Use this metric to quantify prediction error or classification quality only after matching the target scale, aggregation, weighting, and statistical assumptions to the evaluation question.

true Positive

Required integer input.

true Negative

Required integer input.

false Positive

Required integer input.

false Negative

Required integer input.

Compute binary-classification MCC from TP, TN, FP, FN. An undefined zero-denominator case is rejected rather than silently scored as zero. The tool evaluates the supplied inputs together and returns the named outputs below; it does not infer omitted operating conditions or change the units shown.^[1]

coefficient

The resulting coefficient returned as a number.

• Different metrics penalize errors differently and can be dominated by outliers, scale, imbalance, zeros, or threshold choices. A single aggregate hides subgroup, temporal, and distributional failures.
• Use finite inputs in the displayed units and preserve more precision than the final presentation requires. Independently verify safety-critical, financial, compliance, or production decisions.

Report multiple complementary metrics with uncertainty and subgroup breakdowns, inspect residuals, and compare against meaningful baselines.

• Brier Score Calculator

  Compute mean squared probability error for binary outcomes. Probabilities must lie in [0,1], outcomes must be 0 or 1; lower scores are better.

• Huber Loss Calculator

  Compute the mean Huber loss, a quadratic-to-linear blend robust to outlier residuals.

• Mean Absolute Error Calculator

  Calculate mean absolute prediction error.