如何优化以下循环代码? [英] How to optimise the following for loop code?

问题描述

我有一个非常大的数据集，并且正在使用以下代码. 计算花费了太多时间，我想减少迭代次数.

I have a very large dataset and I am using following code. It's taking too much time for computation and I want to reduce number of iterations.

如何提高代码的性能?

import numpy as np

Z=np.asarray([[1,2],
              [3,4],
              [5,6],
              [7,8]])

R=np.asarray([[1,2,3],
              [4,5,6]])

AL=np.asarray([[1,2,3],
               [4,5,6]])

X=np.asarray([[1,2,3],
              [4,5,6],
              [7,8,9],
              [10,11,12]])

N = 4
M = 2
D = 3

result = np.ones([N, D])
for i in range(N):
  for l in range(D):
    temp=[]
    for j in range(M):
      temp.append(Z[i][j]*(R[j][l]+AL[j][l]*X[i][l]))
    result[i][l] = np.sum(temp)   

print(result)

输出为:

array([[ 18.,  36.,  60.],
       [ 95., 156., 231.],
       [232., 360., 510.],
       [429., 648., 897.]])

推荐答案

使用numpy时，更喜欢使用矩阵和数组运算，而不是for迭代.性能大大提高.

When using numpy, prefer using matrix and array operations instead of for iterations. The performance is drastically better.

您的解决方案可以写为:

Your solution can be written as:

result = Z.dot(R) + Z.dot(AL) * X

输出:

array([[ 18.,  36.,  60.],
       [ 95., 156., 231.],
       [232., 360., 510.],
       [429., 648., 897.]])

这篇关于如何优化以下循环代码?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！