Lecture - Perception and Learning for Robotics - Best Practices ML-Project

[@nova2577]

Look forward to your next lecture

[@befreebetrue6500]

Amazing.. Looking forward to your next lecture

[@hamidhussain5488]

Thanks for the wonderful lecture.

[@simon-ottenhaus]

Hi Jonas, great lecture. Nice introduction into the basics. If you want to go a little further, e.g. MLOps you could consider DVC for data versioning.

[@leggedrobotics]

Hi Simon, thanks a lot. I actually never used DVC but this looks like the perfect tool for the job. I will definitely give it a shot in my future projects.

[@galileo3431]

Great video! What are the three bullet points for JAX at 21:44 ? It's covered by the camera view.

[@leggedrobotics]

Thanks a lot. The bullet points state: - New framework / functional - Highly popular in research - Smaller community - Less stable

[@SanjeevKumar-nm7ps]

Thanks for sharing best practices. I have a question: how can we use the Python virtual environment with ROS? Does that mean we need to install the ROS distribution in the virtual environment, or do we install the ROS distribution at the system level and then access its functionality in the virtual environment?

[@leggedrobotics]

Hi, you can simply install ROS on your system and follow the normal virtual environment installation. The virtual environment will, by default, extend the installation of your system's Python. Just make sure to source your ROS and virtual environment correctly and everything should work by default.