Version ML Data

See Extended Examples for complete configuration files and templates.

Implement data version control for machine learning datasets to ensure reproducibility and track data lineage.

When to Use

• Versioning large datasets that don't fit in Git
• Tracking data changes alongside code changes
• Ensuring reproducibility of ML experiments
• Building automated data pipelines with dependency tracking
• Sharing datasets across team members
• Rolling back to previous data versions
• Auditing data lineage for compliance
• Managing multiple dataset variants (train/test splits, feature sets)

Inputs