Experiment Management in Domino Data Lab

In this session with Principal Field Engineer Sameer Wadkar, we'll introduce MLflow, teach you about its core components and how they are used, and do a technical demo that shows experiments creation, execution, comparison, registration, and deployment. You will also learn about how MLflow is secured via Domino Integration, and what additional advantages the combination of an enterprise MLOps platform and MLflow brings to accelerate model velocity.

Domino's integration with MLFlow simplifies machine learning lifecycle management for data scientists. It enables data scientists to track, reproduce, and share machine learning experiments and artifacts within their Domino projects, while Domino's security layer ensures metrics, logs, and artifacts are secured.

Experiment tracking in machine learning is a challenging problem, as keeping track of datasets, algorithms, pre-processing steps, hyperparameters, tools and libraries is essential for maintaining a proper system of record. The open-source, code-first framework MLflow has been gaining popularity among data science practitioners as it enables them to easily compare thousands of experiments and brings a level of transparency and standardization to the way they run experiments.