I want to compare the elements of an array to a scalar and get an array with the maximum of the compared values. That's I want to call
import numpy as np
np.max([1,2,3,4], 3)
and want to get
array([3,3,3,4])
But I get
ValueError: 'axis' entry is out of bounds
When I run
np.max([[1,2,3,4], 3])
I get
[1, 2, 3, 4]
which is one of the two elements in the list that is not the result I seek for. Is there a Numpy solution for that which is fast as the other built-in functions?
This is already built into numpy with the function np.maximum:
a = np.arange(1,5)
n = 3
np.maximum(a, n)
#array([3, 3, 3, 4])
This doesn't mutate a:
a
#array([1, 2, 3, 4])
If you want to mutate the original array as in @jamylak's answer, you can give a as the output:
np.maximum(a, n, a)
#array([3, 3, 3, 4])
a
#array([3, 3, 3, 4])
Docs:
maximum(x1, x2[, out])Element-wise maximum of array elements.
Equivalent tonp.where(x1 > x2, x1, x2)but faster and does proper broadcasting.
np.maximum is another function than np.max (which also exists!). It won't work with max, you have to use maximum :)>>> import numpy as np
>>> a = np.array([1,2,3,4])
>>> n = 3
>>> a[a<n] = n
>>> a
array([3, 3, 3, 4])
