stanfordnlp/CoreNLP · error · ClassCastException

cannot be cast into a…

Error message

{object.getClass()} cannot be cast into a edu.stanford.nlp.ie.KBPStatisticalExtractor

What it means

KBPEnsembleExtractor's main, when loading the statistical model file, requires the deserialized object to be either a LinearClassifier<String,String> or an existing KBPStatisticalExtractor. Any other object type is rejected with this ClassCastException.

Solutions

1. Point the statistical model option at the correct KBPStatisticalExtractor/LinearClassifier serialized file
2. Re-run the KBP statistical trainer to produce a proper model file
3. Inspect the object's actual class in the wrapped file to identify what was serialized
4. Verify you are using matching Stanford CoreNLP versions for training and loading

Example fix

// before
args.put("stat.serialize.model", "models/relation-extractor.ser.gz"); // wrong file
// after
args.put("stat.serialize.model", "models/kbp-statistical-model.ser.gz"); // LinearClassifier file

try (ObjectInputStream in = new ObjectInputStream(new FileInputStream(statModelPath))) {
  Object o = in.readObject();
  if (!(o instanceof LinearClassifier) && !(o instanceof KBPStatisticalExtractor))
    throw new IllegalStateException("stat model file holds " + o.getClass());
}

boolean isStatModel(Object o) {
  return o instanceof LinearClassifier || o instanceof KBPStatisticalExtractor;
}

try {
  runKBPMain(args);
} catch (ClassCastException e) {
  log.error("wrong statistical model file type: " + e.getMessage());
  throw e;
}

When it happens

Trigger: Passing -stat.serialize.model (statistical model) pointing to a file containing an object of an unexpected type (e.g. a different model class, a Map, or text file bytes deserialized as something else).

Common situations: Pointing the stat model option at the wrong serialized file (e.g. the relation extractor model instead of the statistical model); models produced by incompatible KBP training code versions.

Understand the failure class

Background: Type mismatch errors: IllegalArgumentException, TypeError and type guards across 150 open-source libraries — this error's family across 150 libraries.

Related errors

• cannot be cast into a
• First line of input file should be header definition
• Could not parse CoNLL file
• Gabor sucks at logic and he should feel bad about it
• ERROR: Serialized data does not contain an Annotation!

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/5cc933cd9d392ea2. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/ie/KBPEnsembleExtractor.java:89

  }

  public static void main(String[] args) throws IOException, ClassNotFoundException {
    RedwoodConfiguration.standard().apply();  // Disable SLF4J crap.
    ArgumentParser.fillOptions(KBPEnsembleExtractor.class, args);

    KBPRelationExtractor statisticalExtractor;
    if (STATISTICAL_MODEL.length() == 0) {
        logger.info("No statistical model will be used.");
        statisticalExtractor = null;
    } else {
        Object object = IOUtils.readObjectFromURLOrClasspathOrFileSystem(STATISTICAL_MODEL);
        if (object instanceof LinearClassifier) {
          //noinspection unchecked
          statisticalExtractor = new KBPStatisticalExtractor((Classifier<String, String>) object);
        } else if (object instanceof KBPStatisticalExtractor) {
          statisticalExtractor = (KBPStatisticalExtractor) object;
        } else {
          throw new ClassCastException(object.getClass() + " cannot be cast into a " + KBPStatisticalExtractor.class);
        }
        logger.info("Read statistical model from " + STATISTICAL_MODEL);
    }

    KBPRelationExtractor extractor;
    if (statisticalExtractor == null) {
        extractor = new KBPEnsembleExtractor(
            new KBPTokensregexExtractor(TOKENSREGEX_DIR),
            new KBPSemgrexExtractor(SEMGREX_DIR));
    } else {
        extractor = new KBPEnsembleExtractor(
            new KBPTokensregexExtractor(TOKENSREGEX_DIR),
            new KBPSemgrexExtractor(SEMGREX_DIR),
            statisticalExtractor);
    }

    List<Pair<KBPInput, String>> testExamples = KBPRelationExtractor.readDataset(TEST_FILE);

View on GitHub (pinned to 1b7edd19c4)