confitti 0.2.3

Fit conic sections (ellipse, parabola, hyperbola) to a set of 2d points

🎊 Confitti 🎊 - for Conics Fitting

Fit conic sections (ellipse, parabola, hyperbola) to a set of points

See the presentation

What's new

• 2024-11-03: Version 0.2.1 released. Fixed bug in writing save files when eccentricity does not vary.
• 2024-10-16: Version 0.2.0 released. Added ability to save model fit parameters to a file, and to read them back in.
• For more details, see the full change log.

Installation

pip install confitti

or

uv pip install confitti

will install the package plus the required dependencies (numpy, scipy, lmfit).

Optional dependencies, which are used in the example notebooks, may be pip-installed separately:

• emcee for MCMC sampling
• matplotlib, seaborn, and corner for plotting
• astropy and regions for dealing with celestial coordinates

Usage

See the example jupyter notebooks in the notebooks directory. For example,

• demo01-basic.ipynb demonstrates basic usage: finding the best-fit parabola (or general conic) to a set of (x, y) points
• demo02-emcee.ipynb explores uncertainty in the parameters of the best-fit curve by means of mcmc
• demo03-proplyd.ipynb is an example application to real astronomical data (HST image of a bow shock in the Orion Nebula). This also shows how to save the best-fit parameters to a file.
• demo04-other-proplyd.ipynb is a second example application to a different bow shock, which turns out very differently because the best-fit arc is an ellipse, rather than a hyperbola
• demo05-quadrants.ipynb tests that the code can handle axis orientations in all four quadrants and close to the wrap angle of 360 degrees
• demo06-read-save-files.ipynb demonstrates how to read in the multiple saved files from previous fits.

Prior art

This is the successor project to circle-fit

Some of the literature on the topic of fitting conic sections to points is described here.

Funding

We gratefully acknowledge financial support provided by Dirección General de Asuntos del Personal Académico, Universidad Nacional Autónoma de México, through grant IN109823 of the ‘Programa de Apoyo a Proyectos de Investigación e Inovación Tecnológica’.