Neural Networks and Deep Learning week2 Neural Network Basics

最新推荐文章于 2024-09-17 02:13:06 发布

肆思

最新推荐文章于 2024-09-17 02:13:06 发布

阅读量112

点赞数 1

分类专栏： coursera-deeplearning.ai 文章标签：机器学习深度学习

本文链接：https://blog.csdn.net/qq_40488600/article/details/110085105

版权

coursera-deeplearning.ai 专栏收录该内容

10 篇文章 0 订阅

订阅专栏

看别人见解违法coursera荣誉，看懂和做对是两码事
What does a neuron compute?

A neuron computes a linear function (z = Wx + b) followed by an activation function
A neuron computes the mean of all features before applying the output to an activation function
A neuron computes an activation function followed by a linear function (z = Wx + b)
A neuron computes a function g that scales the input x linearly (Wx + b)

神经元做什么，前半部分线性回归，后半部分非线性化变换（激活函数）

Which of these is the "Logistic Loss"?

希望你还记得熵， Y=-ylogy-(1-y)log(1-y)

Suppose img is a (32,32,3) array, representing a 32x32 image with 3 color channels red, green and blue. How do you reshape this into a column vector?

x = img.reshape((32*32*3,1))
x = img.reshape((3,32*32))
x = img.reshape((1,32*32,*3))
x = img.reshape((32*32,3))

图片如何变换，我们经常用的是列向量，所以 32*32*3，1，默认列向量各领域通用

Consider the two following random arrays "a" and "b":What will be the shape of "c"？

a = np.random.randn(2, 3) # a.shape = (2, 3)
b = np.random.randn(2, 1) # b.shape = (2, 1)
c = a + b

c.shape = (3, 2)
c.shape = (2, 3)
The computation cannot happen because the sizes don't match. It's going to be "Error"!
c.shape = (2, 1)

python有广播的性质，（2，3）广播性质就是方便矩阵加常数

Consider the two following random arrays "a" and "b":What will be the shape of "c"?

a = np.random.randn(4, 3) # a.shape = (4, 3)
b = np.random.randn(3, 2) # b.shape = (3, 2)
c = a*b

c.shape = (3, 3)
The computation cannot happen because the sizes don't match. It's going to be "Error"!
c.shape = (4, 3)
c.shape = (4,2)

np.dot和 * 是有区别的，一个是矩阵乘法，一个是数乘，这两个大小不匹配，没法点对点乘法

Suppose you have n_xnx input features per example. Recall that $X = [x^{(1)} x^{(2)} ... x^{(m)}]$ What is the dimension of X?

(m,1)

(1,m)

(n_x, m)

(m,n_x)

x是列向量，X写成这样，那就只能是nx（x向量的特征数），m（x向量的个数）

Recall that "np.dot(a,b)" performs a matrix multiplication on a and b, whereas "a*b" performs an element-wise multiplication.

Consider the two following random arrays "a" and "b":What is the shape of c?

a = np.random.randn(12288, 150) # a.shape = (12288, 150)
b = np.random.randn(150, 45) # b.shape = (150, 45)
c = np.dot(a,b)

c.shape = (12288, 150)
The computation cannot happen because the sizes don't match. It's going to be "Error"!
c.shape = (150,150)
c.shape = (12288, 45)

矩阵乘法前行*后列，前面的列数=后面的行数，大小前行数*后列数

Consider the following code snippet:How do you vectorize this?

# a.shape = (3,4)
# b.shape = (4,1)

for i in range(3):
for j in range(4):
c[i][j] = a[i][j] + b[j]

c = a.T + b.T
c = a.T + b
c = a + b.T
c = a + b

有问题建议画图，画图是一个好习惯，可以捋清自己的思路

Consider the following code:What will be c? (If you’re not sure, feel free to run this in python to find out).

a = np.random.randn(3, 3)
b = np.random.randn(3, 1)
c = a*b

This will invoke broadcasting, so b is copied three times to become (3,3), and *∗ is an element-wise product so c.shape will be (3, 3)
This will invoke broadcasting, so b is copied three times to become (3, 3), and *∗ invokes a matrix multiplication operation of two 3x3 matrices so c.shape will be (3, 3)
This will multiply a 3x3 matrix a with a 3x1 vector, thus resulting in a 3x1 vector. That is, c.shape = (3,1).
It will lead to an error since you cannot use “*” to operate on these two matrices. You need to instead use np.dot(a,b)

python 的广播性质

Consider the following computation graph.

第 10 个问题

Consider the following computation graph. What is the output J?