/** Re-order the action space based on the specified order of names. */
private Collection<Action> orderActions(Collection<Action> actionSpace, List<String> order) {
List<Action> tmp = new ArrayList<>(actionSpace);
List<Action> out = new ArrayList<>();
for (String key : order) {
Iterator<Action> iter = tmp.iterator();
while (iter.hasNext()) {
Action a = iter.next();
if (a.signature().equals(key)) {
out.add(a);
iter.remove();
}
}
}
out.addAll(tmp);
return out;
}
/**
* The core implementation of the search.
*
* @param root The root word to search from. Traditionally, this is the root of the sentence.
* @param candidateFragments The callback for the resulting sentence fragments. This is a
* predicate of a triple of values. The return value of the predicate determines whether we
* should continue searching. The triple is a triple of
* <ol>
* <li>The log probability of the sentence fragment, according to the featurizer and the
* weights
* <li>The features along the path to this fragment. The last element of this is the
* features from the most recent step.
* <li>The sentence fragment. Because it is relatively expensive to compute the resulting
* tree, this is returned as a lazy {@link Supplier}.
* </ol>
*
* @param classifier The classifier for whether an arc should be on the path to a clause split, a
* clause split itself, or neither.
* @param featurizer The featurizer to use. Make sure this matches the weights!
* @param actionSpace The action space we are allowed to take. Each action defines a means of
* splitting a clause on a dependency boundary.
*/
protected void search(
// The root to search from
IndexedWord root,
// The output specs
final Predicate<Triple<Double, List<Counter<String>>, Supplier<SentenceFragment>>>
candidateFragments,
// The learning specs
final Classifier<ClauseSplitter.ClauseClassifierLabel, String> classifier,
Map<String, ? extends List<String>> hardCodedSplits,
final Function<Triple<State, Action, State>, Counter<String>> featurizer,
final Collection<Action> actionSpace,
final int maxTicks) {
// (the fringe)
PriorityQueue<Pair<State, List<Counter<String>>>> fringe = new FixedPrioritiesPriorityQueue<>();
// (avoid duplicate work)
Set<IndexedWord> seenWords = new HashSet<>();
State firstState =
new State(null, null, -9000, null, x -> {}, true); // First state is implicitly "done"
fringe.add(Pair.makePair(firstState, new ArrayList<>(0)), -0.0);
int ticks = 0;
while (!fringe.isEmpty()) {
if (++ticks > maxTicks) {
// System.err.println("WARNING! Timed out on search with " + ticks + " ticks");
return;
}
// Useful variables
double logProbSoFar = fringe.getPriority();
assert logProbSoFar <= 0.0;
Pair<State, List<Counter<String>>> lastStatePair = fringe.removeFirst();
State lastState = lastStatePair.first;
List<Counter<String>> featuresSoFar = lastStatePair.second;
IndexedWord rootWord = lastState.edge == null ? root : lastState.edge.getDependent();
// Register thunk
if (lastState.isDone) {
if (!candidateFragments.test(
Triple.makeTriple(
logProbSoFar,
featuresSoFar,
() -> {
SemanticGraph copy = new SemanticGraph(tree);
lastState
.thunk
.andThen(
x -> {
// Add the extra edges back in, if they don't break the tree-ness of the
// extraction
for (IndexedWord newTreeRoot : x.getRoots()) {
if (newTreeRoot != null) { // what a strange thing to have happen...
for (SemanticGraphEdge extraEdge :
extraEdgesByGovernor.get(newTreeRoot)) {
assert Util.isTree(x);
//noinspection unchecked
addSubtree(
x,
newTreeRoot,
extraEdge.getRelation().toString(),
tree,
extraEdge.getDependent(),
tree.getIncomingEdgesSorted(newTreeRoot));
assert Util.isTree(x);
}
}
}
})
.accept(copy);
return new SentenceFragment(copy, assumedTruth, false);
}))) {
break;
}
}
// Find relevant auxilliary terms
SemanticGraphEdge subjOrNull = null;
SemanticGraphEdge objOrNull = null;
for (SemanticGraphEdge auxEdge : tree.outgoingEdgeIterable(rootWord)) {
String relString = auxEdge.getRelation().toString();
if (relString.contains("obj")) {
objOrNull = auxEdge;
} else if (relString.contains("subj")) {
subjOrNull = auxEdge;
}
}
// Iterate over children
// For each outgoing edge...
