Remember that only PDF submissions are accepted. 1. A dataset D has 64 points. Each point in D is a vector of length 6830, in other words there are 6830 features. If we do Principal Component Analyses (PCA) of this dataset D, how many principal components with non-zero variance would we get? Explain why? 2. (Programming) Read this tech blog (https://sebastianraschka.com/Articles/2014_pca_step-by-step. html) (also uploaded to Canvas) and implement a PCA on your own. Feel free to call any command to compute eigenvector, eigenvalue, covariance matrix, SVD, etc. Just hand-in your code, and you don't have to play with data for now. (But you are encouraged to play with some toy data you find on your own to validate if your implementation is correct or not). You can reuse the code to this question in your mini-project, and play with data then. Remember that only PDF submissions are accepted. 1. A dataset D has 64 points. Each point in D is a vector of length 6830, in other words there are 6830 features. If we do Principal Component Analyses (PCA) of this dataset D, how many principal components with non-zero variance would we get? Explain why? 2. (Programming) Read this tech blog (https://sebastianraschka.com/Articles/2014_pca_step-by-step. html) (also uploaded to Canvas) and implement a PCA on your own. Feel free to call any command to compute eigenvector, eigenvalue, covariance matrix, SVD, etc. Just hand-in your code, and you don't have to play with data for now. (But you are encouraged to play with some toy data you find on your own to validate if your implementation is correct or not). You can reuse the code to this question in your mini-project, and play with data then