https://github.com/manics/flowerai-docker-example
https://github.com/manics/flowerai-docker-example
Last synced: 19 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/manics/flowerai-docker-example
- Owner: manics
- Created: 2026-04-02T22:09:24.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2026-04-03T21:14:16.000Z (2 months ago)
- Last Synced: 2026-04-03T23:09:03.647Z (2 months ago)
- Language: Python
- Size: 7.81 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Flower AI docker Compose example
```sh
docker compose up -d
docker compose logs -f jupyterlab
```
You should see a URL like `http://localhost:8888/lab?token=0123456789abcdef0123456789abcdef0123456789abcdef`
Open the URL in a browser and start a JupyterLab terminal.
Run:
```sh
export FLWR_LOG_LEVEL=DEBUG
flwr config list
flwr run quickstart-pytorch-usps test-deployment --stream
```
This should run the training example.
You should find the model in the `./outputs` directory.
## Containers
- **`superlink`**: The central coordinating service that routes messages between the `ServerApp` and the `SuperNode`s.
- **`supernode-1`**: Represents the first client edge node (configured for data partition 0) and communicates with the SuperLink.
- **`supernode-2`**: Represents the second client edge node (configured for data partition 1) and communicates with the SuperLink.
- **`serverapp-runner`**: Executes the central aggregation logic (`ServerApp`) and connects to the SuperLink.
- **`clientapp-runner-1`**: Executes the local model training and evaluation (`ClientApp`) for the first client, connecting to `supernode-1`.
- **`clientapp-runner-2`**: Executes the local model training and evaluation (`ClientApp`) for the second client, connecting to `supernode-2`.
- **`jupyterlab`**: Provides the web-based IDE and terminal to interactively run commands and trigger the federated learning workload.