emasquil / doom-rl

Deep reinforcement learning agents that play Doom using Python.

Doom-RL

---

Deep Reinforcement Learning agents that play different Doom scenarios using ViZDoom and Tensorflow.

Implementations

• Vanilla Deep Q Learning (DQN)

---

Instructions

To train and test the agents, inside of each folder (E.g dqn/):

• python basic_scenario.py train-model
• python basic_scenario.py test-model