Study Guide

Objective

The student is familiar with common machine learning models supported by online platforms for machine learning. The student can create custom machine learning applications in the cloud without having to write complex code.

Content

Machine learning models supported by the largest cloud providers: binary prediction, category prediction and value prediction.

Building, training and deploying machine learning model in the cloud:
- Preparing the data
- Training the model to learn from the data
- Deploying the model
- Evaluate the model's performance

Materials

All needed materials will be collected in the Moodle workspace. New material will be added from the internet as needed, due to the nature of cloud services and their rapid development rate. Documentations and tutorials of used cloud services, for example, Google Cloud, AWS, CSC and/or MS Azure.

Teaching methods

Lectures and workshops. Practical exercises in classroom.

Exam schedules

The course will be graded based on personal work and exercises

Content scheduling

- Cloud computing in general
- Common cloud environments
- Basics of neural networks and deep learning
- Technological considerations in deep learning
- Basics of image classification and object detection
- Machine learning coding in the cloud
- Other cloud omputing related tools

Evaluation scale

H-5

Assessment criteria, satisfactory (1)

The student can implement a machine learning application in the cloud.

Assessment criteria, good (3)

The student can implement machine learning applications in the cloud using different models.

Assessment criteria, excellent (5)

The student can implement machine learning applications in the cloud using different models and cloud machine learning services.

Assessment methods and criteria

The course will be graded based on exercises done (portfolio), both the quality, quantity and overall challenge will be taken into account. A final report is also required.