Week 4: Demo Proposal #2452
Merged
Week 4: Demo Proposal #2452
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
proposal demo for week 4, mlflow
|
Looks good, good luck! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Assignment Proposal
Title
Dynamic model rollbacks using MLflow
Names and KTH ID
Deadline
Category
Description
We will demonstrate how to implement and automate dynamic model rollbacks using MLflow’s Model Registry.
We will look at a model in production, and simulate a model performance drop or condition (set by us) when deploying a new version of the model.
Based on this condition, we will then trigger a rollback to an earlier reliable model (using Python code and the MLFlow API).
The process will be verified and visualized using the MLflow GUI.
Relevance
Using version management, automatic deployment and automatic reactions based on feedback from the production environment are core parts of the devops philosophy for software development.
By doing this in an MLOps context, we demonstrate how DevOps principles can be extended for use in MLOps.
There is no way to natively do this yet in MLflow, and there is even currently an issue on the main MLflow repository, showing that this is both difficult and relevant.