#numpy中的向量说明，防止一些奇怪的bug出现
import numpy as np
a=np.random.randn(5)

In [2]:

Out[2]:

array([ 1.46701856,  0.63680735,  0.23052999, -0.34198406, -0.16955206])

In [3]:

a.T

Out[3]:

array([ 1.46701856,  0.63680735,  0.23052999, -0.34198406, -0.16955206])

In [4]:

np.dot(a,a.T)

Out[4]:

2.7565121408210516

In [6]:

#用这种有两层括号的结构
a=np.random.randn(5,1)
a

Out[6]:

array([[ 0.87133017],
       [-0.15832112],
       [-0.61651606],
       [ 1.73090703],
       [ 0.29974957]])

In [7]:

a.T

Out[7]:

array([[ 0.87133017, -0.15832112, -0.61651606,  1.73090703,  0.29974957]])

In [8]:

print(np.dot(a,a.T))

[[ 0.75921626 -0.13794997 -0.53718904  1.50819151  0.26118084]
 [-0.13794997  0.02506558  0.09760751 -0.27403914 -0.04745669]
 [-0.53718904  0.09760751  0.38009206 -1.06713199 -0.18480042]
 [ 1.50819151 -0.27403914 -1.06713199  2.99603916  0.51883863]
 [ 0.26118084 -0.04745669 -0.18480042  0.51883863  0.0898498 ]]

In [14]:

#不要用rank 1 array
a=np.random.randn(5)

In [13]:

a=np.random.randn(5,1)
a

Out[13]:

array([[-0.89702644],
       [-1.28384216],
       [-0.15319171],
       [ 1.20351335],
       [ 0.13005488]])

----------------------

Logistic Regression Cost Function的解释

DL study5 numpy的小技巧补充，logistic regression cost function的解释

DL study5 numpy的小技巧补充，logistic regression cost function的解释

相关推荐