Tensors using NumRuby - SciRuby

Tensor word can have slightly different meaning depending upon the nature of study, like it’s defined differently in Physics, slightly different in computer science. In computer terms, it is basically a n-dimensional array. A scalar (one value) is a tensor with 0 dimensions, a vector is a tensor with 1 dimension, a matrix is a tensor with 2 dimensions.It’s possible to define a tensor in Ruby using the Array class of Ruby but it gets tedious when defining multi-dimensional tensors. Also, the Array object is designed to be heterogeneous which means that the elements of the array can be of different type or different classes which would seem as a plus point overall but it has a huge downside. Due to Array being heterogeneous, the memory allocation has to be in such a way that any element of any size can be added or removed from the array, which causes a lot of re-allocations. Also, the indexing and other array functions gets slower due to the heterogeneous nature.