Skip to content
This repository has been archived by the owner on Oct 26, 2024. It is now read-only.
/ RL-Simple Public archive

Simplified version using car on a hill environment

Notifications You must be signed in to change notification settings

scooby359/RL-Simple

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Environment Setup

Dependent Software

The environment needs the following software to be installed:

Used as it creates a container for each project, allowing you to select the Python and dependency versions for your project, without conflicting with any global settings.

No need to add this to path

For version control support

Used by Tensorflow for GPU support

The file cudnn64_7.dll is needed by Tensorflow for GPU support. The download should be extracted and placed somewhere on drive. The path to the DLL will must be manually added to the System Environment PATH variable.

* Registration to NVIDIA required

Process

It is recommended to use a machine with a compatabile NVIDIA graphics card that can be used by TensorFlow. The UCLan Games Lab has PCs with suitable NVIDIA GeForce GTX 1070 cards.

Not using a graphics card is significantly slower and not recommended.

  1. Open Pycharm and checkout code from Github https://github.com/scooby359/RL-Simple.git

  2. When prompted by PyCharm, set the environment to Python 3.6, or manually go to the project options and set.

  3. From the PyCharm terminal, verify Pip is available with pip -V

  4. Install packages for skimage (scikit-image), matplotlib by using the prompts in PyCharm.

  5. Install TensorFlow 1.10 with pip install tensorflow-gpu==1.10. If not using a machine with a suitable graphics card, install tensorflow==1.10 instead.

  6. Install Numpy with pip install numpy==1.17.4 Tensorflow will require version <= 1.14.5, but this isn't compatible with the rest of the project so needs to be manually changed.

  7. Install OpenAI Gym with pip install gym

  8. The project can then be run from PyCharm by right clicking on the Python file and selecting Run.

  9. After completion, result graphs will be displayed. The graphs, raw CSV values and the TensorFlow model are saved in ./result/[TIMESTAMP]/.

About

Simplified version using car on a hill environment

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages