Create a Python Notebook with your response to the following questions. Ensure that your work is presented neatly.Question 1: (50 pts)Create a SVM model (either classification or regression) to...

Create a Python Notebook with your response to the following questions. Ensure that your work is presented neatly.

Question 1: (50 pts)

Create a SVM model (either classification or regression) to predict a target using a dataset of your choice. You can use: 1) a previous dataset that you are familiar with from this course, 2) a new dataset, or 3) one of the builtin datasets in sklearn. You may not use the same dataset that was covered in the lecture.

1. (5 pts) Perform any necessary data preparation steps and EDA.




2. (10 pts) Normalize the numeric features and encode any categorical features. Ensure that you use sklearn to normalize/encode the training data, after which transform the test set (there is an example in the lecture).
   
   
   
   




3. (5 pts) Partition the data into train/test sets.




4. (20 pts) Build the model that predicts the target variable, and use GridSearchCV to tune at least two hyperparameters, and perform cross validation. You'll need to read the documentation of the SVM algorithm to identify the hyperparameters.
   
   
   (Links to an external site.)
   
   
   
   
   
   




5. (5 pts) Display: 1) the best model/parameters and 2) the best score that was obtained.




6. (5 pts) How do you feel about the model? Did it accurately predict the target? Evaluate the scores from GridSearchCV for the training set and the test set to diagnose any bias-variance problems.

Examples of datasets that you can use:

• Pima Indians Diabetes Database




• Iris Dataset




• Boston Housing dataset




• Red Wine Quality




• ...etc
  Note: Keep the dataset small. The focus of this assignment is to experiment with tuning a model.