stanfordnlp/CoreNLP · error · RuntimeException

Cannot enable POSSequence features without POS sequence map.

Error message

Cannot enable POSSequence features without POS sequence map. Use option -frenchMWMap.

What it means

This RuntimeException from FrenchTreebankParserParams.MWTAnnotatingFunction.apply means POS-sequence feature annotation was requested, but the multi-word-expression counter (mwCounter) was never loaded. The counter is only populated when the -frenchMWMap option supplies a POS sequence map, so annotation of French multi-word terminals cannot proceed without it. The library throws eagerly rather than silently producing unannotated features.

Solutions

1. Add the -frenchMWMap option pointing at a valid POS sequence map file when invoking the parser/trainer.
2. Verify the annotation feature that requires POSSequence is actually intended; disable it if the map is unavailable.
3. Load the mwCounter programmatically (setFrenchMWMap path) before running the annotating function.

Example fix

// before
java -mx4g edu.stanford.nlp.parser.lexparser.LexicalizedParser -french frenchFactored.ser.gz input.txt
// after
java -mx4g edu.stanford.nlp.parser.lexparser.LexicalizedParser -french -frenchMWMap mwMap.txt frenchFactored.ser.gz input.txt

// check options before training/annotation
if (op.trainOptions.useMWTFeatures && mwCounter == null) {
    throw new IllegalArgumentException("Provide -frenchMWMap option");
}

When it happens

Trigger: Registering/using the MWTreebankAnnotator / POSSequence annotation function via a Tregex-matched transformation while trainOptions were built without -frenchMWMap (mwCounter stays null).

Common situations: Training or parsing French trees with the -frenchMWMap-dependent annotation pipeline enabled but the map file option omitted from the command line or Options setup.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

• Not POS sequence for tree:
• Error initializing coref system
• Not a valid ellipses style:
• Invalid value for useUnknownWordSignatures:
• %s: Term lacks morpho analysis: %s

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/b9daca7b5d2d835d. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/parser/lexparser/FrenchTreebankParserParams.java:482

  }

  private class AddPOSSequenceFunction implements SerializableFunction<TregexMatcher,String> {

    private final String annotationMark;
    private final boolean doBasicCat;
    private final double cutoff;

    public AddPOSSequenceFunction(String annotationMark, int cutoff, boolean basicCategory) {
      this.annotationMark = annotationMark;
      doBasicCat = basicCategory;
      this.cutoff = cutoff;
    }


    public String apply(TregexMatcher m) {
      if(mwCounter == null)
        throw new RuntimeException("Cannot enable POSSequence features without POS sequence map. Use option -frenchMWMap.");

      Tree t = m.getMatch();
      StringBuilder sb = new StringBuilder();
      for(Tree kid : t.children()) {
        if( ! kid.isPreTerminal())
          throw new RuntimeException("Not POS sequence for tree: " + t.toString());
        String tag = doBasicCat ? tlp.basicCategory(kid.value()) : kid.value();
        sb.append(tag).append(" ");
      }

      if(mwCounter.getCount(t.value(), sb.toString().trim()) > cutoff)
        return annotationMark + sb.toString().replaceAll("\\s+", "").toLowerCase();
      else
        return "";
    }

    @Override
    public String toString() {

View on GitHub (pinned to 1b7edd19c4)