Question

For more control over the translation to C, most Python scientific developers will use the Cython package. Essentially, this is a language that resembles Python with type annotations. The Cython code is then compiled into native code tranaparently. The great advantage of Cythonn over ther approaches are:

• A Python program is also valid Cython program, so optimization can occur incrementally
• Fine degree of control over degree of optimization
• Easy to use - handles details about the C compiler and shared library generation
• Cythonmagic extension comes built into IPyhton notebook
• Can run parallel code with the nogil decorator
• Fully optimized code runs at thee same speed as C in most cases

%load_ext cythonmagic

The Cython magic has been moved to the Cython package, hence
%load_ext cythonmagic is deprecated; please use %load_ext Cython instead.

Though, because I am nice, I'll still try to load it for you this time.

%%cython

import numpy as np
cimport cython
from libc.math cimport sqrt

@cython.boundscheck(False)
@cython.wraparound(False)
def pdist_cython(double[:, ::1] xs):
    cdef int n = xs.shape[0]
    cdef int p = xs.shape[1]
    cdef double tmp, d
    cdef double[:, ::1] D = np.empty((n, n), dtype=np.float)
    for i in range(n):
        for j in range(n):
            d = 0.0
            for k in range(p):
                tmp = xs[i, k] - xs[j, k]
                d += tmp * tmp
            D[i, j] = sqrt(d)
    return np.asarray(D)

print pdist_cython(A)
%timeit pdist_cython(xs)

[[ 0.  5.]
 [ 5.  0.]]
100 loops, best of 3: 7.09 ms per loop