I'm looking for creating a random dimension numpy array, iterate and replace values per 10 for example.
I tried :
# Import numpy library
import numpy as np
def Iter_Replace(x):
print(x)
for i in range(x):
x[i] = 10
print(x)
def main():
x = np.array(([1,2,2], [1,4,3]))
Iter_Replace(x)
main()
But I'm getting this error :
TypeError: only integer scalar arrays can be converted to a scalar index
There is a numpy function for this, numpy.full or numpy.full_like:
>>> x = np.array(([1,2,2], [1,4,3]))
>>> np.full(x.shape, 10)
array([[10, 10, 10],
[10, 10, 10]])
# OR,
>>> np.full_like(x, 10)
array([[10, 10, 10],
[10, 10, 10]])
If you want to iterate you can either use itertools.product:
>>> from itertools import product
>>> def Iter_Replace(x):
indices = product(*map(range, x.shape))
for index in indices:
x[tuple(index)] = 10
return x
>>> x = np.array([[1,2,2], [1,4,3]])
>>> Iter_Replace(x)
array([[10, 10, 10],
[10, 10, 10]])
Or, use np.nditer
>>> x = np.array([[1,2,2], [1,4,3]])
>>> for index in np.ndindex(x.shape):
x[index] = 10
>>> x
array([[10, 10, 10],
[10, 10, 10]])
x and then print. x = np.full_like(x, 10), print(x)You have two errors. There are missing parenthesis in the first line of main:
x = np.array(([1,2,2], [1,4,3]))
And you must replace range(x) by range(len(x)) in the Iter_Replace function.
range(x)does not make sense ifxis array. What were you trying to do?