{
  "schema": "aureon.public_recorded_result.v1",
  "status": "recorded_completed",
  "scope": "Public summary of a recorded internal run. No new inference is executed by this website.",
  "model": "KAO-TerraMind B",
  "precision": "float32, highest matmul precision",
  "scene": "Sentinel-2 L2A, central Thailand",
  "acquisition_date": "2025-01-15",
  "grid": {
    "crs": "EPSG:32647",
    "width": 512,
    "height": 512,
    "transform": [
      10.0,
      0.0,
      632260.0,
      0.0,
      -10.0,
      1583830.0
    ],
    "pixel_size_metres": 10,
    "nodata_class": 255,
    "valid_pixels": 262143,
    "invalid_pixels": 1
  },
  "class_names": [
    "Maize",
    "Sugarcane",
    "Cassava",
    "Rice",
    "Other Crops",
    "Water",
    "Trees",
    "Flooded Veg.",
    "Built Area",
    "Other"
  ],
  "class_pixel_counts": [
    0,
    0,
    0,
    56692,
    140789,
    30262,
    0,
    3497,
    30900,
    3
  ],
  "band_order": [
    "B01",
    "B02",
    "B03",
    "B04",
    "B05",
    "B06",
    "B07",
    "B08",
    "B8A",
    "B09",
    "B11",
    "B12"
  ],
  "units": "surface_reflectance_x10000",
  "source_raster_sha256": "147635528a0c04aef1c3ca482337237c89bc980911907c6b531b6ecfa3234ece",
  "bundle_receipt_sha256": "298515c80ec1c30d3856d3859f87cab86d17d52738d4203a93466772536c5267",
  "original_delivery_receipt_sha256": "262b08c298293c437083af19a4ed7f94c3d858197143eb6cc8b4dff19f79e0b0",
  "original_provenance_sha256": "61b1404a076f23c667cd8a22ce34961cb9470da5c585f21662a24f49f90a823a",
  "preprocessing": {
    "tile_size": 224,
    "stride": 168,
    "tile_count": 9,
    "aggregation": "Ordered logit blending before class selection"
  },
  "parity": {
    "bitwise_equal": true,
    "max_abs_error": 0.0,
    "scene_class_map_equal": true,
    "scene_mask_equal": true,
    "normalization_and_tile_order_equal": true,
    "reference_scope": "Independent eager batch-one execution and unchanged array recipe; same strictly loaded model allocation"
  },
  "quality_scope": "Numerical consistency only. Prediction is not ground truth; no classification accuracy claim.",
  "artifacts": {
    "class_map.tif": {
      "url": "/examples/thailand/class_map.tif",
      "bytes": 17614,
      "sha256": "ad0eb7231bbeeadb8a218edd84fd1b15d72b15fd0ad89bbe917e893bc306c977"
    },
    "valid_mask.tif": {
      "url": "/examples/thailand/valid_mask.tif",
      "bytes": 2085,
      "sha256": "fd1aeef6a1630a78450ed451ec303cde28820a862f5bf3000df89001ac2a455c"
    }
  },
  "public_summary_note": "Curated public fields only. Original artifacts are unchanged; internal paths and raw job records are omitted."
}
