Specifically:
- How can I define a function output data type as
numpy.ndarray? - Is it possible to utilize
cimport numpyinstead ofimport numpyin order to create an array without Python overhead?
If the line
cdef numpy.ndarray array(int start, int end):numpy.ndarray, then the code below functions. However, there still seems to be some Python overhead indicated in the annotation (excluding the initialization of the C++ vector using range(start, end)).
%%cython -a
# distutils: language = c++
import numpy
from libcpp.vector cimport vector
cdef numpy.ndarray array(int start, int end):
cdef vector[int] vect = range(start, end)
return numpy.array(vect)
print(array(1,15))