协方差的特征值、特征向量的几何意义、和pca降维处理流程、矩阵相乘为什么是投影的解释
协方差的特征值、特征向量的几何意义、和pca降维处理流程、矩阵相乘为什么是投影的解释
正态分布几何表示
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430344582.png)
enter description here
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430357223.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430361369.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430364080.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430368713.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430370985.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430377592.png)
维度意义
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430398787.png)
enter description here
特征值特征向量的二维几何意义表示例子
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430931551.png)
enter description here
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430935473.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430950076.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586430952845.png)
pca降维的应用、降维之后准确率几乎不变但是数据处理量下降了一半
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431267227.png)
enter description here
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431272533.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431276094.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431278666.png)
特征向量是极大线性无关组,表示一个坐标系,特征值是坐标的伸缩系数
矩阵相乘的意义,代表坐标代换也就是投影映射,坐标映射
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431974766.png)
enter description here
![enter description here enter description here](./images/1586431977623.png)
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586431981583.png)
kpca是先升維度,再降维度
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586432327790.png)
enter description here
LDA理解
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586432424103.png)
enter description here
![enter description here enter description here](https://raw.githubusercontent.com/miaozhijuan/xiaoshujiang/master/%E5%B0%8F%E4%B9%A6%E5%8C%A0/1586432435321.png)