This project contains two settings. One is about simulations using PySyft framework, wherein we used six benchmark datasets (Synthetic; FEMNIST, CelebA; Sent140, Shakespeare, Reddit) provided by LEAF. The other is about experiments using dockers to communicate with the Ali server, wherein we used two datasets (FEMNIST and Reddit) for simplicity.