This project aims to implement a somewhat okay tensor library that can be used to create neural networks for ML problems. I wanted to refresh my knowledge on templates, STL and overall OOP in C++ and explore new concepts and quirks of the language and as such I have probably overengineered most of the code. I also wanted to understand what goes in to a tensor library such as Tensorflow or PyTorch, and to refresh my knowledge on how a network with backpropagtaion actually works. Everything is thus created through my own understanding of what is correct.
The code is not well documented and completely handwritten, and might as such be a bit unclear and maybe not optimal at times, but the points is not to compete with dedicated libraries, but rather to learn new concepts and refresh old knowledge.