mono-cbp 0.1.4

Pipeline for detecting circumbinary planets in TESS light curves

mono-cbp: Search for Monotransits of Circumbinary Planets

A Python package for detecting circumbinary planets in TESS eclipsing binary light curves through the identification of single transit events ("monotransits").

Overview

mono-cbp is a pipeline designed to systematically search for circumbinary planets by detecting individual transit signatures in TESS eclipsing binary systems. The pipeline automates the complete workflow from masking stellar eclipses, threshold crossing event (TCE) detection, Bayesian vetting, and completeness analysis, making it easy to process large catalogues of eclipsing binaries.

Key Features

• Eclipse Masking: Automatically mask primary and secondary eclipses in eclipsing binary light curves using eclipse positions and widths and binary ephemeris provided by an input catalogue
• Transit Detection: Removes unwanted trends from the input light curves and performs single-event detection using the by identifying Threshold Crossing Events (see Hawthorn et al. 2024)
• Bayesian Model Comparison: Event classification to discern transit-like events and systematics/detrending artefacts
• Injection-Retrieval Testing: Completeness analysis through synthetic transit injection and recovery statistics
• Modular Architecture: Use individual components independently or run the complete integrated pipeline
• Configuration-Driven: Easily customise parameters via Python dictionaries or JSON files without modifying code
• Command-Line Interface: Shell scripts and CLI subcommands for batch processing and reproducibility

Installation

Requirements

• Python 3.8 or higher (tested most rigorously with Python 3.9)

From PyPI (Recommended)

The easiest way to install mono-cbp is from PyPI:

pip install mono-cbp

It is advisable to install mono-cbp into a Python environment using your favourite package manager, e.g. for conda:

conda create --name mono-cbp python=3.9
conda activate mono-cbp
pip install mono-cbp

This installs the package and creates the mono-cbp command-line tool.

From Source

For development or to use the latest unreleased features:

git clone https://github.com/bdrdavies/mono-cbp.git
cd mono-cbp
pip install -e .

This installs the package in editable mode.

Verify Installation

To check that the installation has been successful:

python -c "import mono_cbp; print(mono_cbp.__version__)"
mono-cbp --help

Dependencies

All dependencies are automatically installed when you install mono-cbp.

See pyproject.toml or requirements.txt for the complete dependency list and version constraints.

Troubleshooting Installation

If you encounter issues:

• Python version: The package has been tested most thoroughly with Python 3.9.
• Dependency conflicts: The package pins specific versions of PyMC, ArviZ, and Bokeh for compatibility. If you have conflicts, create a fresh environment
• Import errors: If you see errors related to bokeh or arviz, ensure you have the correct versions installed (see pyproject.toml)

Examples & Tutorials

There are a series of Jupyter notebooks in the examples/ directory to demonstrate how to use the package in your own code:

1. 00_download_light_curves.ipynb - Download TESS light curves in the mono-cbp format using lightkurve
2. 01_complete_pipeline.ipynb - End-to-end workflow on sample data
3. 02_eclipse_masking.ipynb - Eclipse masking demo
4. 03_transit_finding.ipynb - TCE detection example
5. 04_model_comparison.ipynb - Bayesian model comparison example
6. 05_injection_retrieval.ipynb - Completeness testing

Documentation

Documentation is available in the docs/ directory:

• docs/quickstart.md - Quickstart guide
• docs/data_formats.md - Input and output data format specifications
• docs/configuration.md - Configuration system reference
• docs/api_reference.md - API documentation

Support & Contact

For questions, issues, or feature requests:

• Issues: Open an issue on GitHub Issues
• Documentation: Review the full documentation
• Email: [e-mail]