CLOVER (Convnet Line-fitting Of Velocities in Emission-line Regions) is a convolutional neural network (ConvNet) trained to identify spectra with two velocity components along the line of sight and predict their kinematics. It works with Gaussian emission lines (e.g., CO) and lines with hyperfine structure (e.g., NH3). CLOVER has two prediction steps, classification and parameter prediction. For the first step, CLOVER segments the pixels in an input data cube into one of three classes: noise (i.e., no emission), one-component (emission line with single velocity component), and two-component (emission line with two velocity components). For the pixels identified as two-components in the first step, a second regression ConvNet is used to predict centroid velocity, velocity dispersion, and peak intensity for each velocity component.

AREPO is a massively parallel gravity and magnetohydrodynamics code for astrophysics, designed for problems of large dynamic range. It employs a finite-volume approach to discretize the equations of hydrodynamics on a moving Voronoi mesh, and a tree-particle-mesh method for gravitational interactions. AREPO is originally optimized for cosmological simulations of structure formation, but has also been used in many other applications in astrophysics.

WVTICs generates glass-like initial conditions for Smoothed Particle Hydrodynamics. Relaxation of the particle distribution is done using an algorithm based on Weighted Voronoi Tesselations; additional particle reshuffling can be enabled to improve over- and undersampled maxima/minima. The WBTICs package includes a full suite of analytical test problems.

HISS stacks HI (emission and absorption) spectra in a consistent and reliable manner to enable statistical analysis of average HI properties. It provides plots of the stacked spectrum and reference spectrum with any fitted function, of the stacked noise response, and of the distribution of the integrated fluxes when calculating the uncertainties. It also produces a table containing the integrated flux calculated from the fitted functions and the stacked spectrum, among other output files.

EPOS (Exoplanet Population Observation Simulator) simulates observations of exoplanet populations. It provides an interface between planet formation simulations and exoplanet surveys such as Kepler. EPOS can also be used to estimate planet occurrence rates and the orbital architectures of planetary systems.

fgivenx plots a predictive posterior of a function, dependent on sampled parameters, for a Bayesian posterior Post(theta|D,M) described by a set of posterior samples {theta_i}~Post. If there is a function parameterized by theta y=f(x;theta), this script produces a contour plot of the conditional posterior P(y|x,D,M) in the (x,y) plane.

Morphological classification is one of the most demanding challenges in astronomy. With the advent of all-sky surveys, an enormous amount of imaging data is publicly available, and are typically analyzed by experts or encouraged amateur volunteers. For upcoming surveys with billions of objects, however, such an approach is not feasible anymore. PINK (Parallelized rotation and flipping INvariant Kohonen maps) is a simple yet effective variant of a rotation-invariant self-organizing map that is suitable for many analysis tasks in astronomy. The code reduces the computational complexity via modern GPUs and applies the resulting framework to galaxy data for morphological analysis.

PreProFit fits the pressure profile of galaxy clusters using Markov chain Monte Carlo (MCMC). The software can analyze data from different sources and offers flexible parametrization for the pressure profile. PreProFit accounts for Abel integral, beam smearing, and transfer function filtering when fitting data and returns χ2, model parameters and uncertainties in addition to marginal and joint probability contours, diagnostic plots, and surface brightness radial profiles. The code can be used for analytic approximations for the beam and transfer functions for feasibility studies.

a3cosmos-gas-evolution calculates galaxies' cold molecular gas properties using gas scaling functions derived from the A3COSMOS project. By known galaxies' redshifts or cosmic age, stellar masses, and star formation enhancement to galaxies' star-forming main sequence (Delta MS), the gas scaling functions predict their stellar mass ratio (gas fraction) and gas depletion time.

