Overview
This project explores federated learning using the Flower framework, as well as other federated learning implementations with Gated Recurrent Units (GRU) and Long Short-Term Memory (LSTM) models. The goal is to compare and evaluate the performance of these models on the PeMS traffic dataset.
Models
- Federated STG ODE with Flower framework
- Federated STG ODE without Flower framework
- Federated GRU
- Federated LSTM
Dataset
The models are evaluated using the PeMS traffic dataset. This dataset contains traffic flow data collected from various sensors across California's highway network. It is widely used for benchmarking and evaluating traffic prediction models.
GitHub Repository
To access the source code and learn more about the project, visit our GitHub repository.