{
  "dimensions": {
    "context": {
      "strata": {
        "coastal_hard_negative": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.0,
                "precision_case_support": 1,
                "recall": null,
                "recall_case_support": 0,
                "status": "computed"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.6,
                      "mean_confidence": null,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.92,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.8464,
                  "ece": 0.92,
                  "status": "computed"
                },
                "f1": null,
                "false_discovery_rate": 1.0,
                "false_negative": 0,
                "false_positive": 1,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 1,
                    "reference_count": 0
                  }
                },
                "precision": 0.0,
                "precision_ci95_wilson": {
                  "lower": 0.0,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 0.7934506856227626
                },
                "prediction_count": 1,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 1,
                "predictions": 1,
                "ranking_predictions": 1,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "data_contract": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "dense_urban": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 1.0,
                "ap50_95": 0.900990099009901,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 1,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.9,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.15333333333333332,
                  "ece": 0.13333333333333325,
                  "status": "computed"
                },
                "f1": 0.8,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 0,
                "false_positive": 1,
                "map50": 1.0,
                "map50_95": 0.900990099009901,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 1.0,
                    "ap50_95": 0.900990099009901,
                    "prediction_count": 4,
                    "reference_count": 2
                  }
                },
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 3,
                "predictions": 3,
                "ranking_predictions": 4,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.8
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "dense_urban_raster": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        },
        "industrial_change": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "pure_background": {
          "case_support": 1,
          "failure_count": 0,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "rural_buildings": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "suburban_buildings": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "terrain_profile": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "date": {
      "strata": {
        "2023-01-01": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        },
        "2024-05-01/2025-05-01": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "2025-01-01": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "2025-01-15": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "2025-02-01": {
          "case_support": 1,
          "failure_count": 0,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "2025-04-20": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "2025-06-15": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 1.0,
                "ap50_95": 0.900990099009901,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 1,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.9,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.15333333333333332,
                  "ece": 0.13333333333333325,
                  "status": "computed"
                },
                "f1": 0.8,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 0,
                "false_positive": 1,
                "map50": 1.0,
                "map50_95": 0.900990099009901,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 1.0,
                    "ap50_95": 0.900990099009901,
                    "prediction_count": 4,
                    "reference_count": 2
                  }
                },
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 3,
                "predictions": 3,
                "ranking_predictions": 4,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.8
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "2025-07-01": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.0,
                "precision_case_support": 1,
                "recall": null,
                "recall_case_support": 0,
                "status": "computed"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.6,
                      "mean_confidence": null,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.92,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.8464,
                  "ece": 0.92,
                  "status": "computed"
                },
                "f1": null,
                "false_discovery_rate": 1.0,
                "false_negative": 0,
                "false_positive": 1,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 1,
                    "reference_count": 0
                  }
                },
                "precision": 0.0,
                "precision_ci95_wilson": {
                  "lower": 0.0,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 0.7934506856227626
                },
                "prediction_count": 1,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 1,
                "predictions": 1,
                "ranking_predictions": 1,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "2025-10-05": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "difficulty": {
      "strata": {
        "hard": {
          "case_support": 6,
          "failure_count": 10,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "object_detection": {
              "case_support": 3,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "geospatial_data_validation",
            "object_detection",
            "raster_classification"
          ]
        },
        "normal": {
          "case_support": 3,
          "failure_count": 4,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation",
            "terrain_interpretation",
            "vector_comparison"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "municipality": {
      "strata": {
        "Antwerpen": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 1.0,
                "ap50_95": 0.900990099009901,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 1,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.9,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.15333333333333332,
                  "ece": 0.13333333333333325,
                  "status": "computed"
                },
                "f1": 0.8,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 0,
                "false_positive": 1,
                "map50": 1.0,
                "map50_95": 0.900990099009901,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 1.0,
                    "ap50_95": 0.900990099009901,
