Please help me . I need it urgently.

Assignment 3 - How to tell you're reading news Classification It is a widely used ML task, and has many forms. Of course sorting data into categories, but also detection tasks (is there a person in this picture?), tracking (is the same person found in a sequence of images, and if so where?), pattem recognition (did you just say 'Ok Google'?), and so on. Depending on the task, it can be very difficult. LHEN A USER TAKES A PHOTO, THE APP SHOULD CHECK LHETHER THEYRE IN A NATIONAL PARK... Your Task: Download and unpack the starter code. It contains a set You will implement: SURE, EASr GSOKPof functions you need to complete for this assignment. AND CHECK WHETHER THE PHOTO ISOFA BIRD. A Gaussian Class Conditional classifier A Naive Bayes classifier PCA for dimensionality reduction IL NEED A RESEARCH TEAM AND FIVE YEARS. Along the way you will explore issues such as how classification accuracy changes betwe training data and testing data, how the dimensionality of your data affects classification accuracy, and whether or not making assumptions about the independence of features (as we do with the Naive Bayes classifier) Is a reasonable thing to do even when we know features are not actually independent. You will use the same document vectors you clusteres in A2, except this time you have labels for the training and testing sets (these were provided by human IN CS, IT CAN BE HARD To EXPLAIN THE DIFFERENCE BETWEEN THE EASY AND THE VRTUALLY IMPO5SIBLE. Courtesy of xkcd.com: https://kcd.com/1425/ 1) Implement and test the code that does PCA Complete the function that takes an input data set, computes the PCA eigenvectors and the mean of the dataset, and returns a low-dimensional representation of the dataset. I suggest you test this on a simpler dataset than the BBC dataset, for example, feed it the pixels of a colour image (one per row, 3 values per pixel) and make sure you can reconstruct the original data from the PCA representation. Our auto-tester will check your PCA code computes the correct low-dim representation, so make sure your PCA code does the right thing Once you are happy your PCA code works compute a low-dimensional representation of the BBC training dataset (using 3 dimensions), and plot on a 3D graph the low dimensional vectors for the input documents. The colour for each point should be given by the class for the corresponding document choose 5 any colours you like Assignment 3 - How to tell you're reading news Classification It is a widely used ML task, and has many forms. Of course sorting data into categories, but also detection tasks (is there a person in this picture?), tracking (is the same person found in a sequence of images, and if so where?), pattem recognition (did you just say 'Ok Google'?), and so on. Depending on the task, it can be very difficult. LHEN A USER TAKES A PHOTO, THE APP SHOULD CHECK LHETHER THEYRE IN A NATIONAL PARK... Your Task: Download and unpack the starter code. It contains a set You will implement: SURE, EASr GSOKPof functions you need to complete for this assignment. AND CHECK WHETHER THE PHOTO ISOFA BIRD. A Gaussian Class Conditional classifier A Naive Bayes classifier PCA for dimensionality reduction IL NEED A RESEARCH TEAM AND FIVE YEARS. Along the way you will explore issues such as how classification accuracy changes betwe training data and testing data, how the dimensionality of your data affects classification accuracy, and whether or not making assumptions about the independence of features (as we do with the Naive Bayes classifier) Is a reasonable thing to do even when we know features are not actually independent. You will use the same document vectors you clusteres in A2, except this time you have labels for the training and testing sets (these were provided by human IN CS, IT CAN BE HARD To EXPLAIN THE DIFFERENCE BETWEEN THE EASY AND THE VRTUALLY IMPO5SIBLE. Courtesy of xkcd.com: https://kcd.com/1425/ 1) Implement and test the code that does PCA Complete the function that takes an input data set, computes the PCA eigenvectors and the mean of the dataset, and returns a low-dimensional representation of the dataset. I suggest you test this on a simpler dataset than the BBC dataset, for example, feed it the pixels of a colour image (one per row, 3 values per pixel) and make sure you can reconstruct the original data from the PCA representation. Our auto-tester will check your PCA code computes the correct low-dim representation, so make sure your PCA code does the right thing Once you are happy your PCA code works compute a low-dimensional representation of the BBC training dataset (using 3 dimensions), and plot on a 3D graph the low dimensional vectors for the input documents. The colour for each point should be given by the class for the corresponding document choose 5 any colours you like