【CS231n学习笔记】2. python numpy 之numpy

Numpy

数组的创建

import numpy as np

a = np.full((3, 3), 1)
print(a)

a = np.random.random((3, 3))
print(a)

a = np.eye(3)
print(a)

a = np.array([[1, 2, 3, 4],
              [5, 6, 7, 8],
              [9, 10, 11, 12],
              [13, 14, 15, 16]])
print(a)
print(a.shape)

输出：
[[1 1 1]
 [1 1 1]
 [1 1 1]]
[[ 0.09670856  0.44868154  0.43326738]
 [ 0.57400445  0.47124464  0.76310375]
 [ 0.72557452  0.98591433  0.97147127]]
[[ 1.  0.  0.]
 [ 0.  1.  0.]
 [ 0.  0.  1.]]
[[ 1  2  3  4]
 [ 5  6  7  8]
 [ 9 10 11 12]
 [13 14 15 16]]
(4, 4)

数组的访问方法

import numpy as np

a = np.array([[1, 2, 3, 4],
              [5, 6, 7, 8],
              [9, 10, 11, 12],
              [13, 14, 15, 16]])
print(a)
print(a.shape)
print(a[1:3])
print(a[1:-1, 1:-1])
print(a[0, 1])
print(a[1:3, 2])
print(a[2, 1:3])

print(a[[0, 1, 3, 3], [2, 3, 2, 2]])  # print a[0,2],a[1,3],a[3,2],a[3,2]

输出：
[[ 1  2  3  4]
 [ 5  6  7  8]
 [ 9 10 11 12]
 [13 14 15 16]]
(4, 4)
[[ 5  6  7  8]
 [ 9 10 11 12]]
[[ 6  7]
 [10 11]]
2
[ 7 11]
[10 11]
[ 3  8 15 15]

蜜汁用法

import numpy as np

a = np.array([[1, 2, 3, 4],
              [5, 6, 7, 8],
              [9, 10, 11, 12],
              [13, 14, 15, 16]])
print(np.arange(4))
print(np.full([1, 4], 1))
print(a[np.arange(4), 1])
a[np.arange(4), [2, 3, 2, 3]] += 100
print(a)

[0 1 2 3]
[[1 1 1 1]]
[ 2  6 10 14]
[[  1   2 103   4]
 [  5   6   7 108]
 [  9  10 111  12]
 [ 13  14  15 116]]

布尔

import numpy as np

a = np.array([[1, 2, 3, 4],
              [5, 6, 7, 8],
              [9, 10, 11, 12],
              [13, 14, 15, 16]])

b = a > 5  # 还有这种操作？？？
print(b)

print(a[a > 6])

[[False False False False]
 [False  True  True  True]
 [ True  True  True  True]
 [ True  True  True  True]]
[ 7  8  9 10 11 12 13 14 15 16]

数组计算

import numpy as np

a = np.array([1, 2])
b = np.array([3, 4])
print(a + b)
print(a - b)
print(a * b)
print(a / b)
print(a * 2)
print(a + 3)
print(a ** 0.5)

[4 6]
[-2 -2]
[3 8]
[ 0.33333333  0.5       ]
[2 4]
[4 5]
[ 1.          1.41421356]

矩阵乘法&转置

import numpy as np

a = np.array([1, 2])
b = np.array([3, 4])
print(a.dot(b))  # 相当于自动把b竖起来，相当于两个向量内积

a = np.array([[1, 2, 3],
              [4, 5, 6]])
b = np.array([[1, 2, 3],
              [4, 5, 6]])
print(b.T)  # 转置
print(a.dot(b.T))  # 矩阵乘法

11
[[1 4]
 [2 5]
 [3 6]]
[[14 32]
 [32 77]]

求和

import numpy as np

a = np.array([[1, 2, 3],
              [4, 5, 6]])
print(a.sum())  # 求和

各种函数 http://link.zhihu.com/?target=http%3A//docs.scipy.org/doc/numpy/reference/routines.array-manipulation.html

广播

秩不同的矩阵能一起运算

import numpy as np

a = np.array([[1, 2, 3], [1, 2, 3], [1, 2, 3]])
b = np.array([1, 1, 0])
print(a + b)

v = np.array([1, 2, 3])
w = np.array([4, 5])
v.reshape([3, 1])
print(v.reshape(3, 1) + w)
print(w + v.reshape(3, 1))

[[2 3 3]
 [2 3 3]
 [2 3 3]]
[[5 6]
 [6 7]
 [7 8]]
[[5 6]
 [6 7]
 [7 8]]