Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

Expose a detection level for sequences and alerts

Aperta
  • Qualcosa non va?
#718 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 3 giorni

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
5/5
Tempo stimato
Più di una settimana
Idoneità per principianti
35/100
Tipo di issue
Funzionalità
Chiarezza
Abbastanza chiara
Stato di attività
Attiva
Stack tecnologico
fastapi, postgresql, python
Ambito
api, backend

Direzione di ricerca

Start by reading the API's sequence and alert response definitions and the configuration for TEMPORAL_MODEL_THRESHOLD; the issue does not name their files. Then inspect how annotated sequences store is_wildfire and how model scores are recorded. Done means sequences and alerts return documented levels, thresholds are configurable, and their distribution is checked against annotated data; the alert aggregation and model version behavior still need decisions.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

type: feat
Context & Objective

Part of the epic pyronear/roadmap#28 (Alert analysis and follow-up). The API already exposes temporal_model_score per sequence. Users and tenders ask for something readable: how reliable is this alert? We want the API to translate the score into a few levels, so that every client shows the same thing.

What is done currently
  • Each sequence has temporal_model_score with its model versions. An alert groups the sequences that were triangulated together.
  • The score threshold used for validation lives in the API (TEMPORAL_MODEL_THRESHOLD). Nothing translates the score into levels.
What has to be done
  • Define three levels (high, medium, limited) from temporal_model_score, with the thresholds in configuration.
  • Return the level with sequences and alerts. For an alert, derive it from its sequences.
  • Handle sequences without a score (never scored, model unavailable, version change).
  • Look at the distribution of annotated sequences (is_wildfire) by score, to choose the thresholds: which share of each category falls in each level.
Open questions
  • How is an alert level built from its sequences: highest, or a combination?
  • What happens to stored levels when the model version changes?
  • Do we need a quick analysis of the thresholds (for Mateo), or are the validation threshold and its study (pyronear/vision-rd#143, pyronear/vision-rd#156) enough?
Done when
  • Sequences and alerts return a level, documented
  • Thresholds are in configuration and checked on annotated data
Lingua principale
Python
Stelle
30
Fork
15
Merge medio
4g 15h
PR unite (30g)
16

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di pyronear/pyro-api

Tutte le issue di pyronear/pyro-api

Issue simili

Altre issue su Python

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.