RABET is a cross-platform desktop application for annotating animal behaviour from video. It combines frame-accurate playback, keyboard-driven event coding, an interactive timeline, multi-file analysis, raster visualisation, reliability assessment, bout analysis, and first-order transition analysis in one self-contained tool.
RABET ships as a self-contained binary for Windows, macOS, and Linux. No system-wide VLC, FFmpeg, Python, R, or codec-pack installation is required for normal use.
The Annotation view: video player, recording controls, action map, and colour-coded timeline.
Video and annotation
- Frame-accurate playback with single-frame stepping and fast seeking.
- Configurable keyboard-to-behaviour action maps.
- State behaviours for duration events, recorded from key press to key release.
- Point behaviours for instantaneous events, recorded as zero-duration ticks with onset equal to offset.
- Timed recording sessions with pause, resume, rewind handling, undo, and an editable timeline.
Analysis and export
- Multi-file annotation analysis with whole-session and interval summaries.
- Per-behaviour duration and frequency columns.
- Custom latency metrics and overlap-aware total-time metrics.
- Mean and SEM summary rows for quick inspection.
- One-click clipboard copy and CSV export.
Bout and transition analysis
- Bout analysis with user-defined bout criterion interval (BCI), advisory BCI estimation, per-animal bout summaries, and bout raster plots.
- First-order transition analysis with counts, conditional probabilities, odds ratios, expected counts, and adjusted residuals.
- Optional bout-level transition analysis, transition time windows, pooled count summaries, heatmaps, and antecedent-window predictability.
- PNG / SVG / PDF figure export with user-selectable DPI.
Reliability assessment
- Summary mode compares two
summary_table.csvfiles using ICC(2,1), Pearson correlation, and mean absolute difference. - Detailed mode compares two annotation CSVs using time-binned Cohen's kappa, Krippendorff's alpha, raw percentage agreement, and event-raster disagreement review.
- Summary-mode statistics are cross-checked against an R reference script
using
psych::ICC.
Project management
- Keep videos, annotations, action maps, and analysis outputs together in a project directory.
- Save projects as either self-contained copies or references to existing files.
- Persistent UI settings, recent files, and file-dialog locations.
The latest release is published on the GitHub Releases page. The Zenodo concept DOI is kept for citation and long-term archival reference: 10.5281/zenodo.15313025.
Download the asset matching your platform:
| Platform | Asset |
|---|---|
| Windows installer | RABET-Windows-1.4.2-Setup.zip |
| Windows portable | RABET-Windows-1.4.2-portable.zip |
| macOS (Apple Silicon) | RABET-macOS-arm64-1.4.2.dmg |
| macOS (Intel) | RABET-macOS-x86_64-1.4.2.dmg |
| Linux | RABET-Linux-x86_64-1.4.2.AppImage |
For the installer build, unzip RABET-Windows-1.4.2-Setup.zip and run
RABET-Setup.exe. For the portable build, unzip
RABET-Windows-1.4.2-portable.zip and launch RABET.exe directly.
Windows SmartScreen may warn on first launch because the app is not code signed. Choose More info and then Run anyway if you trust the downloaded release.
Open the DMG and drag RABET.app to Applications. The macOS builds are
unsigned and not notarized, so Gatekeeper may report that the app is damaged
or cannot be verified. This is quarantine metadata, not a corrupt download.
The most reliable one-time fix is:
xattr -dr com.apple.quarantine /Applications/RABET.appThen open RABET.app normally.
Make the AppImage executable and run it:
chmod +x RABET-Linux-x86_64-1.4.2.AppImage
./RABET-Linux-x86_64-1.4.2.AppImagegit clone https://github.com/mi2e-K/RABET.git
cd RABET
# Recommended: pinned conda environment
conda env create -f environment.yml
conda activate rabet_build
# Alternative: pip
# pip install -e .
python main.pyPython 3.11 or newer is required. The pinned conda environment uses Python 3.12.
Project website: mi2e-k.github.io/RABET
- User Guide (English)
- User Guide (Japanese)
- CSV format specification
- Reliability assessment reference
- Build instructions for macOS / Linux
Annotation exports contain metadata, an event log, and a per-behaviour summary:
Metadata
RABET Version,1.4.2
Test Duration (seconds),300
Event,Onset,Offset
RecordingStart,0.0000,0.0000
Attack bites,12.4123,12.6480
Head dip,20.2500,20.2500
Behavior,Duration,Frequency
Attack bites,0.24,1
Head dip,0.00,1State events have duration. Point events are instantaneous and therefore have
Onset == Offset; their duration is zero, but their frequency is counted.
Times are seconds with four decimal places. Full details are in
docs/CSV_FORMAT.md.
If RABET supports your research, please cite it. Machine-readable metadata is in CITATION.cff. A human-readable form is:
Mitsui, K. (2026). RABET - Real-time Animal Behavior Event Tagger (Version 1.4.2) [Computer software]. https://github.com/mi2e-K/RABET doi:10.5281/zenodo.15313025
The DOI above is the Zenodo concept DOI, intended to remain stable across RABET versions. When citing a specific binary release, also report the exact RABET version used.
A tool paper describing RABET is in preparation.
Released under the MIT License.
Bug reports, feature requests, and pull requests are welcome.
- Issues: include reproducible steps, the RABET version from Help > About, and the relevant log file when possible.
- Pull requests: keep them scoped, include a short description, and reference the related issue when applicable.
RABET is built on PySide6, PyAV, numpy, pandas, matplotlib, scipy, pingouin, krippendorff, filetype, and PyInstaller. We are grateful to the maintainers and contributors of these projects.