PublicDatasetModelMetrics

  • algorithm
    Type: string · Algorithm
    required

    Human-readable name of the selected algorithm.

  • correlation_validation_name
    Type: string · Correlation Validation Name
    required

    Name of the primary performance metric, such as r2_cv_mean or mcc_cv_mean.

  • correlation_validation_value
    Type: number · Correlation Validation Value
    required

    Primary cross-validation or test-set performance value. Interpret it using correlation_validation_name.

  • error_validation_name
    Type: string · Error Validation Name
    required

    Name of the secondary performance metric, such as mae_cv_mean or roc_auc_cv_mean.

  • error_validation_value
    Type: number · Error Validation Value
    required

    Secondary cross-validation or test-set performance value. Interpret it using error_validation_name.

  • featuriser
    Type: string · Featuriser
    required

    Human-readable name of the selected molecular representation, including the foundation model name when one was selected.

  • is_log
    Type: boolean · Is Log
    required

    Whether the model was trained on log-transformed targets.

  • test_r2
    Type: number · Test R2nullable

    R-squared on user-provided external test data, if any.