Spectral Clustering for Molecular Emission Segmentation

SCIMES identifies relevant molecular gas structures within dendrograms of emission using the spectral clustering paradigm. SCIMES is useful to decompose objects in complex enviroments imaged at high resolution.

Documentation

• Installing SCIMES
  • Requirements
  • Installation
• The concepts behind the algorithm
  • Which dataset to use?
  • Building the dendrogram
  • A dendrogram as a graph
  • From the graph to the affinity matrix
  • Cutting the graph through the Spectral Clustering
  • Clustered dendrogram
  • Giant Molecular Clouds in Orion-Monoceros
• SCIMES tutorial
  • Building the dendrogram and the structure catalog
  • Clustering the dendrogram
  • Clustering results
  • Difference between pixel and physical property-based segmentation
  • Affinity matrix choice
  • SCIMES behaviour at low resolution

Reporting issues and getting help

Please help us improve this package by reporting issues via GitHub. You can also open an issue if you need help with using the package.

Developers

This package is developed by:

• Dario Colombo
• Erik Rosolowsky
• Adam Ginsburg
• Ana Duarte-Cabral
• Annie Hughes

Acknowledgements

Thanks for the help with the documentation to:

• Eric Koch

Citing SCIMES

If you make use of this package in a publication, please add the following acknowledgment:

This research made use of SCIMES, a Python package to find relevant structures into dendrograms of molecular gas emission using the spectral clustering approach and cite the related paper, 2015MNRAS.454.2067C

SCIMES is an astropy affiliated-package. Please also consider adding an acknowledgment for Astropy (see http://www.astropy.org for the latest recommended citation).

Public API

scimes Package

Functions

test([package, test_path, args, plugins, …])

Run the tests using py.test.

Classes

SpectralCloudstering(dendrogram, catalog, header)

Apply the spectral clustering to find the best cloud segmentation out from a dendrogram.