for (SemanticGraphEdge outgoingEdge : tree.outgoingEdgeIterable(rootWord)) {
// Prohibit indirect speech verbs from splitting off clauses
// (e.g., 'said', 'think')
// This fires if the governor is an indirect speech verb, and the outgoing edge is a ccomp
if (outgoingEdge.getRelation().toString().equals("ccomp")
&& ((outgoingEdge.getGovernor().lemma() != null
&& INDIRECT_SPEECH_LEMMAS.contains(outgoingEdge.getGovernor().lemma()))
|| INDIRECT_SPEECH_LEMMAS.contains(outgoingEdge.getGovernor().word()))) {
continue;
}
// Get some variables
String outgoingEdgeRelation = outgoingEdge.getRelation().toString();
List<String> forcedArcOrder = hardCodedSplits.get(outgoingEdgeRelation);
if (forcedArcOrder == null && outgoingEdgeRelation.contains(":")) {
forcedArcOrder =
hardCodedSplits.get(
outgoingEdgeRelation.substring(0, outgoingEdgeRelation.indexOf(":")) + ":*");
}
boolean doneForcedArc = false;
// For each action...
for (Action action :
(forcedArcOrder == null ? actionSpace : orderActions(actionSpace, forcedArcOrder))) {
// Check the prerequisite
if (!action.prerequisitesMet(tree, outgoingEdge)) {
continue;
}
if (forcedArcOrder != null && doneForcedArc) {
break;
}
// 1. Compute the child state
Optional<State> candidate =
action.applyTo(tree, lastState, outgoingEdge, subjOrNull, objOrNull);
if (candidate.isPresent()) {
double logProbability;
ClauseClassifierLabel bestLabel;
Counter<String> features =
featurizer.apply(Triple.makeTriple(lastState, action, candidate.get()));
if (forcedArcOrder != null && !doneForcedArc) {
logProbability = 0.0;
bestLabel = ClauseClassifierLabel.CLAUSE_SPLIT;
doneForcedArc = true;
} else if (features.containsKey("__undocumented_junit_no_classifier")) {
logProbability = Double.NEGATIVE_INFINITY;
bestLabel = ClauseClassifierLabel.CLAUSE_INTERM;
} else {
Counter<ClauseClassifierLabel> scores = classifier.scoresOf(new RVFDatum<>(features));
if (scores.size() > 0) {
Counters.logNormalizeInPlace(scores);
}
String rel = outgoingEdge.getRelation().toString();
if ("nsubj".equals(rel) || "dobj".equals(rel)) {
scores.remove(
ClauseClassifierLabel.NOT_A_CLAUSE); // Always at least yield on nsubj and dobj
}
logProbability = Counters.max(scores, Double.NEGATIVE_INFINITY);
bestLabel = Counters.argmax(scores, (x, y) -> 0, ClauseClassifierLabel.CLAUSE_SPLIT);
}
if (bestLabel != ClauseClassifierLabel.NOT_A_CLAUSE) {
Pair<State, List<Counter<String>>> childState =
Pair.makePair(
candidate.get().withIsDone(bestLabel),
new ArrayList<Counter<String>>(featuresSoFar) {
{
add(features);
}
});
// 2. Register the child state
if (!seenWords.contains(childState.first.edge.getDependent())) {
// System.err.println(" pushing " + action.signature() + " with " +
// argmax.first.edge);
fringe.add(childState, logProbability);
}
}
}
}
}
seenWords.add(rootWord);
}
// System.err.println("Search finished in " + ticks + " ticks and " + classifierEvals + "
// classifier evaluations.");
}
@Override
public Counter<String> apply(Triple<State, Action, State> triple) {
// Variables
State from = triple.first;
Action action = triple.second;
State to = triple.third;
String signature = action.signature();
String edgeRelTaken = to.edge == null ? "root" : to.edge.getRelation().toString();
String edgeRelShort = to.edge == null ? "root" : to.edge.getRelation().getShortName();
