Skip to content

duckietown/template-machine-learning-pytorch

Repository files navigation

Template: template-basic

This template provides a boilerplate repository for developing non-ROS software in Duckietown.

NOTE: If you want to develop software that uses ROS, check out this template.

How to use it

1. Fork this repository

Use the fork button in the top-right corner of the github page to fork this template repository.

2. Create a new repository

Create a new repository on github.com while specifying the newly forked template repository as a template for your new repository.

3. Define dependencies

List the dependencies in the files dependencies-apt.txt and dependencies-py3.txt (apt packages and pip packages respectively).

4. Place your code

Place your code in the directory /packages/ of your new repository.

5. Setup launchers

The directory /launchers can contain as many launchers (launching scripts) as you want. A default launcher called default.sh must always be present.

If you create an executable script (i.e., a file with a valid shebang statement) a launcher will be created for it. For example, the script file /launchers/my-launcher.sh will be available inside the Docker image as the binary dt-launcher-my-launcher.

When launching a new container, you can simply provide dt-launcher-my-launcher as command.

About

Template for ML applications based on PyTorch

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published