TD3_Pytorch_for_Walker Twin Delayed Deep Deterministic Policy Gradient for Gait Learning (Walker2d-v2) by Pytorch Usage To train a new TD3 agent: python train.py To test trained TD3 agent: python test.py Dependancies python 3.7 mujoco-py 2.0.2.13 gym 0.15.4 Result References Official TD3 paper and code https://github.com/nikhilbarhate99/TD3-PyTorch-BipedalWalker-v2 Stable baseline spinning up https://sylim2357.github.io/paper%20review/td3/