一、基础索引和切片

import  numpy as np

a = np.array([[0,1,2,3,4],
              [5,6,7,8,9],
              [10,11,12,13,14],
              [15,16,17,18,19]])
print(a)
print(a[0,0])#0行0列
print(a[2])#2行所有列
print(a[0:2,2:4])#取0、1行，2、3列

二、布尔索引

import  numpy as np

a = np.array([[0,1,2,3,4],
              [5,6,7,8,9],
              [10,11,12,13,14],
              [15,16,17,18,19]])
# 设定筛选规则
b = a>10

print(b)
print('*'*30)
print(a[b])

输出结果：

import  numpy as np

a = np.array([[0,1,2,3,4],
              [5,6,7,8,9],
              [10,11,12,13,14],
              [15,16,17,18,19]])
# 设定筛选规则
b = a[:,3]>5
#打印规则，结果为布尔类型
print(b)
print('*'*30)
# 符合筛选规则的数据加30，注意先定位数据（a[:,3]）再执行筛选规则（a[:,3][b]）
a[:,3][b]+=30
print(a[:,3][b])

三、神奇索引

import  numpy as np

a = np.array([[0,1,2,3,4],
              [5,6,7,8,9],
              [10,11,12,13,14],
              [15,16,17,18,19]])
# 打印第二、三行
print(a[[2,3]])
print('#'*30)
# 打印第[2,3]、[3,4]个元素
print(a[[2,3],[3,4]])