML = toInputXML(document);
File inputFile = File.createTempFile("gutime", ".input");
// Document doc = new Document(inputXML);
PrintWriter inputWriter = new PrintWriter(inputFile);
inputWriter.println(inputXML.toXML());
// new XMLOutputter().output(inputXML, inputWriter);
inputWriter.close();
boolean useFirstDate =
(!document.has(CoreAnnotations.CalendarAnnotation.class)
&& !document.has(CoreAnnotations.DocDateAnnotation.class));
ArrayList<String> args = new ArrayList<String>();
args.add("perl");
args.add("-I" + this.gutimePath.getPath());
args.add(new File(this.gutimePath, "TimeTag.pl").getPath());
if (useFirstDate) args.add("-FDNW");
args.add(inputFile.getPath());
// run GUTime on the input file
ProcessBuilder process = new ProcessBuilder(args);
StringWriter outputWriter = new StringWriter();
SystemUtils.run(process, outputWriter, null);
String output = outputWriter.getBuffer().toString();
Pattern docClose = Pattern.compile("</DOC>.*", Pattern.DOTALL);
output = docClose.matcher(output).replaceAll("</DOC>");
// parse the GUTime output
Element outputXML;
try {
Document newNodeDocument = new Builder().build(output, "");
outputXML = newNodeDocument.getRootElement();
} catch (ParsingException ex) {
throw new RuntimeException(
String.format(
"error:\n%s\ninput:\n%s\noutput:\n%s", ex, IOUtils.slurpFile(inputFile), output));
}
/*
try {
outputXML = new SAXBuilder().build(new StringReader(output)).getRootElement();
} catch (JDOMException e) {
throw new RuntimeException(String.format("error:\n%s\ninput:\n%s\noutput:\n%s",
e, IOUtils.slurpFile(inputFile), output));
} */
inputFile.delete();
// get Timex annotations
List<CoreMap> timexAnns = toTimexCoreMaps(outputXML, document);
document.set(TimexAnnotations.class, timexAnns);
if (outputResults) {
System.out.println(timexAnns);
}
// align Timex annotations to sentences
int timexIndex = 0;
for (CoreMap sentence : document.get(CoreAnnotations.SentencesAnnotation.class)) {
int sentBegin = beginOffset(sentence);
int sentEnd = endOffset(sentence);
// skip times before the sentence
while (timexIndex < timexAnns.size() && beginOffset(timexAnns.get(timexIndex)) < sentBegin) {
++timexIndex;
}
// determine times within the sentence
int sublistBegin = timexIndex;
int sublistEnd = timexIndex;
while (timexIndex < timexAnns.size()
&& sentBegin <= beginOffset(timexAnns.get(timexIndex))
&& endOffset(timexAnns.get(timexIndex)) <= sentEnd) {
++sublistEnd;
++timexIndex;
}
// set the sentence timexes
sentence.set(TimexAnnotations.class, timexAnns.subList(sublistBegin, sublistEnd));
}
}