I want to get the elements of a numpy array using an index array like so
import numpy
a = numpy.arange(6)
ind = [2,3]
now, a[ind] gives me the 3rd and 4th element, but I actually want all the other elements of a.
Is there a one line/ elegant way to do this?
There isn't a straightforward way I know of to get the complement of a set of integer indices. Boolean index negation is easy, which lets you do something like this:
In [100]: a=np.arange(6)
In [101]: ind=[2,3]
In [102]: cind=(a==a)
In [103]: cind[ind]=False
In [104]: a[cind]
Out[104]: array([0, 1, 4, 5])
But it isn't a one line solution.
np.ones(a.shape, 'bool') instead of (a==a)maybe like this:
import numpy
a = numpy.arange(6)
ind = [1,3]
for x in range(6):
if x not in ind: print a[x]
This has been suggested here before, but this is a list comprehension and therefore a oneliner:
numpy.array([a[i] for i in range(len(a)) if i not in ind])
results in
array([0, 1, 4, 5])
