ML/AI Trends

Per-model forecast accuracy weights

Per-model MAE and accuracy score by region and horizon, from the model_accuracy table. Scored against MEASURED ground truth (SNOTEL / Meteo-France / GHCN / resort-reported) on snow days only (>= 1cm observed new snow); the modelled open-meteo fallback is excluded. Regions without a measured truth source are absent from the response rather than reported as zero.

get/api/v1/model-accuracy

Request

  • Base URL: https://www.snowsure.ai
  • URL: https://www.snowsure.ai/api/v1/model-accuracy
  • Auth: none declared

Response

Model accuracy rows

generatedAtstring date-time
bestByRegionobject

Best-performing forecast model keyed by region

Example response

{
  "methodology": {
    "samplePeriodDays": 365,
    "regionsCovered": [
      "alps",
      "andes",
      "cascades",
      "rockies",
      "sierra",
      "southern-alps"
    ]
  }
}

Changes

No recorded changes to this operation across all 1 revision of this API.