The Dashboard API provides a single endpoint that aggregates system-wide KPIs from the database and the ML model evaluation artifacts. Use it to populate a summary view, monitor data completeness, and track model quality over time.Documentation Index
Fetch the complete documentation index at: https://cors-lau.vercel.app/docs/llms.txt
Use this file to discover all available pages before exploring further.
GET /dashboard/metrics
Returns a snapshot of key performance indicators for the current state of the CORS system.How the metrics are computed
| Field | What it reflects |
|---|---|
total_courses | Total courses in your uploaded course catalog |
total_offerings | Total historical enrollment records across all semesters and campuses |
latest_run_id | ID of the most recently completed prediction run (null if none) |
average_model_confidence | ROC-AUC score of the ML model associated with the latest run, as a percentage |
average_fail_ratio | Global average failure rate across all historical offering records, as a percentage |
total_predictions | Number of courses scored in the latest prediction run |
total_recommended_offerings | Number of courses marked offer = true in the latest run |
average_model_confidence is only populated after you have run at least one prediction. If no prediction run exists, or if models need retraining, call POST /predict/train and then run a new analysis via POST /predict/bulk.Responses
Total number of validated course records in the catalog.
Total number of historical enrollment records across all semesters and campuses.
ID of the most recent prediction run.
null if no prediction run has been created yet.ROC-AUC score of the campus model associated with the latest prediction run, expressed as a percentage. Example:
82.3 means 82.3% AUC. Returns 0 if no model artifact is found.Global historical course failure rate as a percentage, averaged across all offering records. Example:
18.7 means 18.7% of enrolled students historically fail.Total number of course entries scored in the latest prediction run.
0 if no run exists.Number of courses in the latest run that received an
offer = true recommendation. 0 if no run exists.