Assignment 4 is available on the CS website

Assignment 3 is available. Please note the different deadlines.
Regards
The STATS team

Please type all of your answers in a single document. The formatting is up to you, but you need to make sure that all answers are clearly numbered.

You must include a signed cover page (statement of originality) as the first page of your document.

The completed exam paper (i.e. report with all the answers) has to be submitted to ClickUP before 7am tomorrow morning. Answer to question 9 in the form of an Excel spreadsheet also has to be submitted on ClickUP.

Good luck!

Final semester marks have been posted.

The exam will take place tomorrow, on 5 June. The exam is a take-home event. The question paper will be released on the CS website (this page) tomorrow at 7:00am, and will be due the next day at 7:00am. You will have exactly 24 hours to complete the question paper. The questions will be theoretical as well as practial. All exam reports as well as the solutions to practical questions will have to be submitted via ClickUP.

If you do not meet the minimum exam entrace requirement (at least 40% semester mark), your exam paper will not be marked.

Updated semester marks have been published. If you have queried your assignment marks, please check whether the updated semester mark reflects the changes. Assignment 3 queries should be addressed to Mr Sollie Millard.

The semester marks are available under Admin/Marks. Note that if your semester mark is below 40% (marked in red), then you have failed the subject, since you do not qualify to write the exam.

The semester mark was calculated as follows:
Assignment 1 - 25%
Assignment 2 - 25%
Assignment 3 - 50%

Refer assignment 1 and 2 queries to Mrs Anna Bosman, refer assignment 3 queries to Mr Sollie Millard.

Assignment 3 marks have been released. All queries should be addressed to Mr Sollie Millard.

Assignment 2 marks have been released and are available from this page, under Admin/Marks. Note that only the final mark is specified. For more detailed feedback, please contact Ms Anna Bosman.

Final marksheet with semester marks will be released later this week.

Please note that Q3 refers to the sheet MitD2. This has also been corrected on the assignment itself.

Preparation for next session. Prepare k-means clustering in such a way that you have a working knowledge of the technique.

Assignment 3 report can be submitted via ClickUp or the CS website.
All code must be submitted via the CS website. 
Printed copies are also accepted in class tomorrow.

Assignment marks have been released and are available from this page, under Admin/Marks. Note that only the final mark is specified. For more detailed feedback, please contact Ms Anna Bosman.

Some important aspects that might not be clear
1. Two deadlines, submit your documents at the latest on the respective dates - in the specified format.
2. Use the same format as for your first assignments, structure and referencing
3. Any additional strategies/techniques are welcome
4. Two report, one for each of the deadlines

The students are encouraged to attend the following event:

http://www.deeplearningindaba.com/

Deep Learning Indaba is a local event that aims to strengthen machine learning in Africa, with a focus on deep learning. Application is open to all, including students, post-doctoral candidates, research staff, industries, startups: to people from across Africa and the world. If you are interested, fill in the application form here:

http://www.deeplearningindaba.com/apply.html

Assignment 2 is available for download. The submission deadline is 16 April, 23:30. No late submissions will be accepted.

To better understand the material discussed in class, as well as to see R examples of applying the discussed techniques to real datasets, the students are advised to study the following chapters in the prescribed textbook:

Chapter 7: Black Box Methods – Neural Networks and Support Vector Machines
Chapter 11: Understanding Ensembles, Bagging, Boosting, Random Forest sections

In addition, the students are referred to the extra material published on the course page. Extra material provides extensive reading on SOMs. To see how some of the examples shown in class were generated in R, refer to:
https://www.r-bloggers.com/self-organising-maps-for-customer-segmentation-using-r/

Assignment 2 will be released shortly.

Note that all late assignment submissions e-mailed to the lecturer will be marked, but the maximum mark that you can expect to receive for a late submission will be 50%. You are given approximately 1 month to complete each assignment, thus there is no excuse for submitting at the last moment.

NB: Any late submissions for future assignments will NOT be marked at all.

Second lecture on machine learning is taking place today, 15:30 – 17:30, GW/HSB 1-15.

Please note that the deadline for Assignment 1 is 11 March 2018 at 23:30. The specifications were corrected accordingly.

The first assignment is available for download.

To better understand the material discussed in class, as well as to see R examples of applying the discussed techniques to real datasets, the students are advised to study the following chapters in the prescribed textbook:

Chapter 3: Lazy Learning – Classification Using Nearest Neighbors
Chapter 5: Divide and Conquer – Classification Using Decision Trees and Rules
Chapter 10: Evaluating Model Performance

The first assignment will be released later today.

The MIT 801 study guide has been released. All assignment release and submission dates are included. You are advised to use this information for time management purposes.

Please note that the same venue will be used for all MIT classes: GW/HSB 1-15 (Human Sciences building, Hatfield campus)

The students are advised to obtain a free copy of "Machine Learning with R" textbook by B. Lantz here: https://www.packtpub.com/packt/free-ebook/r-machine-learning

In preparation for the upcoming lecture, please study chapters 1 and 2. These chapters introduce the concept of machine learning, as well as cover basic data manipulation in R.

To work with R, RStudio IDE is highly recommended: https://www.rstudio.com/products/RStudio/

Dear students,

Welcome to MIT801! This course will teach you the basics of machine and statistical learning. There will be 4 2-hour lectures, first one taking place on 16 February, and 4 practical assignments. Practical assignments will have to be completed in R. R will not be taught in class, but helpful self-study links will be provided. The assignments will make up your semester mark. Students with a semester mark of less than 40% will not be allowed to write the examination. The examination will be a 24-hour take-home question paper. Working students are advised to budget for the exam beforehand, and plan to take a day of leave in June. The exam date will be posted as soon as it becomes available.

A detailed study guide will be released in due course.