Matlab vs Python: Reshape

Question

So I found this:

When converting MATLAB code it might be necessary to first reshape a matrix to a linear sequence, perform some indexing operations

Answer 1

Example:

MATLAB:

>> mafs = [(1:16)' (17:32)']
mafs =
     1    17
     2    18
     3    19
     4    20
     5    21
     6    22
     7    23
     8    24
     9    25
    10    26
    11    27
    12    28
    13    29
    14    30
    15    31
    16    32

>> reshape(mafs,[4 4 2])
ans(:,:,1) =
     1     5     9    13
     2     6    10    14
     3     7    11    15
     4     8    12    16
ans(:,:,2) =
    17    21    25    29
    18    22    26    30
    19    23    27    31
    20    24    28    32

Python:

>>> import numpy as np
>>> mafs = np.c_[np.arange(1,17), np.arange(17,33)]
>>> mafs.shape
(16, 2)
>>> mafs[:,0]
array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16])
>>> mafs[:,1]
array([17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32])

>>> r = np.reshape(mafs, (4,4,2), order="F")
>>> r.shape
(4, 4, 2)
>>> r[:,:,0]
array([[ 1,  5,  9, 13],
       [ 2,  6, 10, 14],
       [ 3,  7, 11, 15],
       [ 4,  8, 12, 16]])
>>> r[:,:,1]
array([[17, 21, 25, 29],
       [18, 22, 26, 30],
       [19, 23, 27, 31],
       [20, 24, 28, 32]])

Answer 2

I was having a similar issue myself, as I am also trying to make the transition from MATLAB to Python. I was finally able to convert a numpy matrix, given in depth, row, col, format to a single sheet of column vectors (per image).

In MATLAB I would have done something like:

output = reshape(imStack,[row*col,depth])

In Python this seems to translate to:

import numpy as np
output=np.transpose(imStack)
output=output.reshape((row*col, depth), order='F')