Description

Assessment weighting:
This assignment is worth 15% of the total marks for this module.
Submission instructions:
Your code should be neatly laid out and formatted and commented appropriately. Submitted code samples (including comments) should adhere to the conventions in the PEP 8 Style Guide for Python Code (https://www.python.org/dev/peps/pep-0008/).
Your code should be submitted to the link marked ‘Submit Assignment 1 Code’ provided on Blackboard in a
single .zip folder (NOT in a .rar, tar.gz, .7z etc.), with the naming convention CT4101_A1_lastname_firstname_code.zip. Code submissions can be in the form of a standard Python module or modules (file extension .py), or in the form of a Jupyter Lab notebook (file extension .ipynb).
Assignment description (25 marks max.)
The goal of this assignment is to learn the basics of using the scikit-learn package to develop machine learning (ML) models for a classification task. To complete this assignment you must write a Python application (extension .py) or a Jupyter notebook (extension .ipynb) that trains two ML models for classification and explores the impact of different hyperparameter values on the accuracy achieved by your models. You must also prepare a short .pdf report (5 pages max.) discussing your findings.
A list of available scikit-learn classifier implementations can be found at: https://scikit-learn.org/stable/auto_examples/classification/plot_classifier_comparison.html
The required sections in the report (5 pages max.) are described below:
1. Description of algorithms (2 x 5 marks): Clearly describe each of your chosen scikit-learn algorithm implementations in turn, paying special attention to discussing the two hyperparameters that you have chosen to tune for each algorithm. You should write a maximum of 1 page per algorithm.
2. Model training and evaluation (2 x 5 marks): For each of your chosen algorithms, you should discuss the results achieved with the default settings, and also discuss on the results you achieved after trying out a selection of different values for the two selected hyperparameters. You should summarise the accuracy results on the training and test data in an appropriate format (e.g., in graphs or tables). You should write a maximum of 1 page per algorithm.
3. Conclusion (5 marks): Briefly sum up your key findings, including e.g., which model performed best, whether the achieved accuracy of one of the models is more sensitive to hyperparameter values than the other, and your recommended hyperparameter values for each algorithm based on your findings. You should write a maximum of 1 page for this section.
For each of the sections above, marks out of 5 will be awarded based on the following scale:
5 – exceptional, 4 – very good, 3 – average, 2 – passable, 1 – incomplete, 0 – section missing