DM_phase maximizes the coherent power of a radio signal instead of its intensity to calculate the best dispersion measure (DM) for a burst such as those emitted by pulsars and fast radio bursts (FRBs). It is robust to complex burst structures and interference, thus mitigating the limitations of traditional methods that search for the best DM value of a source by maximizing the signal-to-noise ratio (S/N) of the detected signal.

exoplanet is a toolkit for probabilistic modeling of transit and/or radial velocity observations of exoplanets and other astronomical time series using PyMC3 (ascl:1610.016), a flexible and high-performance model building language and inference engine. exoplanet extends PyMC3's language to support many of the custom functions and distributions required when fitting exoplanet datasets. These features include a fast and robust solver for Kepler's equation; scalable Gaussian processes using celerite (ascl:1709.008); and fast and accurate limb darkened light curves using the code starry (ascl:1810.005). It also offers common reparameterizations for limb darkening parameters, and planet radius and impact parameters.

Emerge (Empirical ModEl for the foRmation of GalaxiEs) populates dark matter halo merger trees with galaxies using simple empirical relations between galaxy and halo properties. For each model represented by a set of parameters, it computes a mock universe, which it then compares to observed statistical data to obtain a likelihood. Parameter space can be explored with several advanced stochastic algorithms such as MCMC to find the models that are in agreement with the observations.

TLS is an optimized transit-fitting algorithm to search for periodic transits of small planets. In contrast to BLS: Box Least Squares (ascl:1607.008), which searches for rectangular signals in stellar light curves, TLS searches for transit-like features with stellar limb-darkening and including the effects of planetary ingress and egress. TLS also analyses the entire, unbinned data of the phase-folded light curve. TLS yields a ~10% higher detection efficiency (and similar false alarm rates) compared to BLS though has a higher computational load. This load is partly compensated for by applying an optimized period sampling and transit duration sampling constrained to the physically plausible range.

ECLIPS3D (Eigenvectors, Circulation, and Linear Instabilities for Planetary Science in 3 Dimensions) calculates a posteriori energy equations for the study of linear processes in planetary atmospheres with an arbitrary steady state, and provides both increased robustness and physical meaning to the obtained eigenmodes. It was developed originally for planetary atmospheres and includes python scripts for data analysis. ECLIPS3D can be used to study the initial spin up of superrotation of GCM simulations of hot Jupiters in addition to being applied to other problems.

orbitize fits the orbits of directly-imaged objects by packaging the Orbits for the Impatient (OFTI) algorithm and a parallel-tempered Markov Chain Monte Carlo (MCMC) algorithm into a consistent API. It accepts observations in three measurement formats, which can be mixed in the same input file, generates orbits, and plots the computed orbital parameters. orbitize offers numerous ways to visualize the data, including histograms, corner plots, and orbit plots. Generated orbits can be saved in HDF5 format for future use and analysis.

PEXO provides a global modeling framework for ns timing, μas astrometry, and μm/s radial velocities. It can account for binary motion and stellar reflex motions induced by planetary companions and also treat various relativistic effects both in the Solar System and in the target system (Roemer, Shapiro, and Einstein delays). PEXO is able to model timing to a precision of 1 ns, astrometry to a precision of 1 μas, and radial velocity to a precision of 1 μm/s.

LEO-Py uses a novel technique to compute the likelihood function for data sets with uncertain, missing, censored, and correlated values. It uses Gaussian copulas to decouple the correlation structure of variables and their marginal distributions to compute likelihood functions, thus mitigating inconsistent parameter estimates and accounting for non-normal distributions in variables of interest or their errors.

AOTOOLS reduces IR images from adaptive optics. It uses effective dithering, either sky subtraction or dark-subtration, and flat-fielding techniques to determine the effect of the instrument on an image of an object. It also performs bad pixel masking, degrades an AO on-axis PSF due to effects of anisoplanicity, and corrects an AO on-axis PSF due to effects of seeing.

E0102-VR facilitates the characterization of the 3D structure of the oxygen-rich optical ejecta in the young supernova remnant 1E 0102.2-7219 in the Small Magellanic Cloud. This room-scale Virtual Reality application written for the HTC Vive contributes to the exploration of the scientific potential of this technology for the field of observational astrophysics.

