python – MATLAB矩阵乘法性能比NumPy快5倍

MATLAB

close all; clear;

N = 1000 + 100; % a few initial runs to be trimmed off at the end

a = 100;
b = 30;
c = 40;
d = 50;
A = rand(b,c,a);
B = rand(c,d,a);
C = zeros(b,a);

times = zeros(1,N);
for ii = 1:N
    tic
    C = mtimesx(A,B);
    times(ii) = toc;
end

times = times(101:end) * 1e3;

plot(times);
grid on;
title(median(times));

Python

import timeit
import numpy as np
import matplotlib.pyplot as plt


N = 1000 + 100  # a few initial runs to be trimmed off at the end

a = 100
b = 30
c = 40
d = 50
A = np.arange(a * b * c).reshape([a,b,c])
B = np.arange(a * c * d).reshape([a,d])
C = np.empty(a * b * d).reshape([a,d])

times = np.empty(N)

for i in range(N):
    start = timeit.default_timer()
    C = A @ B
    times[i] = timeit.default_timer() - start

times = times[101:] * 1e3

plt.plot(times,linewidth=0.5)
plt.grid()
plt.title(np.median(times))
plt.show()

>我的Python是使用OpenBLAS从pip安装的默认Python.
>我正在使用英特尔NUC i3.

MATLAB代码在1ms运行,而Python在5.8ms运行,我无法弄清楚为什么,因为它们似乎都在使用BLAS.

编辑

来自Anaconda：

In [7]: np.__config__.show()
mkl_info:
    libraries = ['mkl_rt']
    library_dirs = [...]
    define_macros = [('SCIPY_MKL_H',None),('HAVE_CBLAS',None)]
    include_dirs = [...]
blas_mkl_info:
    libraries = ['mkl_rt']
    library_dirs = [...]
    define_macros = [('SCIPY_MKL_H',None)]
    include_dirs = [...]
blas_opt_info:
    libraries = ['mkl_rt']
    library_dirs = [...]
    define_macros = [('SCIPY_MKL_H',None)]
    include_dirs = [...]
lapack_mkl_info:
    libraries = ['mkl_rt']
    library_dirs = [...]
    define_macros = [('SCIPY_MKL_H',None)]
    include_dirs = [...]
lapack_opt_info:
    libraries = ['mkl_rt']
    library_dirs = [...]
    define_macros = [('SCIPY_MKL_H',None)]
    include_dirs = [...]

从numpy使用pip

In [2]: np.__config__.show()
blas_mkl_info:
NOT AVAILABLE
blis_info:
NOT AVAILABLE
openblas_info:
    library_dirs = [...]
    libraries = ['openblas']
    language = f77
    define_macros = [('HAVE_CBLAS',None)]
blas_opt_info:
    library_dirs = [...]
    libraries = ['openblas']
    language = f77
    define_macros = [('HAVE_CBLAS',None)]
lapack_mkl_info:
NOT AVAILABLE
openblas_lapack_info:
    library_dirs = [...]
    libraries = ['openblas']
    language = f77
    define_macros = [('HAVE_CBLAS',None)]
lapack_opt_info:
    library_dirs = [...]
    libraries = ['openblas']
    language = f77
    define_macros = [('HAVE_CBLAS',None)]

编辑2
我试图用np.matmul(A,B,out = C)代替C = A @ B,并且时间缩短2倍,例如大约11ms.这真的很奇怪.