Link Search Menu Expand Document

Machine Learning Operations

Work in progress!

Repository for course 02476 at DTU containing lectures and exercises.

Checkout the homepage!

Course details

  • Course responsible
    • Postdoc Nicki Skafte Detlefsen,
    • Professor Søren Hauberg,
  • 5 ECTS
  • 3 week period of January 2022
  • Master course
  • Grade: Pass/not passed
  • Type of assessment: weekly project updates + final oral examination/presentation
  • Recommended prerequisites: 02456 (Deep Learning) or experience with the following:
    • General understanding of machine learning (datasets, probability, classifiers, overfitting ect.) and basic knowledge about deep learning (backpropagation, convolutional neural networks, auto-encoders ect.)
    • Coding in Pytorch

Course setup

Start by cloning or downloading this repository

git clone

If you do not have git installed (yet) we will touch upon it in the course. The folder will contain all exercise material for this course and lectures. Additionally, you should join our slack channel which we use for communication:

MLOps: What is it?

Machine Learning Operations (MLOps) is a rather new field that has seen its uprise as machine learning and particular deep learning has become a technology that is widely available. The term itself is a compound of “machine learning” and “operations” and covers everything that has to do with the management of the production ML lifecycle.

The lifecycle of production ML can largely be divided into three phases:

  1. Design: The initial phase starts with a investigation of the problem. Based on this analysis, a number of requirements can be prioterized of what we want our future model to actual do. Since machine learning requires data to be trained, we also investigate in this step what data we have and if we need to source it in some other way.

  2. Model development: Based on the design phase we can actually begin to conjour some machine learning algorithm to solve our problems. As always, the initial step often involve doing some data analysis to make sure that our model is actually learning the signal that we want it to learn. Secondly, is the machine learning engenering phase, where the particular model architechture is chosen. Finally, we also need to do validation and testing to make sure that our model is generalizing well.

  3. Operations: Based on the model development phase, we now have a model and we actual want to use. The operations is where create an automatic pipeline that makes sure that whenever we make changes to our codebase they gets automatilly incorporated into our model, such that we do not slow down production. Equally important is also the ongoing monitoring of already deployed models to make sure that they behave exactly as we specified them.

It is important to note that the three steps are in fact a cycle, meaning that we you have successfully deployed a machine learning model that is not the end of it. Your initial requirements may change, forcing you to revisit the design phase. Some new algorithm may show promising results, so you revisit the model development phase to implement this. And finally, you may try to cut the cost of running your model in production, making you revisit the operations phase, trying to optimize some steps.

The focus in this course is particular on the Operations part of MLOps as this is what many data scientist are missing in their toolbox to take all the knowledge they have about data processing and model development into a production setting.

Additional resources (in no particular order):

Learning objectives

General course objective

Introduce the student to a number of coding practises that will help them organisation, scale, monitor and deploy machine learning models either in a research or production setting. To provide hands-on experience with a number of frameworks, both local and in the cloud, for doing large scale machine learning models.

This includes:

  • Organize code in a efficient way for easy maintainability and shareability
  • Understand the importance of reproducibility and how to create reproducible containerized applications and experiments
  • Cable of using version control to efficiently collaborate on code development
  • Knowledge of continuous integration (CI) and continuous machine learning (CML) for automating code development
  • Being able to debug, profile, visualize and monitor multiple experiments to assess model performance
  • Cable of using online cloud based computing services to scale experiments
  • Demonstrate knowledge about different distributed training paradigms within machine learning and how to apply them
  • Deploy machine learning models, both locally and in the cloud
  • Conduct a research project in collaboration with follow students using the frameworks taught in the course


I highly value open-source, and the content of this course is therefore free to use under the Apache 2.0 license. If you use parts of this course in your own work, please cite using:

    author       = {Nicki Skafte Detlefsen},
    title        = {Machine Learning Opeartions},
    howpublished = {\url{}},
    year         = {2021}