ANNz2, a newer implementation of ANNz (ascl:1209.009), utilizes multiple machine learning methods such as artificial neural networks, boosted decision/regression trees and k-nearest neighbors to measure photo-zs based on limited spectral data. The code dynamically optimizes the performance of the photo-z estimation and properly derives the associated uncertainties. In addition to single-value solutions, ANNz2 also generates full probability density functions (PDFs) in two different ways. In addition, estimators are incorporated to mitigate possible problems of spectroscopic training samples which are not representative or are incomplete. ANNz2 is also adapted to provide optimized solutions to general classification problems, such as star/galaxy separation.

MarsLux generates illumination maps of Mars from Digital Terrain Model (DTM), permitting users to investigate in detail the illumination conditions on Mars based on its topography and the relative position of the Sun. MarsLux consists of two Python codes, SolaPar and MarsLux. SolaPar calculates the matrix with solar parameters for one date or a range between the two. The Marslux code generates the illumination maps using the same DTM and the files generated by SolaPar. The resulting illumination maps show areas that are fully illuminated, areas in total shadow, and areas with partial shade, and can be used for geomorphological studies to examine gullies, thermal weathering, or mass wasting processes as well as for producing energy budget maps for future exploration missions.

MiSTree quickly constructs minimum spanning tree graphs for various coordinate systems, including Celestial coordinates, by using a k-nearest neighbor graph (k NN, rather than a matrix of pairwise distances) which is then fed to Kruskal's algorithm to create the graph. MiSTree bins the MST statistics into histograms and plots the distributions; enabling the inclusion of high-order statistics information from the cosmic web to provide additional information that improves cosmological parameter constraints. Though MiSTree was designed for use in cosmology, it can be used in any field requiring extracting non-Gaussian information from point distributions.

ChainConsumer consumes the chains output from Monte Carlo processes such as MCMC to produce plots of the posterior surface inferred from the chain distributions, to plot the chains as walks to check for mixing and convergence, and to output parameter summaries in the form of LaTeX tables. It handles multiple models (chains), allowing for model comparison using AIC, BIC or DIC metrics.

GetDist analyzes Monte Carlo samples, including correlated samples from Markov Chain Monte Carlo (MCMC). It offers a point and click GUI for selecting chain files, viewing plots, marginalized constraints, and LaTeX tables, and includes a plotting library for making custom publication-ready 1D, 2D, 3D-scatter, triangle and other plots. Its convergence diagnostics include correlation length and diagonalized Gelman-Rubin statistics, and the optimized kernel density estimation provides an automated optimal bandwidth choice for 1D and 2D densities with boundary and bias correction. It is available as a standalong package and with CosmoMC (ascl:1106.025).

Cobaya (Code for BAYesian Analysis) provides a framework for sampling and statistical modeling and enables exploration of an arbitrary prior or posterior using a range of Monte Carlo samplers, including the advanced MCMC sampler from CosmoMC (ascl:1106.025) and the advanced nested sampler PolyChord (ascl:1502.011). The results of the sampling can be analyzed with GetDist (ascl:1910.018). It supports MPI parallelization and is highly extensible, allowing the user to define priors and likelihoods and create new parameters as functions of other parameters.

It includes interfaces to the cosmological theory codes CAMB (ascl:1102.026) and CLASS (ascl:1106.020) and likelihoods of cosmological experiments, such as Planck, Bicep-Keck, and SDSS. Automatic installers are included for those external modules; Cobaya can also be used as a wrapper for cosmological models and likelihoods, and integrated it in other samplers and pipelines. The interfaces to most cosmological likelihoods are agnostic as to which theory code is used to compute the observables, which facilitates comparison between those codes. Those interfaces are also parameter-agnostic, allowing use of modified versions of theory codes and likelihoods without additional editing of Cobaya’s source.

OCD (O'Connell Effect Detector) detects eclipsing binaries that demonstrate the O'Connell Effect. This time-domain signature extraction methodology uses a supporting supervised pattern detection algorithm. The methodology maps stellar variable observations (time-domain data) to a new representation known as Distribution Fields (DF), the properties of which enable efficient handling of issues such as irregular sampling and multiple values per time instance. Using this representation, the code applies a metric learning technique directly on the DF space capable of specifically identifying the stars of interest; the metric is tuned on a set of labeled eclipsing binary data from the Kepler survey, targeting particular systems exhibiting the O’Connell Effect. This code is useful for large-scale data volumes such as that expected from next generation telescopes such as LSST.

