Some times it's more comfortable to use numpy instead of LateX to describe things.

• Tensor dimensions written as array slices The matrix: $$ X \in R^{m x n}$$

  can be written as X[:m, :n] literal meaning is show me first m rows and n columns. I write dimension as slices that has the property X == X[:m, :n]

• Here you can find all references I made to this concept. See [[concepts]] page for all other concepts