                    "prediction_count": 4,
                    "reference_count": 2
                  }
                },
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 3,
                "predictions": 3,
                "ranking_predictions": 4,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.8
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "Brussel": {
          "case_support": 2,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection",
            "raster_classification"
          ]
        },
        "Hasselt": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        },
        "Liège": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "Mol": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "Namur": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "Oostende": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.0,
                "precision_case_support": 1,
                "recall": null,
                "recall_case_support": 0,
                "status": "computed"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.6,
                      "mean_confidence": null,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.92,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.8464,
                  "ece": 0.92,
                  "status": "computed"
                },
                "f1": null,
                "false_discovery_rate": 1.0,
                "false_negative": 0,
                "false_positive": 1,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 1,
                    "reference_count": 0
                  }
                },
                "precision": 0.0,
                "precision_ci95_wilson": {
                  "lower": 0.0,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 0.7934506856227626
                },
                "prediction_count": 1,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 1,
                "predictions": 1,
                "ranking_predictions": 1,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "not_applicable": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "object_size": {
      "strata": {
        "empty": {
          "case_support": 2,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 2,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.0,
                "precision_case_support": 1,
                "recall": null,
                "recall_case_support": 0,
                "status": "computed"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.6,
                      "mean_confidence": null,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.92,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.8464,
                  "ece": 0.92,
                  "status": "computed"
                },
                "f1": null,
                "false_discovery_rate": 1.0,
                "false_negative": 0,
                "false_positive": 1,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 1,
                    "reference_count": 0
                  }
                },
                "precision": 0.0,
                "precision_ci95_wilson": {
                  "lower": 0.0,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 0.7934506856227626
                },
                "prediction_count": 1,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 1,
                "predictions": 1,
                "ranking_predictions": 1,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "large": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "mixed": {
          "case_support": 3,
          "failure_count": 6,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 1.0,
                "ap50_95": 0.900990099009901,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 1,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.9,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.15333333333333332,
                  "ece": 0.13333333333333325,
                  "status": "computed"
                },
                "f1": 0.8,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 0,
                "false_positive": 1,
                "map50": 1.0,
                "map50_95": 0.900990099009901,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 1.0,
                    "ap50_95": 0.900990099009901,
                    "prediction_count": 4,
                    "reference_count": 2
                  }
                },
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 3,
                "predictions": 3,
                "ranking_predictions": 4,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.8
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "object_detection",
            "vector_comparison"
          ]
        },
        "not_applicable": {
          "case_support": 3,
          "failure_count": 5,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation",
            "raster_classification",
            "terrain_interpretation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "occlusion": {
      "strata": {
        "none": {
          "case_support": 4,
          "failure_count": 5,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "object_detection": {
              "case_support": 2,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.0,
                "precision_case_support": 1,
                "recall": null,
                "recall_case_support": 0,
                "status": "computed"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.6,
                      "mean_confidence": null,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.92,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.8464,
                  "ece": 0.92,
                  "status": "computed"
                },
                "f1": null,
                "false_discovery_rate": 1.0,
                "false_negative": 0,
                "false_positive": 1,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 1,
                    "reference_count": 0
                  }
                },
                "precision": 0.0,
                "precision_ci95_wilson": {
                  "lower": 0.0,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 0.7934506856227626
                },
                "prediction_count": 1,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 1,
                "predictions": 1,
                "ranking_predictions": 1,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "footprint_segmentation",
            "object_detection"
          ]
        },
        "not_applicable": {
          "case_support": 4,
          "failure_count": 7,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation",
            "raster_classification",
            "terrain_interpretation",
            "vector_comparison"
          ]
        },
        "partial": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 1.0,
                "ap50_95": 0.900990099009901,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 1,
                      "count": 1,
                      "lower": 0.8,
                      "mean_confidence": 0.9,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.15333333333333332,
                  "ece": 0.13333333333333325,
                  "status": "computed"
