The simple version of the horizontal federated learning project to achieve image classification learning. This project is simulated locally, and does not involve network communication details and failure processing. It only involves model aggregation functions, which can quickly verify the relevant algorithms and capabilities of federated learning.
python3 fl_integration.py
It will define a server object and multiple client objects respectively to simulate horizontal federation training scenarios.
Perform model aggregation on the local model uploaded by the selected client.
Receive commands and global models from the server, and use local data to train local models.
References: Federated Learning in Action
定义一个服务端对象和多个客户端对象,用来模拟横向联邦训练场景。
python3 fl_integration.py
将被选择的客户端上传的本地模型进行模型聚合。
接收服务端的下发指令和全局模型,并利用本地数 据进行局部模型训练。
参考资料:联邦学习实战