https://github.com/jlenon7/cartpole_gym_env

🤸 Gym environment solving the CartPole problem using Continuous Q-Learning.
https://github.com/jlenon7/cartpole_gym_env

Last synced: 9 days ago
JSON representation

🤸 Gym environment solving the CartPole problem using Continuous Q-Learning.

• Host: GitHub
• URL: https://github.com/jlenon7/cartpole_gym_env
• Owner: jlenon7
• License: mit
• Created: 2024-06-22T18:59:47.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2024-07-28T13:25:25.000Z (5 months ago)
• Last Synced: 2024-12-06T07:46:55.265Z (17 days ago)
• Language: Python
• Size: 1.13 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# OpenAI Cart Pole Gym Environment 🤸

> Gym environment solving the Cart Pole problem using Continuous Q-Learning.

## Results



> [!TIP]

> To train the model faster, comment the `env.render(mode='human')` call in `bin/agent.py` or change the `render_interval` variable as desired.

### Epoch points log tracker



## TODOs

- [x] Train the agent using Continuous Q-Learning

- [ ] Train the agent using Deep Q-Learning

- [ ] Save the model to be reused with a library like [stable-baselines](https://stable-baselines3.readthedocs.io/en/master/)

## Running

To run the gym environment first create a new Python environment and activate it. I'm using [Anaconda](https://www.anaconda.com/) for setting the python version that pipenv should use to set up the environment. The command bellow will automatically setup the environment with conda and pipenv:

```shell

make env

```

Now install all the project dependencies:

```shell

make install-all

```

To run the game to be played by a human run (use arrows <, >):

```shell

make play

```

To run the game and run random actions run:

```shell

make sample

```

To run the agent to complete the game task run:

```shell

make agent

```