Skip to content

joosthart/cartpole

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

cartpole

The repository contains three reinforcement learning algorithm which solve the CartPole problem. Using the minimal CLI, simple grid search can be run on the different algorithms. Furthermore, simulations can be run which illustrate the performance of the models.

installation

Install the necessary packages using:

pip install -r requirements.txt

Usage

All experiments can be run using:

python main.py -r all

The progress of the Tabular Q-Learning and Deep Q-Learning algorithm runs can be displayed in Tensorboard. To start Tensorboard, run:

tensorboard --log-dir ./log/

Note: The grid search algorithm is far from optimized and might take several hours to run.

With the following command, 5 episodes with a maximum of 150 steps of the DQL agent will be shown:

python main.py --simulate DQL --episodes 5 --max-steps 150

All options can be displayed using:

python main.py --help

Acknoledgments

We had help from some very good blogs and code examples. Below a list of the most helpful ones.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages