- Implementation of PCA from scratch (contains implementation of covariance matrix and implementation of eigenvalues and eigenvectors using power iteration mwthod).
- Prprocessing the mnist dataset making it only zeros and ones to apply the hamming network later on it.
- Tring different number of components in PCA till gets best result.
- Cluster data using k-means (you can use any clustering technique).
- Apply Hamming on unseen data point with PCA and without PCA.
implementing pca not part of hamming network algorithm but we used it to increase the accuarcy .