First Exploratory Data Analysis(EDA) using Karhunen-Louve Transform(KLT). Thereafter KLT Transformation. We'll graph the importance of the eigenvectors. Then feature generation and data visualization at last.
2 eigen vectors are needed to capture 95% of variance
eigen vectors 7 and 6 provide the best accuracy(88.4 %) and hence they can be used to separate the data. We used KNN classifier to find the score.