有人知道如何在Python中从多维数组中提取列吗?
当前回答
>>> x = arange(20).reshape(4,5)
>>> x array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
如果你想要第二列,你可以用
>>> x[:, 1]
array([ 1, 6, 11, 16])
其他回答
>>> import numpy as np
>>> A = np.array([[1,2,3,4],[5,6,7,8]])
>>> A
array([[1, 2, 3, 4],
[5, 6, 7, 8]])
>>> A[:,2] # returns the third columm
array([3, 7])
参见:"numpy。“Arange”和“重塑”来分配内存
示例:(用矩阵(3x4)的形状分配数组)
nrows = 3
ncols = 4
my_array = numpy.arange(nrows*ncols, dtype='double')
my_array = my_array.reshape(nrows, ncols)
如果你有一个数组
a = [[1, 2], [2, 3], [3, 4]]
然后像这样提取第一列:
[row[0] for row in a]
结果是这样的:
[1, 2, 3]
只要使用转置(),就可以像求行一样简单地求列
matrix=np.array(originalMatrix).transpose()
print matrix[NumberOfColumns]
>>> x = arange(20).reshape(4,5)
>>> x array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
如果你想要第二列,你可以用
>>> x[:, 1]
array([ 1, 6, 11, 16])
def get_col(arr, col):
return map(lambda x : x[col], arr)
a = [[1,2,3,4], [5,6,7,8], [9,10,11,12],[13,14,15,16]]
print get_col(a, 3)
Python中的map函数是另一种方法。