                },
                "f1": 0.8,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 0,
                "false_positive": 1,
                "map50": 1.0,
                "map50_95": 0.900990099009901,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 1.0,
                    "ap50_95": 0.900990099009901,
                    "prediction_count": 4,
                    "reference_count": 2
                  }
                },
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 3,
                "predictions": 3,
                "ranking_predictions": 4,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.8
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "region": {
      "strata": {
        "brussels": {
          "case_support": 2,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection",
            "raster_classification"
          ]
        },
        "flanders": {
          "case_support": 4,
          "failure_count": 7,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 2,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection",
            "terrain_interpretation",
            "vector_comparison"
          ]
        },
        "national": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "wallonia": {
          "case_support": 2,
          "failure_count": 3,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "footprint_segmentation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "resolution_m": {
      "strata": {
        "0.1": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "0.25": {
          "case_support": 3,
          "failure_count": 4,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 3,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "0.5": {
          "case_support": 2,
          "failure_count": 3,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "footprint_segmentation"
          ]
        },
        "1": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        },
        "5": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        },
        "__not_applicable__": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "season": {
      "strata": {
        "autumn": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        },
        "multi-date": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "not_applicable": {
          "case_support": 3,
          "failure_count": 5,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation",
            "terrain_interpretation",
            "vector_comparison"
          ]
        },
        "spring": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "summer": {
          "case_support": 2,
          "failure_count": 4,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 2,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "winter": {
          "case_support": 1,
          "failure_count": 0,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "sensor": {
      "strata": {
        "elevation-raster": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        },
        "mixed": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "synthetic-multispectral": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        },
        "synthetic-rgb": {
          "case_support": 4,
          "failure_count": 5,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "object_detection": {
              "case_support": 3,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation",
            "object_detection"
          ]
        },
        "vector": {
          "case_support": 2,
          "failure_count": 4,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "vector_comparison"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "source": {
      "strata": {
        "fixture-contracts": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "fixture-dhmv": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "terrain_interpretation"
          ]
        },
        "fixture-grb": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "fixture-orthophoto": {
          "case_support": 3,
          "failure_count": 4,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 3,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection"
          ]
        },
        "fixture-picc": {
          "case_support": 1,
          "failure_count": 1,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation"
          ]
        },
        "fixture-picc-temporal": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "fixture-thematic-raster": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "raster_classification"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "urbanity": {
      "strata": {
        "industrial": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection"
          ]
        },
        "mixed": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "rural": {
          "case_support": 2,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "footprint_segmentation",
            "terrain_interpretation"
          ]
        },
        "suburban": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "vector_comparison"
          ]
        },
        "urban": {
          "case_support": 4,
          "failure_count": 6,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 3,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection",
            "raster_classification"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    },
    "vegetation": {
      "strata": {
        "low": {
          "case_support": 4,
          "failure_count": 5,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "change_detection": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "footprint_segmentation": {
              "case_support": 1,
              "macro": {
                "f1": 1.0,
                "f1_case_support": 1,
                "mean_boundary_f1": 1.0,
                "mean_boundary_f1_case_support": 1,
                "mean_dice": 0.8881578947368421,
                "mean_dice_case_support": 1,
                "mean_iou": 0.7988165680473372,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 1.0,
                "precision_case_support": 1,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 1.0,
                "false_discovery_rate": 0.0,
                "false_negative": 0,
                "false_positive": 0,
                "miss_rate": 0.0,
                "precision": 1.0,
                "precision_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "prediction_count": 1,
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.20654931437723745,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 1,
                  "upper": 1.0
                },
                "reference_count": 1,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 1,
                "references": 1
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 1.0
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "object_detection": {
              "case_support": 1,
              "macro": {
                "f1": null,
                "f1_case_support": 0,
                "method": "unweighted mean across cases with defined values",
                "precision": null,
                "precision_case_support": 0,
                "recall": null,
                "recall_case_support": 0,
