SimplySAC replicates Soft-Actor-Critic with minimum (~200) lines of code in clean, readable PyTorch style, while trying to use as few additional tricks and hyper-parameters as possible (MuJoCo and PyBullet benchmarks included).
SimplyPPO replicates Proximal-Policy-Optimization with minimum (~250) lines of code in clean, readable PyTorch style, while trying to use as few additional tricks and hyper-parameters as possible (PyBullet benchmarks included).