Playing Black Jack with Reinforcement Learning
- Python3
- Virtualenv
Cloning:
$ git clone https://github.com/jrodrigomg/black-ai
$ cd black-ai/ai
Creating a virtual environment:
$ virtualenv -p python3.5 env
Load environment, this is an example for fish shell:
$ source env/bin/activate.fish
Install requirments:
$ pip install -r requirements.txt
Run the main file:
$ python main.py
In the main directory go to the game carpet:
$ cd game
Run a simple http server with python(Not necesary the enviroment here)
$ python -m SimpleHTTPServer 8080
- Once running both application we need to configure wherever the server ask
- go http://localhost:8080 in the browser
- If just appear "Epoch 1" then refresh the browser
- Thanks to adventuresinML for the incredible post of RL (https://adventuresinmachinelearning.com/reinforcement-learning-tutorial-python-keras/)
- Thanks to Kusnierewicz for the game of blackjack in javascript (https://github.com/Kusnierewicz/Blackjack-game-in-JS)
- For the structure of the enviroment i used some part of Vincent Dutordoir's repo (https://github.com/vdutor/TF-rex)