                "status": "not_evaluable"
              },
              "micro": {
                "ap50": null,
                "ap50_95": null,
                "calibration": {
                  "bins": [],
                  "brier": null,
                  "ece": null,
                  "status": "undefined"
                },
                "f1": null,
                "false_discovery_rate": null,
                "false_negative": 0,
                "false_positive": 0,
                "map50": null,
                "map50_95": null,
                "miss_rate": null,
                "per_class_ap": {
                  "building": {
                    "ap50": null,
                    "ap50_95": null,
                    "prediction_count": 0,
                    "reference_count": 0
                  }
                },
                "precision": null,
                "precision_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "prediction_count": 0,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": null,
                "recall_ci95_wilson": {
                  "lower": null,
                  "status": "undefined",
                  "support": 0,
                  "upper": null
                },
                "reference_count": 0,
                "true_positive": 0
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 0,
                "predictions": 0,
                "ranking_predictions": 0,
                "references": 0
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": null
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "vector_comparison": {
              "case_support": 1,
              "macro": {
                "f1": 0.5,
                "f1_case_support": 1,
                "mean_iou": 1.0,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.5,
                "precision_case_support": 1,
                "recall": 0.5,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "f1": 0.5,
                "false_discovery_rate": 0.5,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.5,
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "prediction_count": 2,
                "recall": 0.5,
                "recall_ci95_wilson": {
                  "lower": 0.09453120573423074,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 0.9054687942657693
                },
                "reference_count": 2,
                "true_positive": 1
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 2,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.5
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "change_detection",
            "footprint_segmentation",
            "object_detection",
            "vector_comparison"
          ]
        },
        "mixed": {
          "case_support": 1,
          "failure_count": 2,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "geospatial_data_validation": {
              "case_support": 1,
              "macro": {
                "blocker_or_critical_miss_count": 1.0,
                "blocker_or_critical_miss_count_case_support": 1,
                "f1": 0.6666666666666666,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.6666666666666666,
                "precision_case_support": 1,
                "recall": 0.6666666666666666,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "blocker_or_critical_miss_count": 1,
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.3333333333333333,
                "false_negative": 1,
                "false_positive": 1,
                "miss_rate": 0.3333333333333333,
                "precision": 0.6666666666666666,
                "precision_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "prediction_count": 3,
                "recall": 0.6666666666666666,
                "recall_ci95_wilson": {
                  "lower": 0.2076596008020477,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 3,
                  "upper": 0.9385080552796037
                },
                "reference_count": 3,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "predictions": 3,
                "references": 3
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "geospatial_data_validation"
          ]
        },
        "moderate": {
          "case_support": 4,
          "failure_count": 7,
          "release_gate_reason": "insufficient_support",
          "release_gate_status": "not_evaluable",
          "task_metrics": {
            "object_detection": {
              "case_support": 2,
              "macro": {
                "f1": 0.8,
                "f1_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "precision": 0.3333333333333333,
                "precision_case_support": 2,
                "recall": 1.0,
                "recall_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "ap50": 0.6666666666666666,
                "ap50_95": 0.5838283828382838,
                "calibration": {
                  "binning": "five fixed equal-width bins",
                  "bins": [
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.0,
                      "mean_confidence": null,
                      "upper": 0.2
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.2,
                      "mean_confidence": null,
                      "upper": 0.4
                    },
                    {
                      "accuracy": null,
                      "count": 0,
                      "lower": 0.4,
                      "mean_confidence": null,
                      "upper": 0.6
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.6,
                      "mean_confidence": 0.6499999999999999,
                      "upper": 0.8
                    },
                    {
                      "accuracy": 0.5,
                      "count": 2,
                      "lower": 0.8,
                      "mean_confidence": 0.91,
                      "upper": 1.0
                    }
                  ],
                  "brier": 0.3266,
                  "ece": 0.27999999999999997,
                  "status": "computed"
                },
                "f1": 0.6666666666666666,
                "false_discovery_rate": 0.5,
                "false_negative": 0,
                "false_positive": 2,
                "map50": 0.6666666666666666,
                "map50_95": 0.5838283828382838,
                "miss_rate": 0.0,
                "per_class_ap": {
                  "building": {
                    "ap50": 0.6666666666666666,
                    "ap50_95": 0.5838283828382838,
                    "prediction_count": 5,
                    "reference_count": 2
                  }
                },
                "precision": 0.5,
                "precision_ci95_wilson": {
                  "lower": 0.15003898915214947,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 4,
                  "upper": 0.8499610108478506
                },
                "prediction_count": 4,
                "ranking_scope": "predictions pooled across cases; class- and sample-aware matching; no averaging of per-case AP",
                "recall": 1.0,
                "recall_ci95_wilson": {
                  "lower": 0.34238022750665303,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 2,
