Re-implementation of the DFP network proposed by Arthur Juliani (https://github.com/awjuliani/dfp).
All the code is based on the O'Reilly article, "Reinforcement Learning for Various, Complex Goals, Using TensorFlow".
To run the project, a conda enviroment file is provided.
- To import the environment
conda env create -f environment.yml.
- Use pycryon as tensroboard portal. You can find the installing instruction at https://github.com/torrvision/crayon