Performance issue in the definition of interpn, neurite/tf/utils/utils.py (P1)

[DLPerf]

Hello, I found a performance issue in the definition of interpn, neurite/tf/utils/utils.py, tf.reshape will be calculated repeatedly during program execution, resulting in reduced efficiency. I think it should be created before the loop.

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.

[adalca]

@DLPerf good find, feel free to submit a PR.

Have you tested how much of a difference it makes? The for loop is only over the corners (so 4 element in 2D and 8 elements in 3D) and reshaping is a quick operation so my intuition would be that it doesnt hurt the performance, but perhaps you saw otherwise?