Containerizing the code to run in parallel/run in a pipeline, and also possible bottlenecks

[Matthew C Tafazoli]

Hello,

We are currently trying to create our version of a vehicle speed checking model and this repository has been a great help. We were easily able to get the sample data to run and work perfectly thanks to the clear instructions.

Now we want to try using Kubernetes (KubeFlow) to break down the code and containerize parts of it to run in parallel (perhaps the 3 tasks mentioned in the description: Vehicle Detection, Vehicle Tracking, Speed Calculation) so that we can get the code to run faster/more efficiently. Would you (or anyone else reading this), have any idea about the feasibility of this? Our team is still new to containerization.

Also, side question, but could you possibly identify any bottlenecks when it comes to running this code that we may be able to speed up?

Thank You and Best Regards,

Matthew