Re: [Numpy-discussion] reshape 2D array into 3D
On Mon, Jul 10, 2017 at 8:46 AM Yarko Tymciurak
On Mon, Jul 10, 2017 at 8:37 AM
wrote: Thanks
Nevertheless it does not work for me and I suspect the python/numpy releases :-(
The server on which I'm working on is under Contos 7 that uses python 2.7 et numpy 1.7 from memory ; I tried to upgrade both of them (plus spyder) but it fails.
Question: Did you try to control the python & numpy versions by creating a virtualenv, or a conda env? I didn't want to impact the other solvers installed on, so I stopped
Paul
a = np.arange(40).reshape(5, 8); print(a) print("b =") b = np.lib.stride_tricks.as_strided(a, (2, 5, 4), (16, 32, 4)); print(b)
[[ 0 1 2 3 4 5 6 7] [ 8 9 10 11 12 13 14 15] [16 17 18 19 20 21 22 23] [24 25 26 27 28 29 30 31] [32 33 34 35 36 37 38 39]] b = [[[ 0 4294967296 1 8589934592] [ 4 21474836480 5 25769803776] [ 8 38654705664 9 42949672960] [ 12 55834574848 13 60129542144] [ 16 73014444032 17 77309411328]]
[[ 2 12884901888 3 17179869184] [ 6 30064771072 7 34359738368] [ 10 47244640256 11 51539607552] [ 14 64424509440 15 68719476736] [ 18 81604378624 19 85899345920]]]
Le 2017-07-10 15:16, eat a écrit :
Hi,
On Mon, Jul 10, 2017 at 3:20 PM,
wrote: Dear All
I'm looking in a way to reshape a 2D matrix into a 3D one ; in my example I want to *move the columns from the 4th to the 8th in the 2nd plane* (3rd dimension i guess)
a = np.random.rand(5,8); print(a)
I tried
a = p.reshape(d, (2,5,4), ) but it is not what I'm expecting
Nota : it looks like the following task (while I want to split it in 2 levels and not in 4), but I've not understood at all
https://stackoverflow.com/questions/31686989/numpy-reshape-and-partition-2d-...
Is this what you are looking for: import numpy as np
a= np.arange(40).reshape(5, 8)
a Out[]: array([[ 0, 1, 2, 3, 4, 5, 6, 7], [ 8, 9, 10, 11, 12, 13, 14, 15], [16, 17, 18, 19, 20, 21, 22, 23], [24, 25, 26, 27, 28, 29, 30, 31], [32, 33, 34, 35, 36, 37, 38, 39]])
np.lib.stride_tricks.as_strided(a, (2, 5, 4), (16, 32, 4)) Out[]: array([[[ 0, 1, 2, 3], [ 8, 9, 10, 11], [16, 17, 18, 19], [24, 25, 26, 27], [32, 33, 34, 35]],
[[ 4, 5, 6, 7], [12, 13, 14, 15], [20, 21, 22, 23], [28, 29, 30, 31], [36, 37, 38, 39]]])
Regards, -eat
Thanks for your support
Paul
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@python.org https://mail.python.org/mailman/listinfo/numpy-discussion
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@python.org https://mail.python.org/mailman/listinfo/numpy-discussion
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@python.org https://mail.python.org/mailman/listinfo/numpy-discussion
"Question: Did you try to control the python & numpy versions by creating a virtualenv, or a conda env?" I've just downloaded (ana)conda, but I've to take care first that it does not substitute to current python release working for for other solvers thanks for the information's and the support Paul
participants (2)
-
paul.carrico@free.fr
-
Yarko Tymciurak