# Tempo AI

Tempo AI is a self-hosted home automation app. Tempo AI forecasts the French Tempo electricity tariff color (blue, white or red) two to four days ahead with a small local LightGBM model trained on public RTE and Open-Meteo data. It is open source and self-hosted, installable on TrueNAS.

Canonical page: https://ourlocalhost.com/app/7609/tempo-ai

## Description

Tempo AI forecasts the French Tempo electricity tariff color (blue, white or red) two to four days ahead with a small local LightGBM model trained on public RTE and Open-Meteo data. It checks its own forecasts every day and exposes a REST API and a web page.

## Platform availability

Available on 1 of 10 platforms: TrueNAS.

## Health score

40/100 (Fair), based on 3 of 4 factors.
- Maintained: 72 (last commit 10 days ago)
- Popular: 0 (0 GitHub stars)
- Easy to install: 25 (1 install platform, install notes)

## Links

- TrueNAS: https://apps.truenas.com/catalog/tempo-ai/
- GitHub: https://github.com/lgm42/tempo-ai

## Common questions

### Which TCP port should I expose for Tempo AI in Docker?

The container serves the app on 8080, with the example publishing it as host port 31090. It should run with a persistent /data volume and TZ=Europe/Paris.

### How long does the first run take, and is a GPU required?

The first start takes about 10 minutes while it downloads historical data and trains its models. Training runs on a CPU in under a minute, and no GPU is needed.

### What data should I persist or back up for Tempo AI?

Persist TEMPO_DATA_DIR, which defaults to /data, because it stores raw data, models, and the SQLite database. The data folder is about 70 MB.

### How can I install Tempo AI on TrueNAS before it appears in the catalog?

Use Apps, Discover, then Install via YAML with the provided custom-app.yaml file. The repository also includes TrueNAS catalog files under truenas/tempo-ai.

---
Source: Our Local Host (https://ourlocalhost.com/app/7609/tempo-ai), the self-hosted and open-source app directory.
