Mission: Make GitLab a tool Data Scientists and Machine Learning Engineers love to use.
Vision: Identify opportunities in our portfolio to explore ways where GitLab can provide a better user experience for Data Science and Machine Learning across the entire Machine Learning life cycle (model creation, testing, deployment, monitoring, and iteration).
Updates
May 1st 2023: Model 15.11 Overview
Current Project
Model Registry & Model Experiments
Mission: Make it dead simple for Data Scientists to manage their model lifecycle within GitLab, from testing different candidates to,
packaging a new model version.
This is a project led by the Incubation Engineer - MLOps
DRI: @eduardobonet
Epic All Merge Requests All Issues Usage Mission Make it dead simple for Data Scientists to track their model experiments within GitLab, while making it easy to access the experiments results across the product (on Model Registry, on MRs, on Issues, etc)
Guiding Principles GitLab enables Data Scientists to use Experiment Tracking without requiring any support from Platform Engineers Data Scientists can start using GitLab Experiment Tracking without any changes to their codebase Experiment Tracking is not just a vertical feature, it integrates across the platform to enhance experience in different steps of model development Approach: GitLab as a backend for MLFlow MlFlow is the most popular open source option for Experiment Tracking, providing a powerful client that allows Data Scientists to log a wide range of models.
This is a project led by the Incubation Engineer - MLOps
DRI: @eduardobonet
Epic All Merge Requests All Issues Mission Make it dead simple for Data Scientists to track their model lifecycle with GitLab, tracking different candidates, promoting them into model versions so that they can be served and deployed.
Guiding Principles GitLab enables Data Scientists to use Model Registry without requiring any support from Platform Engineers Model Registry is not just a vertical feature, it integrates across the platform to enhance experience in different steps of model development Approach: Model registry as Package registry Model registry is implemented using the GitLab package registry.
