Federated learning - Federated Aggregator

Repository that contains a Proof of Concept for the implementation of a Federated Learning framework.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

Prerequisites

• python3
• docker
• docker-compose

Installing

A step by step series that tell you how to get a development env running

git clone [email protected]:DeltaML/federated_aggregator.git
cd federated_aggregator/
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Run

Using command line

    gunicorn -b "0.0.0.0:8080" --chdir federated_aggregator/ wsgi:app --preload

Using Docker

    docker build -t federated_aggregator --rm -f federated_aggregator/Dockerfile
    docker run --rm -it -p 8080:8080 federated_aggregator

Using Pycharm

Script Path: .../federated_aggregator/virtualenv/bin/gunicorn
Parameters: -b "0.0.0.0:8080" wsgi:app --preload
Working directory: ../federated_aggregator

Usage

Register new data owner

curl -v -H "Content-Type: application/json" -X POST "http://localhost:8080/dataowner"

Get data owners registered

curl -v -H "Content-Type: application/json" -X GET "http://localhost:8080/dataowner"

Train model

curl -v -H "Content-Type: application/json" -X POST -d '{"type": "LINEAR_REGRESSION", "call_back_endpoint": "URL_MODEL_BUYER", "call_back_port": 9090,"public_key": "XXXXXXXXXXXXXXXX"}' "http://localhost:8080/model"

Configuration

ACTIVE_ENCRYPTION = False
N_ITER = 100 # El numero de iteraciones aceptables utilizando PheEncryption por ahora es 4
DATA_OWNER_PORT = 5000

Configuration details

• ACTIVE_ENCRYPTION: TODO
• N_ITER: TODO
• DATA_OWNER_PORT: TODO

Versioning

We use SemVer for versioning. For the versions available, see the tags on this repository.

Authors

• Fabrizio Graffe - Dev - GFibrizo
• Agustin Rojas - Dev - agrojas

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE.md file for details