索引与切片
1.整数索引
- 要获取数组的单个元素,指定元素的索引即可。
import numpy as np
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
print(x[2]) # [21 22 23 24 25]
print(x[3,3]) # 29
print(x[3][3]) # 29
2.切片索引
- 切片操作是指抽取数组的一部分元素生成新数组。对 python 列表进行切片操作得到的数组是原数组的副本,而对 Numpy 数据进行切片操作得到的数组则是指向相同缓冲区的视图。
- 如果想抽取(或查看)数组的一部分,必须使用切片语法,也就是,把几个用冒号( start:stop:step )隔开的数字置于方括号内。
- 省去start,numpy 会认为是0;省去stop,numpy 则会认为是数组的最大索引值;如省去step,它将会被理解为1。
注意切片参数为负数时,即为倒数的索引:
x = np.array([1, 2, 3, 4, 5, 6, 7, 8])
print(x[0:2]) # [1 2]
print(x[1:5:2]) # [2 4]
print(x[2:]) # [3 4 5 6 7 8]
print(x[:2]) # [1 2]
print(x[-2:]) # [7 8]
print(x[:-2]) # [1 2 3 4 5 6]
print(x[:]) # [1 2 3 4 5 6 7 8]
print(x[::-1]) # [8 7 6 5 4 3 2 1]
- 对于二维数组,我们的第一片定义了行的切片,第二片定义了列的切片。
难点是二维数组的切片:
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
print(x[1:3]) # [[16 17 18 19 20]
# [21 22 23 24 25]]
print(x[2, :]) # [21 22 23 24 25]
print(x[:, 2]) # [13 18 23 28 33]
print(x[0, 1:4]) # [12 13 14]
print(x[1:4, 0]) # [16 21 26]
print(x[1:3, 2:4]) # [[18 19]
# [23 24]]
3.dots索引
- NumPy 允许使用 … 表示足够多的冒号来构建完整的索引列表。
x[1,2,…] 等于 x[1,2,:,:,:]
x[…,3] 等于 x[:,:,:,:,3]
x[4,…,5,:] 等于 x[4,:,:,5,:]
4.整数数组索引
- 方括号内传入多个索引值,可以同时选择多个元素。
一维
x = np.array([1, 2, 3, 4, 5, 6, 7, 8])
k = [1,3,5]
print(x[k]) # [2 4 6]
k = [-1,-2,7]
print(x[k]) # [8 7 8]
二维(难点)
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
k = [0, -1]
print(x[k]) # [[11 12 13 14 15]
# [31 32 33 34 35]]
r = [1, 2, 3]
c = [2, 3, 4]
print(x[r,c]) # [18 24 30]
x = np.array([1, 2, 3, 4, 5, 6, 7, 8])
r = np.array([[0, 1], [3, 4]])
print(x[r])
# [[1 2]
# [4 5]]
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
r = np.array([[0, 1], [3, 4]])
print(x[r])
# [[[11 12 13 14 15]
# [16 17 18 19 20]]
#
# [[26 27 28 29 30]
# [31 32 33 34 35]]]
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
print(x[[0, 0, 4, 4], [0, 4, 0, 4]])
#
[11 15 31 35]
r = np.array([[0, 0], [4, 4]])
c = np.array([[0, 4], [0, 4]])
print(x[r, c])
# [[11 15]
# [31 35]]
5.切片 : 与整数数组组合
x = np.array([[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20],
[21, 22, 23, 24, 25],
[26, 27, 28, 29, 30],
[31, 32, 33, 34, 35]])
y = x[1:4, [1, 3, -1]]
print(y)
# [[17 19 20]
# [22 24 25]
# [27 29 30]]
6.布尔索引
x = np.array([1, 2, 3, 4, 5, 6])
y = x > 3
print(y) # [False False False True True True]
print(x[y]) # [4 5 6]
x = np.array([np.nan, 1, 2, np.nan, 3, 4, 5])
y = np.logical_not(np.isnan(x)) # logical_not 逻辑取反
print(y) # [False True True False True True True]
print(x[y]) # [1. 2. 3. 4. 5.]
函数作图
import matplotlib.pyplot as plt
x = np.linspace(0, 2 * np.pi, 50) # 生成0~2pi之间50个点
y = np.sin(x)
# print(y)
plt.show()
mask = (y >= 0)
print(mask)
plt.plot(x[mask], y[mask])
plt.plot(x[mask], y[mask], 'bo')
plt.show()