Astronomical Data
The Astronomical Data group solves hard problems in measurement and discovery. Depending on the context, the challenges in these problems can stem from the sheer size or complexity of the datasets, the precision requirements for measurements, or the subtlety or structure of the signals of interest.
The group develops and maintains advanced tools for the astrophysics community, especially ones for building and using probabilistic or generative models, but it does so by carrying out in-house data analysis projects that answer important scientific questions. Currently the team is concentrating on extrasolar planet discovery and characterization, precision measurement of stellar chemical abundances, and precise mapping of the Milky Way and of large-scale cosmological structure. As new data become available and the challenges evolve, so will the scope of the work. By sharing all its results, and by developing and maintaining open-source software, the Astronomical Data group directly supports scientific reproducibility and open science.
Research Highlights
An Ultraviolet-Optical Color-Metallicity relation for Red Clump Stars using GALEX and Gaia
S. Mohammed, D. Schiminovich, K. Hawkins, B. Johnson, D. Wang, D. Hogg
arXiv:1805.03236 The information content in cold stellar streams
A. Bonaca, D. Hogg
Cold stellar streams—produced by tidal disruptions of clusters—are long-lived, coherent dynamical features in the halo of the Milky Way. Due…
arXiv:1804.06854News & Announcements
Group Members
Software
Astrometry.net
If you have astronomical imaging of the sky with celestial coordinates you do not know—or do not trust—then Astrometry.net is for you. Input an image and we'll give you back astrometric calibration meta-data, plus lists of known objects falling inside the field of view.
emcee
emcee is an extensible, pure-Python implementation of Goodman & Weare's Affine Invariant Markov chain Monte Carlo (MCMC) Ensemble sampler. It's designed for Bayesian parameter estimation.
celerite
celerite is a library for fast and scalable Gaussian Process (GP) Regression in one dimension with implementations in C++, Python, and Julia. The Python implementation is the most stable and it exposes the most features but it relies on the C++ implementation for computational efficiency.
DAFT
Daft is a Python package that uses matplotlib to render pixel-perfect probabilistic graphical models for publication in a journal or on the internet. With a short Python script and an intuitive model-building syntax you can design directed (Bayesian Networks, directed acyclic graphs) and undirected (Markov random fields) models and save them in any formats that matplotlib supports (including PDF, PNG, EPS and SVG).
George
George is a fast and flexible Python library for Gaussian Process Regression. It capitalizes on the Hierarchical Off-Diagonal Low-Rank formalism to make controlled approximations for fast execution.