回答此问题可获得 20 贡献值,回答如果被采纳可获得 50 分。
<p>我在wikipedia上读到python是一种实现行主顺序的语言。但是我试图用两个矩阵相乘,我使用row major方法和column major方法访问。但是当我执行代码时,column major方法总是更快。
我将在这里发布我的代码片段:</p>
<pre><code>class Matrix:
#flag indicates the type of matrix
def __init__(self, num_lines, num_cols, flag):
self.num_lines = num_lines
self.num_cols = num_cols
if flag == 0:
# First Matrix
self.matrix = [1]*num_lines*num_cols
elif flag == 1:
# Second Matrix
self.matrix = [0]*num_lines*num_cols
for i in range(num_lines):
for j in range(num_cols):
self.matrix[i*num_cols+j] = i+1
elif flag == 2:
# Result Matrix
self.matrix = self.matrix = [0]*num_lines*num_cols
def setMatrixValue(self, line, column, value):
self.matrix[line*self.num_cols + column] = value
def getMatrixValue(self, line, column):
return self.matrix[line*self.num_cols + column]
def multiplyMatrices(num_lines, num_cols, flag):
matrix_a = Matrix(num_lines, num_cols, 0)
matrix_b = Matrix(num_cols, num_lines, 1)
matrix_result = Matrix(num_lines, num_lines, 2)
# Column-major approach
if flag == 0:
start_time = time.time()
for i in range(matrix_result.num_lines):
for j in range(matrix_result.num_cols):
temp = 0
for k in range(matrix_a.num_cols):
temp += matrix_a.getMatrixValue(i, k) * matrix_b.getMatrixValue(k, j)
matrix_result.setMatrixValue(i,j,temp)
# Row-major approach
elif flag == 1:
start_time = time.time()
for i in range(matrix_result.num_lines):
for k in range(matrix_result.num_cols):
for j in range(matrix_a.num_lines):
matrix_result.setMatrixValue(i,j, matrix_result.getMatrixValue(i,j) + (matrix_a.getMatrixValue(i,k) * matrix_b.getMatrixValue(k,i)))
end_time = time.time()
print matrix_result.matrix
diffTime(start_time, end_time)
</code></pre>
<> P>我意识到Python中的矩阵乘法比java或C++慢得多。有什么原因吗?在</p>