Thank you sir. Brilliant methods and explanations to help a near beginner trying to grasp the nuance of eigenvalues and eigenvectors as applied in PCA and unsupervised machine learning. I'm not there yet, but I sure feel like I can and will after listening to you. Who needs machines ... lol?
Thank you verymuch It covers 1. column std. matrix 2.Co variance matrix 3.eigen vector 4.eigen value 5.Orthogonal matrix of v1, v2, .... vd 6.projection 7.projection 100%, 75%,60%,50% In general, for any matrix, the eigenvectors are NOT always orthogonal. But for a special type of matrix, symmetric matrix, the eigenvalues are always real and the corresponding eigenvectors are always orthogonal.
Applied AI Course In 1:57 covariance matrix says Cov of X(S)= Transpose(X).X But in previous vedio we saw Cov of X(S)=1/n( Transpose(X).X ) ..Why this difference??
The goal of PCA is to find projection directions along which the variance of the projected data points is maximum. If you would write this goal in mathematical terms, and simplify the resulting objective, you would find a matrix in the resulting eigenvalue equation. This matrix is nothing but the covariance matrix. Hence, the covariance matrix naturally arises from the goal/objective of PCA. Please go through the link: www.visiondummy.com/2014/04/geometric-interpretation-covariance-matrix/
Yes, these lectures are in a sequence. You need to first watch the videos for linear algebra and then watch PCA for a clear understanding. You can watch these videos in order on our website as these are our sample/free videos. Please login to AppliedAICourse.com and check out the sample videos which are in clean order starting from basics of Python
