Skip to content

Inverted Pendulum classic reinforcement learning problem.

Notifications You must be signed in to change notification settings

gegelati/app-pendulum

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pendulum

This application teaches a learning agent built with the GEGELATI library how to keep a pendulum at an equilibrium position.

How to Build?

The build process of applications relies on cmake to configure a project for a wide variety of development environments and operating systems. Install cmake on your system before building the application.

Under windows

  1. Copy the gegelatilib-<version> folder containing the binaries of the GEGELATI library into the lib folder.
  2. Open a command line interface in the bin folder.
  3. Enter the following command to create the project for your favorite IDE cmake ...
  4. Open the project created in the bin folder, or launch the build with the following command: cmake --build ..

CodeGen example

The folder src/CodeGen contains an example of use case for the code generation feature of GEGELATI. There are 4 targets for this example:

  • pendulumCodeGenCompile: Import the TPG_graph.dot and launch the code gen to generate the sources files. If you want to run this target you need to set your working directory as the current build directory of your build system. You can use the following variable $CMakeCurrentBuildDir$.
  • pendulumCodeGenGenerate: A custom command to execute the previous target (after it is compiled)
  • pendulumCodeGenInference: Uses the generated file and link them with the learning environment of the directory è src/Learn. This target depend on the previous, so building it will automatically trigger a build of the two previous.
  • pendulumTPGInference: Import the TPG_graph.dot and run it within the pendulum learning environment, in the exact same condition as within the pendulumCodeGenGenerate target. This target enables comparing the identical behavior of the generated code and the original TPG.

About

Inverted Pendulum classic reinforcement learning problem.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •