This repository has the code from my O'Reilly article 'Reinforcement Learning w/ TensorFlow' published on.
This code contains tools for implementing the reinforcement learning algorthing REINFORCE w/ baselines using TensorFlow and a convolutional Neural Network
There are two ways you can install these packages: by using Docker or by using native Python 3.5. If you use Docker you will not be able to render the environment.
-
Download and install Docker. If using Ubuntu 14.04/16.04 I wrote my own instructions for installing docker here.
-
Download and unzip this entire repo from GitHub, either interactively, or by entering
git clone https://github.com/wagonhelm/Deep-Policy-Gradient.git
-
Open your terminal and use
cdto navigate into the directory of the repo on your machinecd Deep-Policy-Gradient -
To build the Dockerfile, enter
docker build -t dpg_dockerfile -f dockerfile .If you get a permissions error on running this command, you may need to run it with
sudo:sudo docker build -t dpg_dockerfile -f dockerfile . -
Run Docker from the Dockerfile you've just built
docker run -it -p 8888:8888 -p 6006:6006 dpg_dockerfile bash
or
sudo docker run -it -p 8888:8888 -p 6006:6006 dpg_dockerfile bash
if you run into permission problems.
-
Launch Jupyter and Tensorboard both by using tmux
tmux jupyter notebook --allow-root
Press CTL+BthenCto open a new tmux window, thentensorboard --logdir='/tmp/dpg'To switch windows
Press CTL+Bthenwindow #Once both jupyter and tensorboard are running, using your browser, navigate to the URLs shown in the terminal output if those don't work try http://localhost:8888/ for Jupyter Notebook and http://localhost:6006/ for Tensorboard.
- Install system requirements
sudo apt-get install git cmake zlib1g-delibjpeg-dev libboost-all-dev libsdl2-dev wget unzip gitboostbuild-essential zlib1g-dev libsdl2-dev libjpeg-dev nasm tar libbz2-dev libgtk2.0-dev cmake libfluidsynth-dev libgme-dev libopenal-dev timidity libwildmidi-dev python3-pip python3-dev python3-wheel- Download and unzip this entire repo from GitHub, either interactively, or by entering
git clone https://github.com/wagonhelm/Deep-Policy-Gradient.git- Install Python Requirement
cd Deep-Policy-Gradient
pip3 install -r requirements.txt
pip3 install tensorflow jupyter- Launch Jupyter
jupyter notebook- Launch Tensorboard
tensorboard --logdir='/tmp/dpg'Once both jupyter and tensorboard are running, using your browser, navigate to the URLs shown in the terminal output if those don't work try http://localhost:8888/ for Jupyter Notebook and http://localhost:6006/ for Tensorboard.