The AOtools package offers generic adaptive optics processing tools in addition to astronomy-specific methods; among these are analyzing data in the pupil plane, images and point spread functions in the focal plane, wavefront sensors, modeling of atmospheric turbulence, physical optical propagation of wavefronts, and conversion functions to convert stellar brightness into photon flux for a given waveband. The software also calculates integrated atmospheric parameters, such as coherence time and isoplanatic angle from atmospheric turbulence and wind speed profile.

qnm computes the Kerr quasinormal mode frequencies, angular separation constants, and spherical-spheroidal mixing coefficients. The qnm package includes a Leaver solver with the Cook-Zalutskiy spectral approach to the angular sector, and a caching mechanism to avoid repeating calculations. A large cache of low ℓ, m, n modes is available for download and can be installed with a single function call and interpolated to provide good initial guess for root-polishing at new values of spin.

We have developed a method to efficiently simulate the dynamics of the magnetic flux in the solar network. We call this method Network Flux Transport (NFT). Implemented using a Spherical Centroidal Voronoi Tessellation (SCVT) based network model, magnetic flux is advected by photospheric plasma velocity fields according to the geometry of the SCVT model. We test NFT by simulating the magnetism of the Solar poles. The poles of the sun above 55 deg latitude are free from flux emergence from active regions or ephemeral regions. As such, they are ideal targets for a simplified simulation that relies on the strengths of the NFT model. This simulation method reproduces the magnetic and spatial distributions for the solar poles over two full solar cycles.

PLAN (PLanetesimal ANalyzer) identifies and characterizes planetesimals produced in numerical simulations of the Streaming Instability that includes particle self-gravity with code Athena (ascl:1010.014). PLAN works with the 3D particle output of Athena and finds gravitationally bound clumps robustly and efficiently. PLAN — written in C++ with OpenMP/MPI — is massively parallelized, memory-efficient, and scalable to analyze billions of particles and multiple snapshots simultaneously. The approach of PLAN is based on the dark matter halo finder HOP (ascl:1102.019), but with many customizations for planetesimal formation. PLAN can be easily adapted to analyze other object formation simulations that use Lagrangian particles (e.g., Athena++ simulations). PLAN is also equipped with a toolkit to analyze the grid-based hydro data (VTK dumps of primitive variables) from Athena, which requires the Boost MultiDimensional Array Library.

uvplot makes nice plots of deprojected interferometric visibilities (often called uvplots). It implements plotting functionality, handles MS tables with spectral windows with different number of channels, and can import visibilities from ASCII to MS Table. It also allows export of specific channels. uvplot can be installed inside the NRAO CASA package (ascl:1107.013).

OpenSPH runs hydrodynamical and N-body simulations and was written for asteroid collisions and subsequent gravitational evolution. The code offers SPH and N-body solvers with several different equations of state and material rheologies. It is written in C++14 with a modular object-oriented design, focused on extensibility and maintainability, and it can be used either as a library or as a standalone graphical program that allows to set up the problem in a convenient graphical node editor. The graphical program further allows real-time visualization of the simulation, diagnostics and tools for analysis of the results.

PypeIt reduces data from echelle and low-resolution spectrometers; the code can be run in several modes of reduction that demark the level of sophistication (e.g. quick and dirty vs. MonteCarlo) and also the amount of output written to disk. It also generates numerous data products, including 1D and 2D spectra; calibration images, fits, and meta files; and quality assurance figures.

MARTINI (Mock APERTIF-like Radio Telescope Interferometry of the Neutal ISM) creates synthetic resolved HI line observations (data cubes) of smoothed-particle hydrodynamics simulations of galaxies. The various aspects of the mock-observing process are divided logically into sub-modules handling the data cube, source, beam, noise, spectral model and SPH kernel. MARTINI is object-oriented: each sub-module provides a class (or classes) which can be configured as desired. For most sub-modules, base classes are provided to allow for straightforward customization. Instances of each sub-module class are then given as parameters to the Martini class. A mock observation is then constructed by calling a handful of functions to execute the desired steps in the mock-observing process.

