History

juvdiaz 240a55e826 Homelab Main / deploy (push) Successful in 1m33s Details adding mlops		2026-05-29 09:24:47 -06:00
..
app	adding mlops	2026-05-29 09:24:47 -06:00
models	adding mlops	2026-05-29 09:24:47 -06:00
training	adding mlops	2026-05-29 09:24:47 -06:00
Dockerfile	adding mlops	2026-05-29 09:24:47 -06:00
README.md	adding mlops	2026-05-29 09:24:47 -06:00
requirements.txt	adding mlops	2026-05-29 09:24:47 -06:00

README.md

MLOps Platform Demo

Production-shaped inference demo for the portfolio site. The model is intentionally small: logistic regression coefficients trained with scikit-learn and exported to JSON so the runtime stays light enough for the homelab.

Endpoints

GET /healthz reports service, track, and active model metadata.
POST /predict scores service health risk from latency, error rate, CPU, memory, and queue depth.
GET /metrics exposes Prometheus metrics for request count, latency, errors, model version, confidence, and drift score.

Model Rollout

MODEL_VERSION=v1, MODEL_TRACK=blue is the stable route.
MODEL_VERSION=v2, MODEL_TRACK=green is the canary route.
Kubernetes service selectors choose the active track, so rollback is a service selector change instead of an image rebuild.