Health Check Endpoint
Checks the readiness status of the Student Purchase Prediction API and its components.
Endpoint
Response
Returns the health status of the API service and its loaded artifacts.
Overall readiness status. Returns true only when model, threshold, and feature config are all loaded.
Whether the trained model artifact has been successfully loaded into memory.
Whether the drift monitoring baseline file has been loaded. Used for distribution drift detection.
Status Codes
- 200 OK - Health check completed successfully
Example Request
curl -X GET "http://localhost:8000/health" \
-H "accept: application/json"
Example Response
{
"ready": true,
"predictor_loaded": true,
"drift_baseline_loaded": true
}
Service Not Ready
{
"ready": false,
"predictor_loaded": false,
"drift_baseline_loaded": false
}
Implementation Details
Defined in src/api.py:275-281
Response Model: HealthResponse (src/api.py:50-53)
class HealthResponse(BaseModel):
ready: bool
predictor_loaded: bool
drift_baseline_loaded: bool
_MODEL: Trained scikit-learn model_THRESHOLD: Classification threshold_FEATURE_CFG: Feature engineering configuration_DRIFT_BASELINE: Baseline statistics for drift monitoring
Use Cases
- Container health probes - Kubernetes liveness/readiness checks
- Load balancer health checks - Ensure traffic only routes to ready instances
- Monitoring dashboards - Track service availability
- Pre-deployment validation - Verify artifacts loaded correctly after deployment