数据分析和科学计算可视化(6)

当前位置:
首页 > temp > 简明python教程 >

数据分析和科学计算可视化(6)
			
									矩阵不同维度上的计算

									>>> x = np.matrix(np.arange(0,10).reshape(2,5))  # 二维矩阵

									>>> x

									matrix([[0, 1, 2, 3, 4],

									        [5, 6, 7, 8, 9]])

									>>> x.sum()                                      # 所有元素之和

									45

									>>> x.sum(axis=0)                                # 纵向求和

									matrix([[ 5,  7,  9, 11, 13]])

									>>> x.sum(axis=1)                                # 横向求和

									matrix([[10],

									        [35]])

									>>> x.mean()                                     # 平均值

									4.5

									>>> x.mean(axis=1)

									matrix([[ 2.],

									        [ 7.]])

									>>> x.mean(axis=0)

									matrix([[ 2.5,  3.5,  4.5,  5.5,  6.5]])

									>>> x.max()                                # 所有元素最大值

									9

									>>> x.max(axis=0)                          # 纵向最大值

									matrix([[5, 6, 7, 8, 9]])

									>>> x.max(axis=1)                          # 横向最大值

									matrix([[4],

									        [9]])

									>>> weight = [0.3, 0.7]                    # 权重

									>>> np.average(x, axis=0, weights=weight)

									matrix([[ 3.5,  4.5,  5.5,  6.5,  7.5]])

									>>> x = np.matrix(np.random.randint(0, 10, size=(3,3)))

									>>> x

									matrix([[3, 7, 4],

									        [5, 1, 8],

									        [2, 7, 0]])

									>>> x.std()                         # 标准差

									2.6851213274654606

									>>> x.std(axis=1)                   # 横向标准差

									matrix([[ 1.69967317],

									        [ 2.86744176],

									        [ 2.94392029]])

									>>> x.std(axis=0)                   # 纵向标准差

									matrix([[ 1.24721913,  2.82842712,  3.26598632]])

									>>> x.var(axis=0)                   # 纵向方差

									matrix([[  1.55555556,   8.        ,  10.66666667]])
栏目列表