We have now put together all the pieces for the deep Q-learner and are ready to train the agent! Be sure to check out/pull/download the latest code from this book's code repository.
You can pick any environment from the list of Atari environments and train the agent we developed using the following command:
(rl_gym_book) praveen@ubuntu:~/HOIAWOG/ch6$ python deep_Q_learner.py --env "ENV_ID"
In the previous command, ENV_ID is the name/ID of the Atari Gym environment. For example, if you want to train the agent on the pong environment with no frame skip, you would run the following command:
(rl_gym_book) praveen@ubuntu:~/HOIAWOG/ch6$ python deep_Q_learner.py --env "PongNoFrameskip-v4"
By default, the training ...