Sorting an numpy array in numba no-python mode

Question

Numba documentation suggests the following code should compile

@njit()
def accuracy(x,y,z):
    x.argsort(axis=1)
    # compare accuracy, this code works without the above line  
    accuracy_y = int(((np.equal(y, x).mean())*100)%100)
    accuracy_z = int(((np.equal(z, x).mean())*100)%100)
    return accuracy_y,accuracy_z

It fails on x.argsort(), I have also tried the following with and without axis arguments

np.argsort(x)
np.sort(x)
x.sort()

However I get the following failed to compile error (or similar):

numba.core.errors.TypingError: Failed in nopython mode pipeline (step: nopython frontend)
No implementation of function Function(<function sort at 0x000001B3B2CD2EE0>) found for signature:
 
 >>> sort(array(int64, 2d, C))
 
There are 2 candidate implementations:
  - Of which 2 did not match due to:
  Overload of function 'sort': File: numba\core\typing\npydecl.py: Line 665.
    With argument(s): '(array(int64, 2d, C))':
   No match.

During: resolving callee type: Function(<function sort at 0x000001B3B2CD2EE0>)



File "accuracy.py", line 148:
def accuracy(x,lm,sfn):
    <source elided>
    # accuracy
    np.sort(x)
    ^

What am I missing here?

You seem to be missing the part that declares and fills z, y and z. — Mark Setchell
– Mark Setchell, Commented Sep 7, 2021 at 20:30
I pass arrays into this in the function however in it's current form it doesn't matter what you try to pass it won't compile due to the np.sort(x) line being passed into numba. — tjaqu787
– tjaqu787, Commented Sep 7, 2021 at 20:35
It seems you are passing a 2d array but numba only supports 1d — MaxNoe
– MaxNoe, Commented Sep 7, 2021 at 20:40
What you pass to it defines what is compiled. That is the whole point of Numba. I would guess the two dimensional arrays you haven’t shown are the problem. The compiler probably only has a lowering for 1D inputs — talonmies
– talonmies, Commented Sep 7, 2021 at 20:42

Rutger Kassies · Accepted Answer · 2021-09-08 06:48:49Z

You could also consider using guvectorize if it fits your use case. That gives the benefit of being able to specify the axis to sort over. Sorting over more than 1 dimension can be done by repeated calls over a different axis.

@guvectorize("(n)->(n)")
def sort_array(x, out):
    out[:] = np.sort(x)

Using a slightly different example array that also has columns unordered.

arr = np.array([
    [6,5,4],
    [3,2,1],
    [9,8,7]],
)

sort_array(arr, out, axis=0)
sort_array(out, out, axis=1)

Shows:

array([[1, 2, 3],
       [4, 5, 6],
       [7, 8, 9]])

Making a simple wrapper would allow sorting of an arbitrary number of dimensions at once. I don't think Numba's overload supports guvectorize, otherwise you could even use it to make np.sort work inside your jitted functions without having to change anything.

https://numba.pydata.org/numba-doc/dev/extending/overloading-guide.html

Testing the output compared to Numpy:

for _ in range(20):
    
    arr = np.random.randint(0, 99, (9,9))

    # numba sorting
    out_nb = np.empty_like(arr)
    sort_array(arr, out_nb, axis=0)
    sort_array(out_nb, out_nb, axis=1)

    # numpy sorting
    out_np = np.sort(arr, axis=0)
    out_np = np.sort(out_np, axis=1)

    np.testing.assert_array_equal(out_nb, out_np)

tjaqu787 · Accepted Answer · 2021-09-07 22:01:11Z

0

Thank you for the comments! The following function sorts a 2d array!

from numba import njit
import numpy  as np

@njit()
def sort_2d_array(x):
    n,m=np.shape(x)
    for row in range(n):
        x[row]=np.sort(x[row])
    return x

arr=np.array([[3,2,1],[6,5,4],[9,8,7]])
y=sort_2d_array(arr)
print(y)

edited Sep 7, 2021 at 22:01

answered Sep 7, 2021 at 21:52

tjaqu787

3273 silver badges18 bronze badges

1 Comment

tjaqu787 Over a year ago

Note: there's probably and easy way to use @vectorize on this function

Collectives™ on Stack Overflow

Sorting an numpy array in numba no-python mode

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related