Skip to content

Latest commit

 

History

History
35 lines (22 loc) · 1.88 KB

remote_development.md

File metadata and controls

35 lines (22 loc) · 1.88 KB

Remote development

VS Code gives a convinient way for remote development using devcontainer configuration files. It is suggested for AWML users to setup configuration by themselves based on current environment.

Configuration

Before you run container, you might need to configure environment yourself. This consists of:

  • devcontainer.json:
    • "BASE_IMAGE": "autoware-ml" can be exchanged with your current project image.
    • "--volume=${env:HOME}/autoware_data:/home/autoware/autoware_data" can be exchanged with you external storage path. The new user inside Docker container is autoware, therefore you might need to change symlink within /workspace/data inside Docker container thought.
    • "extensions": list can be suited for your personal preferences.
  • Dockerfile can be extended with you preferable tools.

Run container

VS Code

Using the Visual Studio Code with the Remote - Containers extension, you can develop Autoware in the containerized environment with ease.

Get the Visual Studio Code's Remote - Containers extension. And reopen the workspace in the container by selecting Remote-Containers: Reopen in Container from the Command Palette (F1).

Terminal

If you already run container and prefer to use system terminal rather than itegrated one in VS Code, you can log to current container using provided script:

./.devcontainer/enter.sh

Debug

Debugging train loops can be perforemd as its described in Python debugging in VS Code. For extra parameter you can update "args": [] in launch.json.