if (edgeRelShort.contains("_")) {
edgeRelShort = edgeRelShort.substring(0, edgeRelShort.indexOf("_"));
}
// -- Featurize --
// Variables to aggregate
boolean parentHasSubj = false;
boolean parentHasObj = false;
boolean childHasSubj = false;
boolean childHasObj = false;
Counter<String> feats = new ClassicCounter<>();
// 1. edge taken
feats.incrementCount(signature + "&edge:" + edgeRelTaken);
feats.incrementCount(signature + "&edge_type:" + edgeRelShort);
// 2. last edge taken
if (from.edge == null) {
assert to.edge == null || to.originalTree().getRoots().contains(to.edge.getGovernor());
feats.incrementCount(signature + "&at_root");
feats.incrementCount(
signature + "&at_root&root_pos:" + to.originalTree().getFirstRoot().tag());
} else {
feats.incrementCount(signature + "¬_root");
String lastRelShort = from.edge.getRelation().getShortName();
if (lastRelShort.contains("_")) {
lastRelShort = lastRelShort.substring(0, lastRelShort.indexOf("_"));
}
feats.incrementCount(signature + "&last_edge:" + lastRelShort);
}
if (to.edge != null) {
// 3. other edges at parent
for (SemanticGraphEdge parentNeighbor :
from.originalTree().outgoingEdgeIterable(to.edge.getGovernor())) {
if (parentNeighbor != to.edge) {
String parentNeighborRel = parentNeighbor.getRelation().toString();
if (parentNeighborRel.contains("subj")) {
parentHasSubj = true;
}
if (parentNeighborRel.contains("obj")) {
parentHasObj = true;
}
// (add feature)
feats.incrementCount(signature + "&parent_neighbor:" + parentNeighborRel);
feats.incrementCount(
signature
+ "&edge_type:"
+ edgeRelShort
+ "&parent_neighbor:"
+ parentNeighborRel);
}
}
// 4. Other edges at child
int childNeighborCount = 0;
for (SemanticGraphEdge childNeighbor :
from.originalTree().outgoingEdgeIterable(to.edge.getDependent())) {
String childNeighborRel = childNeighbor.getRelation().toString();
if (childNeighborRel.contains("subj")) {
childHasSubj = true;
}
if (childNeighborRel.contains("obj")) {
childHasObj = true;
}
childNeighborCount += 1;
// (add feature)
feats.incrementCount(signature + "&child_neighbor:" + childNeighborRel);
feats.incrementCount(
signature + "&edge_type:" + edgeRelShort + "&child_neighbor:" + childNeighborRel);
}
// 4.1 Number of other edges at child
feats.incrementCount(
signature
+ "&child_neighbor_count:"
+ (childNeighborCount < 3 ? childNeighborCount : ">2"));
feats.incrementCount(
signature
+ "&edge_type:"
+ edgeRelShort
+ "&child_neighbor_count:"
+ (childNeighborCount < 3 ? childNeighborCount : ">2"));
// 5. Subject/Object stats
feats.incrementCount(signature + "&parent_neighbor_subj:" + parentHasSubj);
feats.incrementCount(signature + "&parent_neighbor_obj:" + parentHasObj);
feats.incrementCount(signature + "&child_neighbor_subj:" + childHasSubj);
feats.incrementCount(signature + "&child_neighbor_obj:" + childHasObj);
// 6. POS tag info
feats.incrementCount(signature + "&parent_pos:" + to.edge.getGovernor().tag());
feats.incrementCount(signature + "&child_pos:" + to.edge.getDependent().tag());
feats.incrementCount(
signature
+ "&pos_signature:"
+ to.edge.getGovernor().tag()
+ "_"
+ to.edge.getDependent().tag());
feats.incrementCount(
signature
+ "&edge_type:"
+ edgeRelShort
+ "&pos_signature:"
+ to.edge.getGovernor().tag()
+ "_"
+ to.edge.getDependent().tag());
}
return feats;
}