Coach is a python framework that models the interaction between an agent and an environment in a modular way. With Coach, it is possible to model an agent by combining various building blocks, and training the agent on multiple environments. The available environments allow testing the agent in different fields such as robotics, autonomous driving, games and more. It exposes a set of easy-to-use APIs for experimenting with new RL algorithms and allows simple integration of new environments to solve. Coach collects statistics from the training process and supports advanced visualization techniques for debugging the agent being trained. Coach supports many state-of-the-art reinforcement learning algorithms, which are separated into three main classes - value optimization, policy optimization, and imitation learning. Coach supports a large number of environments which can be solved using reinforcement learning.
Features
- Coach is a python reinforcement learning framework containing implementation of many state-of-the-art algorithms
- Training an agent to solve an environment is easy
- It exposes a set of easy-to-use APIs for experimenting with new RL algorithms
- Distributed Multi-Node Coach
- Coach has only been tested on Ubuntu 16.04 LTS, and with Python 3.5.
- Coach outputs several signals, per trained algorithm, in order to track algorithmic performance