{
  "name": "Scikit-Learn Binary Classification Evaluation Tools Comparison",
  "description": "Comparative reference of binary classification evaluation functions and display classes in scikit-learn, detailing output structures, prediction score requirements, threshold dependency, and clinical diagnostic focus compiled from scikit-learn documentation.",
  "page": "https://casrai.org/guides/roc-curve-in-python",
  "publisher": "CASRAI",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "dateModified": "2026-11-20T01:27:33Z",
  "temporalCoverage": "2026",
  "spatialCoverage": "Global",
  "columns": [
    {
      "name": "Evaluation Tool",
      "unit": "",
      "description": "Scikit-learn metric or display function"
    },
    {
      "name": "Output Type",
      "unit": "",
      "description": "Format of returned evaluation data"
    },
    {
      "name": "Input Requirement",
      "unit": "",
      "description": "Expected prediction format: continuous scores or discrete labels"
    },
    {
      "name": "Threshold Dependent",
      "unit": "",
      "description": "Whether evaluation evaluates a single threshold or full spectrum"
    },
    {
      "name": "Display Utility",
      "unit": "",
      "description": "Associated visualization class in sklearn.metrics"
    },
    {
      "name": "Diagnostic Focus",
      "unit": "",
      "description": "Clinical or statistical performance aspect assessed"
    }
  ],
  "rows": [
    {
      "Evaluation Tool": "roc_curve",
      "Output Type": "Tuple of arrays (FPR, TPR, thresholds)",
      "Input Requirement": "Continuous probability or decision scores",
      "Threshold Dependent": "Full spectrum (all unique thresholds)",
      "Display Utility": "RocCurveDisplay",
      "Diagnostic Focus": "Discrimination across all decision thresholds"
    },
    {
      "Evaluation Tool": "roc_auc_score",
      "Output Type": "Scalar float (AUROC)",
      "Input Requirement": "Continuous probability or decision scores",
      "Threshold Dependent": "Threshold-independent summary",
      "Display Utility": "RocCurveDisplay",
      "Diagnostic Focus": "Rank-order discrimination capability"
    },
    {
      "Evaluation Tool": "precision_recall_curve",
      "Output Type": "Tuple of arrays (Precision, Recall, thresholds)",
      "Input Requirement": "Continuous probability or decision scores",
      "Threshold Dependent": "Full spectrum (all unique thresholds)",
      "Display Utility": "PrecisionRecallDisplay",
      "Diagnostic Focus": "Performance under severe class imbalance"
    },
    {
      "Evaluation Tool": "confusion_matrix",
      "Output Type": "2x2 integer array (TN, FP, FN, TP)",
      "Input Requirement": "Discrete predicted class labels",
      "Threshold Dependent": "Single operational threshold",
      "Display Utility": "ConfusionMatrixDisplay",
      "Diagnostic Focus": "Operational error counts at fixed threshold"
    },
    {
      "Evaluation Tool": "classification_report",
      "Output Type": "Text summary dictionary (precision, recall, f1)",
      "Input Requirement": "Discrete predicted class labels",
      "Threshold Dependent": "Single operational threshold",
      "Display Utility": "None",
      "Diagnostic Focus": "Multimetric summary at fixed threshold"
    },
    {
      "Evaluation Tool": "det_curve",
      "Output Type": "Tuple of arrays (FPR, FNR, thresholds)",
      "Input Requirement": "Continuous probability or decision scores",
      "Threshold Dependent": "Full spectrum (normal deviate scale)",
      "Display Utility": "DetCurveDisplay",
      "Diagnostic Focus": "Detection error trade-off in verification systems"
    }
  ],
  "sources": [
    {
      "title": "scikit-learn: sklearn.metrics.roc_curve documentation",
      "url": "https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_curve.html",
      "accessed": "2026-11-19"
    },
    {
      "title": "scikit-learn: sklearn.metrics.RocCurveDisplay documentation",
      "url": "https://scikit-learn.org/stable/modules/generated/sklearn.metrics.RocCurveDisplay.html",
      "accessed": "2026-11-19"
    },
    {
      "title": "scikit-learn: 3.4. Metrics and scoring: quantifying the quality of predictions",
      "url": "https://scikit-learn.org/stable/modules/model_evaluation.html",
      "accessed": "2026-11-19"
    }
  ]
}