Data Versioning and Experiment Tracking for Machine Learning
Chapter 1: The Need for Reproducibility in Machine Learning
Challenges in Managing ML Projects
Why Git Alone Is Not Sufficient
Defining Reproducibility in ML
Components of a Reproducible ML Workflow
Introduction to Data Versioning Concepts
Introduction to Experiment Tracking Concepts
Quiz for Chapter 1
Chapter 2: Versioning Data with DVC
Data Versioning Strategies
Introducing Data Version Control (DVC)
Setting Up DVC in a Project
Tracking Data Files and Directories
Storing and Retrieving Data Versions
Connecting DVC to Remote Storage (S3, GCS, Azure Blob)
Switching Between Data Versions
Hands-on Practical: Versioning a Dataset
Quiz for Chapter 2
Chapter 3: Tracking Experiments with MLflow
The Importance of Experiment Tracking
Introducing MLflow Tracking
Setting up MLflow
Logging Parameters and Metrics
Logging Artifacts (Models, Plots, Files)
Organizing Runs with Experiments
Using the MLflow UI
Comparing Experiment Runs
Practice: Tracking a Training Run
Chapter 4: Integrating DVC and MLflow for Reproducible Workflows
Connecting Data Versions to Experiments
Structuring Projects for Integration
Logging DVC Metadata in MLflow
Creating DVC Pipelines
Reproducing DVC Pipelines
Tracking DVC Pipeline Metrics
Combining DVC Pipelines and MLflow Tracking
Best Practices for Integrated Workflows
Hands-on Practical: Building an Integrated Pipeline
Comparing Experiment Runs
Was this section helpful?
- MLflow Tracking: Record and Compare Runs, MLflow Project, 2024 (MLflow Project) - Official documentation describing MLflow's tracking capabilities, including the UI for visual comparisons and the API for programmatic analysis of experiment runs.
- Building Machine Learning Powered Applications: Going from Idea to Product, Emmanuel Ameisen, 2020 (O'Reilly Media) - A practical guide to the MLOps lifecycle, emphasizing the need for structured experiment tracking and comparison to develop and iterate on models effectively.
- MLflow: An Open Platform for the Machine Learning Lifecycle, Andrew Chen, Aaron J. Meng, Shreya Makkar, Sue Ann Hong, Michael Armbrust, Sameer Singh, Andrew Fogarty, Matei Zaharia, 2020 Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data (SIGMOD '20) (ACM) DOI: 10.1145/3318534.3329623 - The original paper introducing MLflow, detailing its architecture and features, including the tracking component for managing and comparing machine learning experiments.