In this assignment, students need to develop and evaluate various regression models to solve the real- world problem chosen in Assignment 1. They should also communicate the results by writing a short report.
Dataset uploading, and Data preparation
Students need to write a module to read/upload the dataset(s) and save them in appropriate variables and data structures. If it is required appropriate data preparation/pre-processing should also be done. Students will use the datasets they have selected in Assignment 1. The data to be use must have one output variable and one input variable.
Machine learning models
Students will create multiple regression models using linear and polynomial regression. Students need to use the machine learning techniques taught (and beyond is possible) in Modules 1, 2, and 3.
Model Generalisation and assessment
Students need to apply appropriate generalisation and assement methods on the created models. Students must split data into train, test. Validation sets and create multiple models and assess them. Modules 2 and 3 cover these techniques.
Visualisation and demonstration of data and results
Appropriate visualisation methods should be developed to present data and results.
Comments in the source code
The source code should have appropriate comments within the code to make sure the code is readable.
Report - Explanations, Format, Referencing
A short report consisting of at least 500 words should be written to present the solutions and also to communicate the results.
You are required to submit two files. The two files must exactly conform to the requirements below otherwise, you will lose grade:
Your files should be submitted exactly as given below otherwise, you will lose grade:
This assignment goes through the process we cover from weeks 1 - 3. If you have been putting work in, we expect that this problem will be interesting and a little challenging for you but not very difficult. What we are looking for is that you know how to tackle such a problem and know the process through which to approach the solution. You should develop your solution in a thoughtful manner and identify the interesting features of the problem. Working solutions which do not give insight into how they were arrived at will not be marked favourably. Moreover, banal/’professional’ and jargon-laden writing which does not convey much meaning or insight and mostly conveys straightforward information/talking points will also not be judged favourably.
The reason we have a strict requirement of what kind of files we want is that we all want them in the same form so we can mark them easily. Moreover, in the real world, you are usually given specific details on how the desired result will look like and if it doesn’t look like that, there can be huge issues.
Just imagine a contractor builds a powerplant in your country, but the output voltage and current of
the powerplant are not compatible with your country’s voltage and current. It would be a disaster!
You can go beyond what we have taught you (in fact we encourage it!!) but you should state why and how you did that.
A rubric has been provided to help students with marking. Moreover, a description of the rubric is also provided in the following.
An interview of 10-15 minutes may be conducted for each student to answer some questions about the model and explain their code. If you cannot answer questions about your solutions, it will inform us that you did not solve the problem yourself and we will adjust your grade accordingly.
If you meet the minimal requirements for submission above, you can achieve the following grades:
Pass: To achieve a pass mark you must show a basic understanding of the process of solving the given problem and have a basic working solution that gives a reasonable error.
Credit: To achieve a credit grade you must show a good understanding of the process and have a working solution with a reasonable error. Moreover, you can partially identify the interesting features of the problem, and some difficulties you faced and how to tackle them.
Distinction: To achieve a distinction you must show an excellent understanding of the process and have a well-implemented working solution with low error. Furthermore, you can identify the majority of the interesting features of the problem, the implementation issues the requirements pose and how you solved all these issues.
High Distinction: Everything in distinction but at an outstanding level and even going beyond. This means you will demonstrate a deep understanding of the problem and the process employed to solve the problem.