Managing Datasets

This guide covers best practices for organizing and managing datasets in Mission Control.

Organization Strategies

By Project

Organize datasets by the project they belong to:

• Group related data together
• Use clear naming conventions
• Add descriptions for context

By Data Type

Separate datasets by data format:

• Images for 2D annotation
• Point clouds for 3D annotation
• Video sequences

By Stage

Track data through your pipeline:

• Raw uploads
• In progress (being annotated)
• Completed (ready for training)

Tagging Best Practices

Use consistent tags for easy filtering:

Search and Filter

Mission Control supports powerful search:

tags:annotated visibility:private

Search Operators

• field:value - Exact match
• "quoted phrase" - Phrase match
• -term - Exclude term

Filterable Fields

Bulk Operations

Select multiple datasets for bulk actions:

• Bulk tag: Add or remove tags from many datasets
• Bulk delete: Remove datasets (with confirmation)
• Bulk visibility: Change visibility settings

Dataset Settings

Visibility

• Private: Only you and collaborators can access
• Public: Listed in public marketplace

Collaborators

Add team members who can view or edit your dataset:

1. Go to dataset Settings
2. Click Add Collaborator
3. Search for user by username or email
4. Set permission level

Best Practices

1. Use descriptive names: “Training Images Batch 3” is better than “data”
2. Add descriptions: Help teammates understand the dataset contents
3. Tag consistently: Establish tagging conventions for your team
4. Archive completed work: Move finished datasets to archive to keep workspace clean