stanfordnlp/CoreNLP · error · UnsupportedOperationException

Predict not implemented for max margin

Error message

Predict not implemented for max margin

What it means

The maxMargin loss function intentionally has no predict() implementation; calling predict on a Loss returned by SimpleLinearClassifier.maxMargin(h) throws UnsupportedOperationException. Max-margin losses only define the hinge derivative for training, not a scoring function.

Solutions

1. Do not call predict with a maxMargin loss; use a differentiable/scoreable loss such as logistic for inference
2. Rebuild the classifier with logistic loss before scoring
3. Use the raw weights dot-product yourself instead of Loss.predict if you need margin scores

Example fix

// before
SimpleLinearClassifier clf = new SimpleLinearClassifier(SimpleLinearClassifier.maxMargin(1.0), modelFile);
double score = clf.predict(example, features, compressor); // throws
// after
SimpleLinearClassifier clf = new SimpleLinearClassifier(Loss.logistic, modelFile);
double score = clf.predict(example, features, compressor);

if (loss instanceof SimpleLinearClassifier.Loss) {
  // ensure loss is not maxMargin before calling predict
  assert !isMaxMargin(loss) : "maxMargin loss does not support predict";
}

try {
  double score = classifier.predict(example, feats, compressor);
} catch (UnsupportedOperationException e) {
  throw new IllegalStateException("use a scoreable loss (e.g. logistic) for predict", e);
}

When it happens

Trigger: Calling classifier.predict(example, ...) (SimpleLinearClassifier.predict at line 184) when the classifier was constructed with the Loss returned by maxMargin(h).

Common situations: Training a coref classifier with max-margin loss, then reusing it for inference/scoring instead of switching to a loss that supports prediction (e.g. logistic).

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

• LogPrior.getSigmaSquaredM is undefined for any prior but…
• If you want to ask for the probability, you must train a…
• Not implemented for this class.
• Error running testGibbs inference!
• Index out of bounds:

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/4b73387f2c1bb8f9. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/coref/statistical/SimpleLinearClassifier.java:184

            (mistake - 1) * label / gamma : -label);
      }

      @Override
      public String toString() {
        return String.format("quadraticallySmoothed(%s)", gamma);
      }
    };
  }

  public static Loss hinge() {
    return quadraticallySmoothedSVM(0);
  }

  public static Loss maxMargin(final double h) {
    return new Loss() {
      @Override
      public double predict(double product) {
        throw new UnsupportedOperationException("Predict not implemented for max margin");
      }

      @Override
      public double derivative(double label, double product) {
        return product < -h ? 0 : 1;
      }

      @Override
      public String toString() {
        return String.format("max-margin(%s)", h);
      }
    };
  }

  public static Loss risk() {
    return new Loss() {
      @Override
      public double predict(double product) {

View on GitHub (pinned to 1b7edd19c4)