beekeeping 🐝

Manage video metadata for animal behaviour experiments.

Overview

beekeeping is a web-based dashboard built with Dash-Plotly for managing video metadata in animal behaviour experiments. It provides an intuitive interface for creating, editing, and organizing metadata files associated with experimental videos.

Key Features:

• Upload and manage project configurations
• Interactive metadata table with editing capabilities
• Bulk import from spreadsheets (CSV/Excel)
• Automatic detection of videos missing metadata
• Export metadata as YAML files
• Responsive web interface with Bootstrap theming

It is based on an earlier codebase called WAZP, but focuses specifically on metadata management.

Installation

Install beekeeping inside a conda environment:

conda create -n beekeeping-env -c conda-forge python=3.12
conda activate beekeeping-env
git clone https://github.com/neuroinformatics-unit/beekeeping.git
cd beekeeping
pip install .

Getting Started

1. Launch the application

   start-beekeeping
   

   The app opens in your browser at http://localhost:8050

2. Upload Project Configuration

   • Navigate to the Home page
   • Upload your project_config.yaml file containing:

     videos_dir_path: /path/to/your/videos
     metadata_fields_file_path: /path/to/metadata_fields.yaml
     metadata_key_field_str: File
     

3. Manage Video Metadata

   • Navigate to the Metadata page to view/edit your video metadata table
   • Each row represents one video file's metadata

Core Operations

Viewing and Editing Metadata

• Browse: Scroll through the paginated table (25 rows per page)
• Sort: Click column headers to sort data
• Hide/Show columns: Use column visibility controls
• Edit: Click any editable cell to modify values directly
• Row selection: Click checkboxes to select rows for batch operations

Adding Missing Videos

• Click "Check for missing metadata files"
• The app scans your videos directory and adds rows for videos without .metadata.yaml files
• File extensions supported: .avi, .mp4

Manual Data Entry

• Click "Add empty row" to create a new metadata entry
• Fill in the fields as needed
• The filename field links to your video file

Batch Operations

• Select All/Unselect All: Mass row selection controls
• Export Selected: Click "Export selected rows as yaml" to save .metadata.yaml files for selected videos
• Import from Spreadsheet: Upload CSV/Excel files to bulk generate metadata files

Spreadsheet Import Process

1. Prepare your spreadsheet (CSV or Excel) with columns matching your metadata fields
2. Click "Generate yaml files from spreadsheet"
3. Upload your file
4. The app will:
   • Match spreadsheet columns to metadata fields
   • Add missing columns with empty values
   • Only process rows with corresponding video files
   • Generate .metadata.yaml files in your videos directory
   • Show confirmation with count of files generated

License

⚖️ BSD 3-Clause