ATHOS provides on-the-fly stellar parameter determination of FGK stars based on flux ratios from optical spectra. Once configured properly, it will measure flux ratios in the input spectra and deduce the stellar parameters effective temperature, iron abundance (a.k.a [Fe/H]), and surface gravity by employing pre-defined analytical relations. ATHOS can be configured to run in parallel in an arbitrary number of threads, thus enabling the fast and efficient analysis of huge datasets.

planetplanet models exoplanet transits, secondary eclipses, phase curves, and exomoons, as well as eclipsing binaries, circumbinary planets, and more. The code was originally developed to model planet-planet occultation (PPO) light curves for the TRAPPIST-1 system, but it is generally applicable to any exoplanet system. During a PPO, a planet occults (transits) the disk of another planet in the same planetary system, blocking its thermal (and reflected) light, which can be measured photometrically by a distant observer. planetplanet is coded in C and wrapped in a user-friendly Python interface.

HeatingRate calculates the nuclear heating rates [erg/s/g] of beta-decay, alpha-decay, and spontaneous fission of r-process nuclei, taking into account for thermalization of gamma-rays and charged decay products in r-process ejecta. It uses the half-lives and injection energy spectra from an evaluated nuclear data library (ENDF/B-VII.1). Each heating rate is computed for given abundances, ejecta mass, velocity, and density profile. HeatingRate also computes the bolometric light curve and the evolution of the effective temperature for given abundances, ejecta mass, velocity, and density profile assuming opacities independent of the wavelength.

frbpoppy conducts fast radio burst population synthesis and continues the work of PSRPOP (ascl:1107.019) and PsrPopPy (ascl:1501.006) in the realm of FRBs. The code replicates observed FRB detection rates and FRB distributions in three steps. It first simulates a cosmic population of one-off FRBs and allows the user to select options such as models for source number density, cosmology, DM host/IGM/Milky Way, luminosity functions, and emission bands as well as maximum redshift and size of the FRB population. The code then generates a survey by adopting a beam pattern using various survey parameters, among them telescope gain, sampling time, receiver temperature, central frequency, channel bandwidth, number of polarizations, and survey region limits. Finally, frbpoppy convolves the generated intrinsic population with the generated survey to simulate an observed FRB population.

Fruitbat estimates the redshift of Fast Radio Bursts (FRB) from their dispersion measure. The code combines various dispersion measure (DM) and redshift relations with the YMW16 galactic dispersion measure model into a single easy to use API.

IDG (Image Domain Gridding) is an imager that makes w-term corrections and a-term corrections computationally very cheap. It works with WSClean (ascl:1408.023) and supports the same cleaning and data selections options that WSClean offers in normal mode (such as cotton-schwab, multi-frequency multi-scale cleaning, and auto-masking). IDG also allows gridding with a time-variable beam including the LOFAR, AARTFAAC and MWA beam and can perform full beam or differential correction. The code requires measurement sets with four polarizations (e.g. XX/XY/YX/YY), can apply a spatially varying time-variable TEC term that can additionally be different for different antennas and output channels, and performs extremely well on GPUs.

Zeltron is an explicit 3D relativistic electromagnetic Particle-In-Cell code suited for modeling particle acceleration in astrophysical plasmas. The code is efficiently parallelized with the Message Passing Interface, and can be run on a laptop computer or on multiple cores on current supercomputers. Zeltron takes into account the effect of the radiation reaction force on the motion of the particles; it assigns variable weights to the macro-particles to model particle density gradients, and does not strictly conserve the total energy. The code uses linear interpolation to deposit the charges and currents generated by each particle at the nodes of the computational grid, and computes the charge and current densities for Maxwell's equations. Zeltron contains a large set of analysis tools, including plasma density, particle spectrum, optically thin synchrotron and inverse Compton spectra, angular distributions, and stress-energy tensor.

