How do i use numpy.view with string data in numba

user1894205 0
Viewed 26

I can't get the following code to jit using numba. Is there an alternative method? I want the result to be an array of strings. The code runs exactly as expected without the numba jit.

from numba import jit
import numpy as np

@jit(nopython=True)
def test():
    chars = np.array([[97, 98, 99, 0, 0],[99, 98, 97, 0, 0]], dtype=np.uint8)
    return chars.view(dtype='<S5').astype(str).squeeze()

test()
numpy numba python
1 Answers

I asked about this on the numba gitter channel was told that this isn't supported. I settled for the following. There is additional processing in the real function which is jittable and pulling out this last small part allowed the speed ups to take place on all the rest to great effect.

from numba import jit
import numpy as np

@jit(nopython=True)
def test():
    chars = np.array([[97, 98, 99, 0, 0],[99, 98, 97, 0, 0]], dtype=np.uint8)
    dtypes5 = np.dtype('<S5')
    return chars.view(dtypes5)

test().astype(str).squeeze()
user18942050
Related