Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Pierrot-01/Hackathon_epis_2026/llms.txt

Use this file to discover all available pages before exploring further.

The /api/stats endpoint aggregates student risk classifications into a compact summary used by the admin.html dashboard to populate the four top-level statistics cards: total students, high risk count, medium risk count, and low risk count. It is designed to be fast regardless of dataset size.

​
GET /api/stats

Returns global risk-level counts and percentages for all students in the dataset. No parameters are required.

​
Caching Behavior

The endpoint checks whether the in-memory session cache (_resultados_cache) is populated:
  • Cache available — reads nivel_riesgo directly from the cached EstudianteResultado objects. This is essentially instantaneous.
  • Cache empty — loads the full dataset from data/estudiantes.json and runs clasificar_estudiante() on each record synchronously, without calling the AI service. This makes the cold-start path significantly faster than GET /api/estudiantes, at the cost of not populating the AI explanation cache.
In both cases the counts and percentages returned are consistent with the classifier’s current rule set.

​
Response Fields

​
total
integer
required
Total number of student records in the dataset. Example: 20.
​
alto
integer
required
Count of students classified as 🔴 high risk.
​
medio
integer
required
Count of students classified as 🟡 medium risk.
​
bajo
integer
required
Count of students classified as 🟢 low risk.
​
insuficiente
integer
required
Count of students classified as ⚪ insufficient data — too many variables were null to produce a meaningful risk level.
​
pct_alto
float
required
Percentage of students at 🔴 high risk, rounded to one decimal place. Calculated as alto / total * 100. Returns 0 if total is 0.
​
pct_medio
float
required
Percentage of students at 🟡 medium risk, rounded to one decimal place.
​
pct_bajo
float
required
Percentage of students at 🟢 low risk, rounded to one decimal place.

​
Example

curl http://localhost:8000/api/stats

{
  "total": 20,
  "alto": 4,
  "medio": 7,
  "bajo": 7,
  "insuficiente": 2,
  "pct_alto": 20.0,
  "pct_medio": 35.0,
  "pct_bajo": 35.0
}
pct_alto + pct_medio + pct_bajo will not equal 100% when insuficiente > 0, because ⚪ students are counted in total but not included in any percentage field.

​
Dashboard Integration

The admin.html frontend calls GET /api/stats on page load to populate the summary statistics cards at the top of the administration view. Because the stats endpoint uses the classifier-only fast path when the cache is cold, these cards load quickly even on a fresh server start — before the full AI-enriched student list has been fetched.

​
GET /api/health

The health endpoint provides a simple liveness check for the server. It is typically called on startup to confirm the API is running before the frontend attempts any data requests. No parameters are required.

​
Response

{
  "status": "ok",
  "version": "1.0.0"
}
​
status
string
required
Always "ok" when the server is running and able to handle requests.
​
version
string
required
Current API version string. Value: "1.0.0".

​
Example

curl http://localhost:8000/api/health
This endpoint is safe to call repeatedly — it performs no I/O and returns in under 1 ms. Use it for:
  • Startup probes: confirm the server is ready before loading the dashboard.
  • Health monitoring: integrate with process managers (e.g., systemd, PM2) or uptime monitors to detect server crashes.
  • Development: quickly verify the API is reachable after a code change or restart.

Build docs developers (and LLMs) love

Get started for freeTalk to us