search

simoninithomas / Deep_reinforcement_learning_Course

Implementations from the free course Deep Reinforcement Learning with Tensorflow and PyTorch
http://www.simoninithomas.com/deep-rl-course
3.79k stars 1.23k forks source link

Fix bugs in "Watch our Agent play" #3

Closed princessofpillows closed 6 years ago

princessofpillows commented 6 years ago

The function "stack_frames" accepts three arguments, where only two are currently given. "True" is passed as the third argument to fix this bug. Furthermore, stack_frames returns two objects, and they are both being assigned to state currently. Fix is to separate into "state, frames = stack_frames()".

simoninithomas commented 6 years ago

Hello, thanks for you submission!

You've corrected the former version of the notebook, effectively there was a bug in Doom DQN, I completely reworked this implementation you can see the new notebook one week ago here https://github.com/simoninithomas/Deep_reinforcement_learning_Course/blob/master/DQN/doom/Deep%20Q%20learning%20with%20Doom.ipynb.