stanfordnlp/CoreNLP · error · java.lang.IllegalArgumentException

Unexpected matrix initialization type

Error message

Unexpected matrix initialization type ${op.trainOptions.transformMatrixType}

What it means

DVModel.randomTransformMatrix uses a switch over op.trainOptions.transformMatrixType to build random transform matrices; the default branch throws IllegalArgumentException for any unrecognized type. This indicates the TrainingOptions field was set to a value outside the supported enum/constants (e.g. via options file or programmatic misuse).

Solutions

  1. Set transformMatrixType to a supported value (e.g. RANDOM ||Diagonal||) in trainOptions
  2. Remove the custom override and use the default transform matrix initialization
  3. Check the CoreNLP version's TrainOptions for the valid constants and correct the config

Example fix

// before
op.trainOptions.transformMatrixType = "DIAGONAL"; // unsupported
// after
op.trainOptions.transformMatrixType = TrainOptions.TransformMatrixType.RANDOM;
Defensive patterns

Strategy: validation

Validate before calling

String t = op.trainOptions.transformMatrixType;
Set<String> supported = Set.of("RANDOM", "ZERO", "DIAGONAL"); // check TrainOptions for your version
if (!supported.contains(t)) {
    throw new IllegalArgumentException("Unsupported transformMatrixType: " + t);
}

Try / catch

try {
    DVModel model = new DVModel(op, stateIndex, wordlist, dvWordVectorsFile);
} catch (IllegalArgumentException e) {
    if (e.getMessage().contains("matrix initialization type")) {
        op.trainOptions.transformMatrixType = "RANDOM";
    }
}

Prevention

When it happens

Trigger: Setting trainOptions.transformMatrixType to an unsupported value before DVModel construction, which then calls randomTransformMatrix from the unary/left/right matrix initializers during model build.

Common situations: Hand-editing parser options files with an invalid matrix type name; copy-pasting option values between CoreNLP versions where the constant set differs; typos in -transformMatrixType style flags.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/845b099088976e87. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/parser/dvparser/DVModel.java:238

    case OFF_DIAGONAL:
      matrix = SimpleMatrix.random_DDRM(numRows,numCols,-1.0/Math.sqrt((double)numCols * 100.0),1.0/Math.sqrt((double)numCols * 100.0),rand).plus(identity);
      for (int i = 0; i < numCols; ++i) {
        int x = rand.nextInt(numCols);
        int y = rand.nextInt(numCols);
        int scale = rand.nextInt(3) - 1;  // -1, 0, or 1
        matrix.set(x, y, matrix.get(x, y) + scale);
      }
      break;
    case RANDOM_ZEROS:
      matrix = SimpleMatrix.random_DDRM(numRows,numCols,-1.0/Math.sqrt((double)numCols * 100.0),1.0/Math.sqrt((double)numCols * 100.0),rand).plus(identity);
      for (int i = 0; i < numCols; ++i) {
        int x = rand.nextInt(numCols);
        int y = rand.nextInt(numCols);
        matrix.set(x, y, 0.0);
      }
      break;
    default:
      throw new IllegalArgumentException("Unexpected matrix initialization type " + op.trainOptions.transformMatrixType);
    }
    return matrix;
  }

  public void addRandomUnaryMatrix(String childBasic) {
    if (unaryTransform.get(childBasic) != null) {
      return;
    }

    ++numUnaryMatrices;

    // scoring matrix
    SimpleMatrix score = SimpleMatrix.random_DDRM(1, numCols, -1.0/Math.sqrt((double)numCols),1.0/Math.sqrt((double)numCols),rand);
    unaryScore.put(childBasic, score.scale(op.trainOptions.scalingForInit));

    SimpleMatrix transform;
    if (op.trainOptions.useContextWords) {
      transform = new SimpleMatrix(numRows, numCols * 3 + 1);

View on GitHub (pinned to 1b7edd19c4)