Thoughts on hardware setup for Computer Vision software 

Hi Elnasir, The optimization is a complex operation and often more complex then the entire project itself. I can give you here some guidelines on what to look for. Mainly there are 4 things to look into: - python code must be optimized, avoid using python loops and instead use operation on the array with numpy library - you can use libraries like timeit , to check the speed of every chunk of your code on python and try to optimize it as much as possible. - the deep learning model must be converted into tensorRT format, optimised for the size of the data you need to pass into it (number of images per batch and size of image), tailored to the specific GPU you're using. - you need to leverage multiprocessing and split the task of your software into multiple process to leverage the power of the CPU