Tested on Anaconda Envs (py36-pytorch-tf-gpu.yml)
- python 3.6
- tensorflow 1.10.0
Reinforce Learning Example based on Gym library
- how to prepare the data
- how to train the data
- how to save the model
- how to load the model and evaluate
- how to predict the data
- how to improve the data-> now repeat the same procedures
to do : pick the best score as the last model
files:
README.md : logs
main.py : simple example in library, we can see how Gym libs work
RL-main.py : the first designed NN