Hello!
For this article we take Tensor as a N-dimensional array whose last two axes might be interpreted as Matrix and/or the last axis might be interpreted as Vector. For TLDR the result is here.
- Tensors (N-dimensional storage)
- Implementation of methods and functions for working with tensors as with data storages