The datasets for this demo are the following:

1. The smiley face images we have used in class created by me :)

2. AT&T face dataset available publicly at:

https://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html

Citation:
“Parameterisation of a Stochastic Model for Human Face Identification”, F.S. Samaria ; A.C. Harter,  Proceedings of 2nd IEEE Workshop on Applications of Computer Vision, Sarasota FL, December 1994


3. Mnist digit dataset publicly available at:

http://yann.lecun.com/exdb/mnist/

Citation:
"Gradient-based learning applied to document recognition.”, Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner.  Proceedings of the IEEE, 86(11):2278-2324, November 1998


The python code for the demo are:

1. highdimgauss.ipynb:  This is the code showing that in d dimensions, most points are at distnce sqrt(d) from the mean for standard gaussian distributed points

2. smileyface_KPCA.ipynb:  smiley face dataset using kernel pca

3. kpcaVSisomapVStsne.ipynb: kernel PCA Versus Isomap Versus t-sne on ATT face dataset and Mnist dataset
