Incompatible shape for array on vector function

[TomMaullin]

Reporting a bug

• I am using the latest released version of Numba (most recent is visible in
  the change log (https://github.com/numba/numba/blob/master/CHANGE_LOG).
• I have included below a minimal working reproducer (if you are unsure how
  to write one see http://matthewrocklin.com/blog/work/2018/02/28/minimal-bug-reports).

Bug

I have the following helper function to vectorize a 2D matrix. If I run it without numba it works well.

def mat2vec(matrix):
  
  #Return vectorised matrix
  return(matrix.transpose().reshape(matrix.shape[0]*matrix.shape[1],1))

# Example:
matrix = np.random.randn(3,3)
print(matrix)
print(mat2vec(matrix))

If I run it with @numba.jit(nopython=True) prepended to the function I get:

---------------------------------------------------------------------------
NotImplementedError                       Traceback (most recent call last)
<ipython-input-46-bce81c0a0a5c> in <module>()
      8 matrix = np.random.randn(3,3)
      9 print(matrix)
---> 10 print(mat2vec(matrix))

NotImplementedError: incompatible shape for array

Apologies if this is documented behaviour/a non-issue and I have misunderstood but I cannot work out why this happens.

[sklam]

Confirmed bug. The reshape on the original non-transposed matrix works fine. Seems to be a logic error when the layout is changed.