I have a very huge numpy array like this:
np.array([1, 2, 3, 4, 5, 6, 7 , ... , 12345])
I need to create subgroups of n elements (in the example n = 3) in another array like this:
np.array([[1, 2, 3],[4, 5, 6], [6, 7, 8], [...], [12340, 12341, 12342], [12343, 12344, 12345]])
I did accomplish that using normal python lists, just appending the subgroups to another list. But, I'm having a hard time trying to do that in numpy.
Any ideas how can I do that?
Thanks!
You can use np.reshape(-1, 3), where the -1 means "whatever's left".
>>> array = np.arange(1, 12346)
>>> array
array([ 1, 2, 3, ..., 12343, 12344, 12345])
>>> array.reshape(-1, 3)
array([[ 1, 2, 3],
[ 4, 5, 6],
[ 7, 8, 9],
...,
[12337, 12338, 12339],
[12340, 12341, 12342],
[12343, 12344, 12345]])
-1, the documentation mentioned "One shape dimension can be -1.", and this is a clever implementation of that.np.reshape():From the documentation (link in title):
numpy.reshape(a, newshape, order='C')Gives a new shape to an array without changing its data.
Here is an example of how you can apply it to your situation:
>>> import numpy as np
>>> a = np.array([1, 2, 3, 4, 5, 6, 7, 8, 12345])
>>> a.reshape((int(len(a)/3), 3))
array([[1, 2, 3],
[4, 5, 6],
[7, 8, 12345]], dtype=object)
Note that obviously, the length of the array (len(a)) has to be a multiple of 3 to be able to reshape it into a 2-dimensional numpy array, because they must be rectangular.
np.reshape?