ATLAS performs the tracking, projecting, power-spectrum-making, and ring-fitting needed to turn a set of Dopplergram images into a set of frequency shift measurements. This code is essentially a combination of three codes, FRACK (FORTRAN Tracking), PSPEC (Power SPECtrum), and MRF (Multi-Ridge Fitting), included in the ATLAS package. ATLAS reads in a list of longitude/latitude coordinates corresponding to the desired tile centers and a set of full-disk Dopplergram images and outputs frequency shift measurements from each wave mode of each tile. The code relies on both distributed-memory (MPI) and shared-memory (OpenMP) parallelism to scale up to around 1000 processes. Due to the immense volume of data produced by the tracking and projecting steps, the intermediate data products (tiles, power spectra) are never written out.

MORDI (Massively-Overlapped Ring-Diagram Inversion) performs three-dimensional ring-diagram inversions. The code reads in frequency shift measurements and their associated sensitivity kernels and outputs two-dimensional slices of the subsurface flow field at a constant depth and (optionally) the associated averaging kernels. It relies on both distributed-memory (MPI) and shared-memory (OpenMP) parallelism to scale efficiently up to a few thousand processors, but can also run reasonably well on small machines (1-4 cpus). The actions of the code are modified by command-line parameters, which enable a significant amount of flexibility when setting up an inversion.

LATTICEEASY creates lattice simulations of the evolution of interacting scalar fields in an expanding universe. The program can do runs with different parameters and new models can be easily introduced for evaluation. Simulations can be done in one, two, or three dimensions by resetting a single variable. Mathematica notebooks for plotting the output and a range of models are also available for download; a parallel processing version of LATTICEEASY called CLUSTEREASY (ascl:1911.016) is also available.

CLUSTEREASY is a parallel programming extension of the simulation program LATTICEEASY (ascl:1911.015); running the program in parallel greatly extends the range of scales and times that can be simulated. The program is particularly useful for the study of reheating and thermalization after inflation.

HLattice simulates scalar fields and gravity in the early universe. The code allows the user to select between symplectic integrators, descretization schemes, and metrics such as Minkowski or FRW backgrounds and adaptice schemes in an "all-in-one" configuration file.

WhereWolf tracks (sub)haloes even if they have been lost by a halo finder in cosmological simulations and supplements halo catalogs such as VELOCIraptor (ascl:1911.020) with these recovered (sub)haloes. The code can improve measurements of the subhalo/halo mass function and present estimates of the distribution of radii at which subhaloes merge.

OrbWeaver extracts orbits from halo catalogs, enabling large statistical studies of their orbital parameters. The code is run in two stages. For the first run, a configuration file is used to modify orbit host selection and the region around orbit host used for the superset of orbiting halos. Each orbit host has a orbit forest (containing halos that passed within the region of interest); the code generates a pre-processed catalog which contains a superset of orbiting halo for each identified orbit host. The second run uses the file list generated in the first stage for the creation of the orbit catalog, which is the final output.

VELOCIraptor-STF, formerly STructure Finder (ascl:1306.009), is a 6-Dimensional Friends-of-Friends (6D-FoF) phase space halo finder and constructs halo catalogs. The code uses using MPI and OpenMP APIs and can be compiled as a library for on-the-fly halo finding within an N-body/hydrodynamnical code. There is an associated halo merger tree code TreeFrog (ascl:1911.021).

TreeFrog reads in particle IDs information between various structure catalogs and cross matches catalogs, assuming that particle IDs are unique and constant across snapshots. Though it is built as a cross correlator (in that it can match particles across several different catalogs), its principle use is as halo merger tree builder. TreeFrog produces links between objects found at different snapshots (or catalogs) and uses several possible functions to evaluate the merit of a link between one object at a given snapshot (or in a given catalog) to another object in a previous snapshot (or different catalog). It can also produce a full graph. The code utilizes MPI and OpenMP. It is optimzed for reading VELOCIraptor (ascl:1911.020) output but can also read output from other structure finders such as AHF (ascl:1102.009).