Course name: Machine Learning Development and Deployment (ML-dev)

COURSE SYLLABUS, third-cycle courses

ECTS/HP: 7.5 ECTS

Course code:
SRT0005

Educational level: third-cycle course

Entry requirements:

• Basic knowledge in Machine Learning
• Programming knowledge
• Admission to third-cycle studies involving Machine Learning or Machine Learning applications

Course content:

The course focuses on the dissemination of machine learning tools to the general public or the issues of reproducibility in machine learning research. It is important to promote collaboration, reusability, and the efficiency through tools and workflows. The course covers source code management, machine learning deployment, collaboration in programming, computational resources, data management, and implementation.

Learning outcomes:

• improving and applying practical skills in machine learning related programming
• contribution to bigger software projects
• being able to deploy machine learning algorithms for practical applications.
• managing high performance computing resources
• efficiently running machine learning projects with the ML-dev framework

Course methods:

The course introduces into the most important methods for machine learning deployment, software implementation and collaboration, and the ML-dev framework of the Machine Learning Group.
The Machine Learning Development Framework (ML-dev) comprises the general functionalities Machine Learning projects to implement algorithms and get projects going more efficiently.

The course will be realized in Canvas and comprises several modules for which credits are given individually:

• Source Code, Management (Git, GitHub, semantic commit messages, formatting, repositories, access, benefits for research, version control system concepts/general, ...) (1 ECTS)
• Source Code, Collaboration (branches, workflows, pull requests, issue management, access rights, documentation, ...) (1 ECTS)
• Dev Project, Setup (development environment, Docker, develop in containers, base images + libs, reproducibility, Visual Studio Code, tests, Python example, ...) (1 ECTS)
• Dev Project, Computational Resources (Clusters, Docker remote host, GPU usage, resource allocation, logging, notifications, long run training, deployment, ...) (1 ECTS)
• Implementation, Technology Stack (Technology Radar, libraries, methodologies, CRISP-DM, software development principals, architectures, Jupyter Notebook, ...) (1 ECTS)
• Implementation, Data Management (databases, SQL, NoSQL, data types, Docker, access rights, security, versioning, reports, results, ...) (1 ECTS)
• Implementation, ML-dev framework (concept, benefits, architecture, usage, deployment, contribute, ...) (1.5 ECTS)

Examination form:
Successful completion of all the quizzes in Canvas as well as the practical tasks for each module.

Grading scale: Pass/Fail

Course literature:

to be decided together with and approved by a senior member in the Machine Learning subject.

Education cycle:
Any stage of PhD studies

Course is given periodically: Continuously

Send application to: any senior member of the machine learning group
Doctoral student enter name, civic registration number, e-mail, Division and Department in the application

Deadline for application:
continuous participation possible

Course open for application by doctoral students admitted to other universities than LTU: Yes □

Limited number of students: No □

Tuition:
If the course is allocated resources via internal resource allocation system, the course is free of charge for doctoral students admitted at LTU, for doctoral students admitted to other universities, course fees may be required. For other courses the Department decides on course fees.

Contact person:
Rajkumar Saini (Rajkumar.Saini@ltu.se)

Examiner:
Marcus Liwicki

Course syllabus decided by:
Johan Carlson, forskarutbildningsledare SRT, LTU-2634-2026

Date of decision: 2026-03-24