7_ Principal component analysis

7_ Principal component analysis （Principal Component Analysis）

Usually the textbook will use a lot of mathematical concepts to explain PCA, In this course , We will find another way , Get around the heavy math concepts , Sister method using gradient descent ： Gradient rise method to solve PCA problem , And then understand PCA The basic principle of , How to use PCA Reduce the dimension of data . We'll also give you more PCA Application scenarios of , It's not just for you to practice PCA The power of dimension reduction , Let's see PCA Noise reduction , Face recognition, etc …

7-1 What is? PCA

• First step : Classify the average value of the sample as 0(demean)
  
  
  
  
  
  

7-2 Using gradient rise method to solve PCA problem

7-3 Find the main components of the data PCA

7-4 Before data n A principal component

7-5 High dimensional data is mapped to low dimensional data

7-6 scikit-learn Medium PCA

7-7 Try your hand MNIST Data sets

7-8 About MNIST The latest way to obtain data sets

7-9 Use PCA Noise reduction of data

7-10 Face recognition and feature face