Using ROS with fast.ai

I am interested in applying deep learning in the field of robotics in order to obtain real world applications.

A project i am working on consists of a robot that can detect weeds and exterminate them.the robot will use the ROS framework for navigation and the like.

I want to apply what i learned here about the fast.ai library in that project,and my question is whether or not it is feasible,since ros uses python 2.7 while fast.ai and pytorch use python 3

ps: ROS 2.0 is not an option , at least for this particular project :stuck_out_tongue:

1 Like

sys.path.append("…/…/…/FastAi")
sys.path.remove(’/opt/ros/kinetic/lib/python2.7/dist-packages’)

1 Like

that woul allow you to use either ros or fastai,not both of them.anyway found out the best way to do this is to install fastai without conda and indicate in the shebang of your ros node that you are using python3.then you can import fast ai normally.

1 Like