A federated learning simulation framework to train language model on social media data.
Hangrui Cao, Jiayi Pan, Zhiyuan Chen
Link: https://www.yelp.com/dataset/documentation/main Yelp Dataset Links: https://archive.org/details/stackexchange Superuser dataset
How to process dataset: Use the build_vocab to build the vocabulary for yelp/superuser dataset
python main_fed.py --dataset yelp --model nlpmodel --epochs 400Reference: Graphsage link: https://github.com/williamleif/GraphSAGE