How to code the questions provided

Question 1 (10 points) In the lecture, we used the DrugBank data as a regression example where a few regressors were applied to predict the binding affinity score y given the Morgan Fingerprint vector x of a molecule. Now, given dataset { drugbank_gpx4_x. txt , drugbank_gpx4_y . txt ), you are required to do the following tasks: . Load the dataset { drugbank_gpx4_x. txt , drugbank_gpx4_y . txt } using function np. loadtxt( ) to numpy arrays drugbankgpx4_x and drugbankgpx4_y respecitvely, and then discretize the values in drugbankgpx4_y into Os and is using threshold -6.0. If a value in drugbankgpx4_y is less than or equal to -6.0, replace the value with 1; otherwise 0. (1 mark) . Partition the data into training and test sets. (1 mark) . Apply any two classifers introduced in the lecture to learn on the training set and test on the test set. (4 marks) . Apply any classifier (such as nearest neighbor, decision tree, nearest centroid, naive Bayes, etc.), where was not covered in the lecture but available in scikit-learn (refer to https://scikit-learn.org/stable/supervised_learning.html and https://scikit-learn.org/stable/modules/classes.html), to your training set (for learning the model) and test set (for test performance). (2 marks) Furthermore, use at least 50 words and at most 100 words to introduce how this classifier works in theory. (1 mark) (To address this subtask, you need to google this classifier, do some reading, and understand this model you choose to use.) . Generate a table using markdown language to compare the test performances of these three classifiers you used above, and highlight the classifier that performs the best in terms of test accuracy. (1 mark)