将numpy数组传递给Cython

Question

I am learning Cython. I have problem with passing numpy arrays to Cython and don't really understand what is going on. Could you help me?

I have two simple arrays:

a = np.array([1,2])
b = np.array([[1,4],[3,4]])

I want to compute a dot product of them. In python/numpy everything works fine:

>>> np.dot(a,b)
array([ 7, 12])

I translated the code to Cython (as here: http://docs.cython.org/src/tutorial/numpy.html ):

import numpy as np
cimport numpy as np

DTYPE = np.int
ctypedef np.int_t DTYPE_t

def dot(np.ndarray a, np.ndarray b):
    cdef int d = np.dot(a, b)
    return d

It compiled with no problems but returns an error:

>>> dot(a,b)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "test.pyx", line 8, in test.dot (test.c:1262)
    cdef int d = np.dot(a, b)
TypeError: only length-1 arrays can be converted to Python scalars

Could you tell me why and how to do it correctly? Unfortunately Google was not helpful...

Thanks!

Answer 1

Your result is np.ndarray, not int. It fails trying to convert first one to latter. Do instead

def dot(np.ndarray a, np.ndarray b):
    cdef np.ndarray d = np.dot(a, b)
    return d