Clean, single file, minimalistic and easy to read implementations of Deep Reinforcement Learning algorithms

These implementations are designed to be as simple as possible. No unnecessary python magic.

Currently implemented:

• DQN (DQN.py)
• REINFORCE (REINFORCE.py)
• Actor-Critic (AC.py)
• A2C
• PPO (PPO.py)
• SAC