PCA on Olivetti dataset Load the Olivetti face images dataset from sklearn. (sklearn.datasets.fetch_olivetti_faces). Vectorize the images and apply PCA. Plot the eigenvalues. Display the mean image and first 20 principal components. Represent a face image using different numbers of principal components (say, 5, 10, 40, 200) and compare the representations with the original image. Use scatter plot to represent all images on a 3D space using the first 3 principal components