Projects

Practical work is an essential part of learning. We offer you an opportunity to test your newly acquired machine learning skills on real-life problems. Please read this page carefully before asking questions. For questions, use the #questions-ml26 channel in Slack.

Key dates:

• Team formation (team size is 3–4 students), until September 27
• Project plan, deadline October 11 (2 points out of 25 points)
• Intermediate presentations November 16–18 (5 points out of 25 points)
• Final presentations December 14–16 (15 points out of 25 points)
• Team evaluation, deadline December 20 (3 points out of 25 points)

Choosing a project

The first step is to find a project that is aligned with your interests. Below are several options:

From the partners:

A list of projects proposed by our partners will be made available here before the start of the course: TBA.

When choosing a project from a partner, consider not only the topic and the description of the idea, but also whether the data is readily available and whether the project's complexity is reasonable and aligns well with your expectations. You take full responsibility for working on this project, so be mindful.

When you sign up for a project from a partner, make sure that your team does not exceed the limit on the number of teams that the partner is willing to supervise. This will be indicated in the project list.

If you choose a project from one of the partners, connect with the representative who proposed the project to obtain data and guidance. Note that extra time and care must be invested in communicating with a project owner and meeting their expectations.

Project owners will be invited to the final presentations and given an opportunity to provide feedback that will be considered in the final assessment.

From Kaggle.com:

You can do a project by participating in a suitable Kaggle competition or by working with a Kaggle dataset.

Kaggle competitions and datasets are often accompanied by public notebooks where people document and publish their analysis and solutions. You must declare all existing notebooks, code, or solutions that you use in your project and convince the instructors during the presentations that your own contribution is substantially different from existing solutions and constitutes a sufficient amount of work for a 25-point project.

Come up with a project yourself:

If neither option above is of interest to you, feel free to propose your own project. For example, you may decide to analyze publicly available data from Estonian Open Data, UNICEF, Wikidata, Open Climate Data, or the World Bank.

Make sure there is a clear machine learning component in your project. You can consult with the instructors if in doubt.

If you already have a team, describe your project and team in the project document: TBA.

If you do not have a team, you can advertise your project in the #introductions channel in Slack and try to attract team members.

Team formation (deadline: September 27)

After you have chosen a project or come up with one yourself, you need to assemble a team or join an existing one. You can either advertise your project in the #introductions channel or comment on someone else's post, letting them know that you would like to join.

Remember: team size must be 3–4 people.

If someone has already formed a team, get their approval before adding your name to the existing team in the project document.

By September 27 at 23:59, each team must register its project in the project document: TBA.

You will need to specify:

• project name;
• project type (partner project, Kaggle, or your own);
• project description;
• team members;
• preferred presentation day (Monday, Tuesday, or Wednesday).

This is an obligatory step for continuing with the project.

If you plan to work on a partner project, also specify the corresponding project identifier from the partner-project list.

Your choice of presentation day will determine which practice-session instructors will grade your intermediate and final presentations.

Project plan: October 11 (2 points out of 25 points)

Fill in and submit the project plan using the provided project-plan template: TBA.

We recommend taking enough time to understand the data and the problem domain before writing the plan, so make sure to allocate some time for preliminary research.

We also strongly recommend following basic principles of team and project management. For example:

• hold regular weekly meetings;
• divide responsibilities clearly;
• keep track of what has been completed;
• agree on concrete tasks for the following week.

These activities take time and energy, but they help keep the project on track.

We will also provide a simple project-management spreadsheet that you can copy and modify for your team: TBA.

Using this spreadsheet is not graded, and you are free to use another project-management tool instead.

Intermediate presentation: November 16–18 (5 points out of 25 points)

The order of presentations will be published later.

General guidelines for intermediate presentations:

• All team members should attend the presentation. If one person is presenting, the rest of the team should participate in the Q&A. There is also a possibility to present online.
• Make sure your presentation fits into 5 minutes. Going over time may result in a penalty.
• Add your Google Slides to the corresponding presentation folder. Links for Monday, Tuesday, and Wednesday will be provided here later: TBA.
• Introduce your team and project owner, if applicable.
• Briefly describe the problem you are trying to solve and explain why it is worth solving.
• Describe the progress you have made so far.
• Explain any blockers or problems you have encountered.
• Specify which team member is responsible for which part of the work.
• Describe your next steps and what you plan to accomplish before the final presentation.

Intermediate presentations are compulsory in order to proceed to the final presentations.

Final presentation: December 14–16 (15 points out of 25 points)

• All team members should attend the presentation. If one person is presenting, the rest of the team should participate in the Q&A. There is also a possibility to present online.
• As with the intermediate presentations, the final presentations will be held in person, with Zoom participation possible when needed, and at least two instructors will grade each presentation.
• You have up to 5 minutes for the final presentation.
• Add your Google Slides to the corresponding presentation folder. Links for Monday, Tuesday, and Wednesday will be provided here later: TBA.
• Introduce your team and project owner, if applicable.
• Briefly remind us of the problem you are trying to solve and why it is worth solving.
• Explain your approach to the problem and the methods you used.
• Present the results you obtained and discuss how they compare with your initial expectations.
• Specify which team member was responsible for which part of the work.
• Describe a few lessons you learned while working on the project.
• Add a link to your repository to your presentation. If you have a private repository, give access to the course instructors.

The order of teams and the detailed schedule will be published later.

Team evaluation: deadline December 20 (3 points out of 25 points)

After you have finished the project and completed the final presentation, you need to fill in a questionnaire about each of your teammates separately.

The questionnaire will be made available here before the final presentations: TBA.

The questionnaire will ask about each teammate's contribution, including whether the person completed the work they agreed to do, attended meetings, and contributed to the project.

Your team-evaluation score will be calculated based on the evaluations provided by your teammates, so it may differ between members of the same team.

Grading

Your project grade consists of:

• project plan: 2 points;
• intermediate presentation: 5 points;
• final presentation: 15 points;
• team evaluation: 3 points.

The presentations will be graded based roughly on the following criteria:

• amount and complexity of work performed (40%);
• quality of the presentation (30%);
• degree to which you completed the initial task (25%);
• staying within the allocated presentation time (5%).

Project-plan and presentation points are assigned equally to all team members unless there is a good reason to believe that some team members contributed significantly less or more than others.

Both intermediate and final presentations are normally held during practice sessions. Each project will be graded independently by at least two instructors, and the final grade will be their average.

There is no written report for the project. The only things we will ask you to submit at the end are a link to your project repository and your presentation.

Short summary of what you have to do:

1. Choose a project: a) from a partner, b) from Kaggle, or c) come up with your own project idea.
2. By September 27, form a team of 3–4 people using the #introductions Slack channel or your contacts in the course, and register your project and team in the project document.
3. By October 11, submit your project plan.
4. Deliver an intermediate presentation during November 16–18 in your chosen practice session (Monday, Tuesday, or Wednesday).
5. Deliver a final presentation during December 14–16 in your chosen practice session.
6. Fill out the team-evaluation questionnaire by December 20.