Principal Component Analysis

week 8 延了 一週 因為上禮拜的 machine learning 課程太過瘋狂 XDDD
好~ GOGO!

Principal Component Analysis (PCA): problem formulation

PCA: problem formulation

• Reduce from 2-dimension to 1-dimension: Find a direction (a vector) onto which to project the data so as to minimize the projection error.

• Reduce from n-dimension to k-dimension: Find k vectors onto which to project the data, so as to minimize the projection error.

不得不說實話，還是大神教的好 ! 真的棒啊

PCA is not linear regression

Principal Component Analysis (PCA): Algorithm

Data preprocessing

Applying PCA:Reconstruction from Compressed Representation

Reconstuction:

Applying PCA:Choosing the Number of Principal Components

Choosing k

variance 還是會保持住!!!

大神自己用octave 的時候 也直接用 svd funcs 來做 XDD

Advice for Applying PCA

大神教到這邊的時候，我突然想到一個問題，在 python sklearn 的套件下，我們要給 n_components 的數值，疑問就來了~ 那要如何決定要幾個?
速度查，看到這篇 好棒棒的文章

Supervised learning speedup

用 PCA 來做 數據壓縮!!! (我看 書 上寫的名詞 XDDD)

結果無意間發現這 超級超級棒棒的資源, 資源源源

Bad use of PCA: To prevent overfitting

PCA is sometimes used where it shouldn’t be

Quiz