Developer Guide for Intel® Data Analytics Acceleration Library 2019 Update 1
The forward reshape layer generates a tensor from input argument X of size n 1 x n 2 x ... x n p with modified size of dimensions m 1 x m 2 x ... x m q without modifying data and its order. For more details, see Forward Reshape Layer. The backward reshape layer generates output tensor Z of size n 1 x n 2 x ... x n p from input tensor G of dimensions m 1 x m 2 x ... x m q .
Given the dimension sizes of the output tensor n 1 x n 2 x ... x n p , such that
The p-dimensional result tensor Z of size n 1 x n 2 x ... x n p contains the same data from G in unmodified order.