Protect Environment: The project environment is in a unity game. The objective is to collect as many yellow bannanas (+1 point) and avoid as many purple bananas (-1 point) as possible. The observation/state space has 37 values including velocity and a ray based world perception. The 4 actions forward back left and right. The state space is considered solved when the algorithm gets an average of 13 points over 100 episodes; however, my algorithm does achieve better performance than that.

Getting Started:

Dependencies tensorflow==1.7.1 Pillow>=4.2.1 matplotlib numpy>=1.11.0 jupyter pytest>=3.2.2 docopt pyyaml protobuf==3.5.2 grpcio==1.11.0 torch==0.4.0 pandas scipy ipykernel

Downloading unity environment: Github: https://github.com/Unity-Technologies/ml-agents Mac: https://s3-us-west-1.amazonaws.com/udacity-drlnd/P1/Banana/Banana.app.zip Windows (64-bit): https://s3-us-west-1.amazonaws.com/udacity-drlnd/P1/Banana/Banana_Windows_x86_64.zip

Instructions: Run all of the code cells in Navigation.ipynb. This is dependant on the model and agent scripts so ensure they are in the same directory.