I have two numpy arrays of shape
(129873, 12)
(129873,)
I would like to concatenate these so they are in the shape of:
(129873, 13)
I have tried numpy.concatenate and numpy.vstack but seem to be getting errors:
ValueError: all the input arrays must have same number of dimensions
Any suggestions on how to do this?
I think this was already answered here:
Numpy: Concatenating multidimensional and unidimensional arrays
import numpy
a = numpy.array([[1, 2, 3, 4, 5], [1, 2, 3, 4, 5], [1, 2, 3, 4, 5]])
b = numpy.array([5, 6, 6])
c = numpy.column_stack((a,b))
print a
print b
print c
print a.shape
print b.shape
print c.shape
This gives you:
[[1 2 3 4 5]
[1 2 3 4 5]
[1 2 3 4 5]]
[5 6 6]
[[1 2 3 4 5 5]
[1 2 3 4 5 6]
[1 2 3 4 5 6]]
(3, 5)
(3,)
(3, 6)
So one array has 2 dimensions, the other 1:
(129873, 12)
(129873,)
You need to change the 2nd to have shape (129873,1). Then you can concatenate on axis 1.
There are a number of way of do this. The [:,None] or np.newaxis indexing is my favorite:
In [648]: A=np.ones((3,4),int)
In [649]: B=np.ones((3,),int)
In [650]: B[:,None].shape
Out[650]: (3, 1)
In [651]: np.concatenate((A,B[:,None]),axis=1).shape
Out[651]: (3, 5)
B.reshape(-1,1) also works. Also np.atleast_2d(B).T and np.expand_dims(B,1).
np.column_stack((A,B)) uses np.array(arr, copy=False, subok=True, ndmin=2).T to ensure each array has the right number of dimensions.
While there are friendly covers to concatenate like column_stack, it's important to know how to change dimensions directly.