                  "upper": 1.0
                },
                "reference_count": 2,
                "true_positive": 2
              },
              "minimum_case_support": 5,
              "observation_support": {
                "calibration_predictions": 4,
                "predictions": 4,
                "ranking_predictions": 5,
                "references": 2
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.f1",
                "value": 0.6666666666666666
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "raster_classification": {
              "case_support": 1,
              "macro": {
                "accuracy": 0.8333333333333334,
                "accuracy_case_support": 1,
                "macro_f1": 0.8222222222222223,
                "macro_f1_case_support": 1,
                "mean_iou": 0.7222222222222222,
                "mean_iou_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "status": "computed"
              },
              "micro": {
                "accuracy": 0.8333333333333334,
                "accuracy_ci95_wilson": {
                  "lower": 0.5519691377470266,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 0.9530348578161463
                },
                "correct_pixel_count": 10,
                "evaluated_reference_pixel_count": 12,
                "macro_f1_across_classes": 0.8222222222222223,
                "mean_iou": 0.7222222222222222,
                "per_class": {
                  "0": {
                    "f1": 1.0,
                    "false_discovery_rate": 0.0,
                    "false_negative": 0,
                    "false_positive": 0,
                    "iou": 1.0,
                    "miss_rate": 0.0,
                    "precision": 1.0,
                    "precision_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "prediction_count": 4,
                    "recall": 1.0,
                    "recall_ci95_wilson": {
                      "lower": 0.5101091635454027,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 4,
                      "upper": 1.0
                    },
                    "reference_count": 4,
                    "true_positive": 4
                  },
                  "1": {
                    "f1": 0.8000000000000002,
                    "false_discovery_rate": 0.2,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.6666666666666666,
                    "miss_rate": 0.2,
                    "precision": 0.8,
                    "precision_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "prediction_count": 5,
                    "recall": 0.8,
                    "recall_ci95_wilson": {
                      "lower": 0.37553462976252533,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 5,
                      "upper": 0.9637758913675698
                    },
                    "reference_count": 5,
                    "true_positive": 4
                  },
                  "2": {
                    "f1": 0.6666666666666666,
                    "false_discovery_rate": 0.3333333333333333,
                    "false_negative": 1,
                    "false_positive": 1,
                    "iou": 0.5,
                    "miss_rate": 0.3333333333333333,
                    "precision": 0.6666666666666666,
                    "precision_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "prediction_count": 3,
                    "recall": 0.6666666666666666,
                    "recall_ci95_wilson": {
                      "lower": 0.2076596008020477,
                      "method": "Wilson score interval",
                      "status": "computed",
                      "support": 3,
                      "upper": 0.9385080552796037
                    },
                    "reference_count": 3,
                    "true_positive": 2
                  }
                },
                "prediction_coverage": 1.0,
                "prediction_coverage_ci95_wilson": {
                  "lower": 0.7575059933447592,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 12,
                  "upper": 1.0
                },
                "prediction_present_pixel_count": 12
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated_pixels": 12
              },
              "primary_metric": {
                "direction": "higher_is_better",
                "name": "micro.mean_iou",
                "value": 0.7222222222222222
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            },
            "terrain_interpretation": {
              "case_support": 1,
              "macro": {
                "bias": 0.02499999999999991,
                "bias_case_support": 1,
                "coverage": 0.8,
                "coverage_case_support": 1,
                "mae": 0.2749999999999999,
                "mae_case_support": 1,
                "method": "unweighted mean across cases with defined values",
                "rmse": 0.2872281323269015,
                "rmse_case_support": 1,
                "status": "computed"
              },
              "micro": {
                "bias": 0.02499999999999991,
                "coverage": 0.8,
                "coverage_ci95_wilson": {
                  "lower": 0.37553462976252533,
                  "method": "Wilson score interval",
                  "status": "computed",
                  "support": 5,
                  "upper": 0.9637758913675698
                },
                "evaluated_count": 4,
                "mae": 0.2749999999999999,
                "reference_count": 5,
                "rmse": 0.2872281323269015,
                "status": "computed",
                "units": "m_TAW"
              },
              "minimum_case_support": 5,
              "observation_support": {
                "evaluated": 4,
                "references": 5
              },
              "primary_metric": {
                "direction": "lower_is_better",
                "name": "micro.rmse",
                "value": 0.2872281323269015
              },
              "release_gate_reason": "insufficient_support",
              "release_gate_status": "not_evaluable",
              "status": "insufficient_support"
            }
          },
          "tasks": [
            "object_detection",
            "raster_classification",
            "terrain_interpretation"
          ]
        }
      },
      "worst_stratum_by_task": {
        "change_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "footprint_segmentation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "geospatial_data_validation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "object_detection": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "raster_classification": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "terrain_interpretation": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        },
        "vector_comparison": {
          "reason": "no stratum meets minimum support with a defined primary metric",
          "status": "not_evaluable"
        }
      }
    }
  },
  "minimum_case_support": 5,
  "overall_status": "not_evaluable",
  "schema_version": 2,
  "support_unit": "independent benchmark cases; pixel/object counts are reported separately and do not replace case support"
}