Course content

This course has the aim of providing the foundations and key concepts of supervised and unsupervised learning with the most popular python libraries such as sklearn and keras. Here, you will learn about regression, classification, and clustering. You will learn how to use perceptron, and multilayer perceptron for solving real world problems. You will learn data preprocessing, feature engineering, feature selection, and feature construction techniques. Throughout the course you will learn learning modes, model selection skills, model evaluation and tuning. You will work with various types of problems and various datasets including images, videos, text, and time series. You will learn the key concepts underlying deep learning and how to use Python to develop machine learning pipelines to tackle real world problems using images, videos, text data and time-series.

Learning outcome

Upon Completion of This Course, You'll Have:

1. An understanding of the capabilities and limitations of machine learning (ML), and the knowledge of how to formulate your problem to solve it effectively.
2. An effective process for developing your machine learning pipeline to tackle real world problems such as machine vision, text understanding and time series prediction.
3. The skills required to deploying, monitoring, and evaluating the ML model, as well as assessing its relevance, and the uses of different ML models.
4. The basis required to collect, process, and utilize data efficiently.
5. The basic skills required to select the right platform to deploy your model (cloud, edge device, hybrid) and how to configure it to achieve the required performance.
6. The ability to document and communicate the results of your ML approach and guide your coding and ML efforts in the right direction.

Learning methods and activities

• The course will be offered in approximately 12 weeks (physically and digitally).
• Teaching approach: 5 hours each (lectures - practice - project work).
• Evaluation: exam will be in the form of a portfolio assessment where samples of work and mini projects will be used to evaluate the intended learning outcomes (ILOs) achievement throughout the course.
• Participants will get a certificate of course completion.

Further on evaluation

Portfolio assessment in terms of project(s) report and presentation. Apart from academic excellence, presentation skills are also important and will be evaluated. Ensure that the work submitted is clearly laid out and has legible figures, drawings, and diagrams. Report layout will be decided during the course. A basic report layout consists of an introduction where you summarize the state-of-the-art in this area and give a brief summary of your work. methodology, results, concluding remarks and a reference list. It is worth noting that it is a good habit of adding references wherever needed. The portfolio contains assignments that are carried out, digitally documented and submitted during the term. Both individual and team assignments may be given. Assignments are designed to help students achieve specific course learning outcomes, and formative feedback is given during the period of the portfolio.

Specific conditions

Recommended previous knowledge

Basic linear algebra and basic Python programming skills.

Course materials

A reading list will be continuedly updated before the start date of the course. To name a few:

1. Andreas C. Muller and Sarah Guido (2017). Introduction to machine learning with python. O'Reilly.
2. Ankur A. Patel (2019). Hands-on unsupervised learning using python. O'Reilly.
3. Alice Zhen and Amanda Casari (2018). Feature engineering for machine learning: principles and techniques for data scientists. O'Reilly.

Sebastin Raschka and Vahid Mirjalili (2018). Python machine learning: machine learning and deep learning with Python, scikit-learn, and tensorflow. 2nd ed. Packt Publishing.

Credit reductions