python 性能并不慢_为什么MATLAB/Numpy/Scipy性能很慢并且达不到CPU能力（flops）？

最新推荐文章于 2023-10-31 10:32:06 发布

青城山说

最新推荐文章于 2023-10-31 10:32:06 发布

阅读量558

点赞数

文章标签： python 性能并不慢

本文链接：https://blog.csdn.net/weixin_33138569/article/details/113965930

版权

博主对比了MATLAB和Python(Numpy, Scipy)在元素乘法操作上的性能，发现Python的性能显著低于预期的CPU Flops能力。在相同的操作中，MATLAB/Numpy/Scipy并未充分利用CPU潜能，引发了关于性能优化的讨论。" 110405345,10382342,Android开发入门教程：从创建项目到运行应用,"['Android开发', 'Android Studio', '用户界面']

摘要由CSDN通过智能技术生成

首先，我知道有多个线程涉及这个问题，但是我不能得到一个直接的答案，并遇到了一些失败的计算错误。在

我准备了一个MATLAB和Python的元素乘法基准测试。这是一种最简单、最前沿的方法，可以很容易地计算flop计数。在

它使用NxN数组(matrix)，但不执行矩阵乘法，而是按元素进行乘法。这一点很重要，因为当使用矩阵乘法时，运算数不是N^3！！！在

然而，随机生成的数的元素乘法运算必须在N^2次运算中执行

我有一个intel i7-4770(我想它有4个物理核和8个虚拟核)@3.5GHz。所以如果假设每个周期有4次浮点运算，那么每个核心应该是14次浮点运算！在

MATLAB/Numpy/Scipy别靠近它。在

为什么？在

MATLAB软件：%element wise multiplication benchmark

N = 10^4;

nOps = N^2;

m1 = randn(N);

m2 = randn(size(m1));

m = randn(size(m1));

m1 = single(m1);

m2 = single(m2);

% clear m

tic

m1 = m1 .* m2;

t = toc;

gflops = nOps/t*1e-9;

t_gflops = [t gflops]

% clear m

tic

m1 = m1.*m2;

t = toc;

gflops = nOps/t*1e-9;

t_gflops = [t gflops]

% clear m

tic

m1 = m1.*m2;

t = toc;

gflops = nOps/t*1e-9;

t_gflops = [t gflops]

version('-blas')

version('-lapack')

结果是：

^{pr2}$

现在Python：import numpy as np

# import gnumpy as gnp

import scipy as sp

import scipy.linalg as la

import time

if __name__ == '__main__':

N = 10**4

nOps = N**2

a = np.random.randn(N,N).astype(np.float32)

b = np.random.randn(N,N).astype(np.float32)

t = time.time()

c = a*b

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = np.multiply(a, b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = sp.multiply(a, b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = sp.multiply(a, b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

a = np.random.randn(N,1).astype(np.float32)

b = np.random.randn(1,N).astype(np.float32)

t = time.time()

c1 = np.dot(a, b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = np.dot(a, b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = la.blas.dgemm(1.0,a,b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

t = time.time()

c = la._fblas.dgemm(1.0,a,b)

dt = time.time()-t

gflops = nOps/dt*1e-9

print("dt = ", dt, ", gflops = ", gflops)

print("numpy config")

np.show_config()

print("scipy config")

sp.show_config()

# numpy

结果是：dt = 0.16301608085632324 , gflops = 0.6134364136022663

dt = 0.16701674461364746 , gflops = 0.5987423610209003

dt = 0.1770176887512207 , gflops = 0.5649152957845881

dt = 0.188018798828125 , gflops = 0.5318617107612401

dt = 0.151015043258667 , gflops = 0.6621856858903415

dt = 0.17201733589172363 , gflops = 0.5813367558659613

dt = 0.3080308437347412 , gflops = 0.3246428142959423

dt = 0.39503931999206543 , gflops = 0.253139358385916

numpy配置

mkl\U信息：define_macros = [('SCIPY_MKL_H', None), ('HAVE_CBLAS', None)]

libraries = ['mkl_core_dll', 'mkl_intel_lp64_dll', 'mkl_intel_thread_dll']