MAPPINGS III is a general purpose astrophysical plasma modelling code. It is principally intended to predict emission line spectra of medium and low density plasmas subjected to different levels of photoionization and ionization by shockwaves. MAPPINGS III tracks up to 16 atomic species in all stages of ionization, over a useful range of 102 to 108 K. It treats spherical and plane parallel geometries in equilibrium and time-dependent models. MAPPINGS III is useful for computing models of HI and HII regions, planetary nebulae, novae, supernova remnants, Herbig-Haro shocks, active galaxies, the intergalactic medium and the interstellar medium in general. The present version of MAPPINGS III is a large FORTRAN program that runs with a simple TTY interface for historical and portability reasons. A newer version of this software, MAPPINGS V (ascl:1807.005), is available.

MapCurvature, written in IDL, can create map projections with Goldberg-Gott indicatrices. These indicatrices measure the flexion and skewness of a map, and are useful for determining whether features are faithfully reproduced on a particular projection.

The MapCUMBA package applies a multigrid fast iterative Jacobi algorithm for map-making in the context of CMB experiments.

MANTA-Ray (Modified Absorption of Non-spherical Tiny Aggregates in the RAYleigh regime) calculates the absorption efficiency and absorption cross-section of fractal aggregates. This fast tool can be used to estimate the amount of absorption that occurs in dust and haze particles of any shape, provided that the electromagnetic radiation is in the Rayleigh regime, often in the context of planetary atmospheres or protoplanetary disks. MANTA-Ray obtains values within 10-20% of those calculated by a rigorous benchmark model (the discrete dipole approximation), but is 10¹³ times faster.

Mangrove uses Graph Neural Networks to regress baryonic properties directly from full dark matter merger trees to infer galaxy properties. The package includes code for preprocessing the merger tree, and training the model can be done either as single experiments or as a sweep. Mangrove provides loss functions, learning rate schedulers, models, and a script for doing the training on a GPU.

Mangle deals accurately and efficiently with complex angular masks, such as occur typically in galaxy surveys. Mangle performs the following tasks: converts masks between many handy formats (including HEALPix); rapidly finds the polygons containing a given point on the sphere; rapidly decomposes a set of polygons into disjoint parts; expands masks in spherical harmonics; generates random points with weights given by the mask; and implements computations for correlation function analysis. To mangle, a mask is an arbitrary union of arbitrarily weighted angular regions bounded by arbitrary numbers of edges. The restrictions on the mask are only (1) that each edge must be part of some circle on the sphere (but not necessarily a great circle), and (2) that the weight within each subregion of the mask must be constant. Mangle is complementary to and integrated with the HEALPix package (ascl:1107.018); mangle works with vector graphics whereas HEALPix works with pixels.

The MaNGA Data Reduction Pipeline (DRP) processes the raw data to produce flux calibrated, sky subtracted, coadded data cubes from each of the individual exposures for a given galaxy. The DRP consists of two primary parts: the 2d stage that produces flux calibrated fiber spectra from raw individual exposures, and the 3d stage that combines multiple flux calibrated exposures with astrometric information to produce stacked data cubes. These science-grade data cubes are then processed by the MaNGA Data Analysis Pipeline (ascl:2203.017), which measures the shape and location of various spectral features, fits stellar population models, and performs a variety of other analyses necessary to derive astrophysically meaningful quantities from the calibrated data cubes.

The MaNGA data analysis pipeline (MaNGA DAP) analyzes the data produced by the MaNGA data-reduction pipeline (ascl:2203.016) to produced physical properties derived from the MaNGA spectroscopy. All survey-provided properties are currently derived from the log-linear binned datacubes (<i>i.e.</i>, the LOGCUBE files).

The Maneage (Managing data lineage; ending pronounced like "lineage") framework produces fully reproducible computational research. It provides full control on building the necessary software environment from a low-level C compiler, the shell and LaTeX, all the way up to the high-level science software in languages such as Python without a third-party package manager. Once the software environment is built, adding analysis steps is as easy as defining "Make" rules to allow parallelized operations, and not repeating operations that do not need to be recreated. Make provides control over data provenance. A Maneage'd project also contains the narrative description of the project in LaTeX, which helps prepare the research for publication. All results from the analysis are passed into the report through LaTeX macros, allowing immediate dynamic updates to the PDF paper when any part of the analysis has changed. All information is stored in plain text and is version-controlled in Git. Maneage itself is actually a Git branch; new projects start by defining a new Git branch over it and customizing it for a new project. Through Git merging of branches, it is possible to import infrastructure updates to projects.

MAMPOSSt (Modeling Anisotropy and Mass Profiles of Observed Spherical Systems) is a Bayesian code to perform mass/orbit modeling of spherical systems. It determines marginal parameter distributions and parameter covariances of parametrized radial distributions of dark or total matter, as well as the mass of a possible central black hole, and the radial profiles of density and velocity anisotropy of one or several tracer components, all of which are jointly fit to the discrete data in projected phase space. It is based upon the MAMPOSSt likelihood function for the distribution of individual tracers in projected phase space (projected radius and line-of-sight velocity) and the CosmoMC Markov Chain Monte Carlo code (ascl:1106.025), run in generic mode. MAMPOSSt is not based on the 6D distribution function (which would require triple integrals), but on the assumption that the local 3D velocity distribution is an (anisotropic) Gaussian (requiring only a single integral).

MALU visualizes integral field spectroscopy (IFS) data such as CALIFA, MANGA, SAMI or MUSE data producing fully interactive plots. The tool is not specific to any instrument. It is available in Python and no installation is required.

MaLTPyNT (Matteo's Libraries and Tools in Python for NuSTAR Timing) provides a quick-look timing analysis of NuSTAR data, properly treating orbital gaps and exploiting the presence of two independent detectors by using the cospectrum as a proxy for the power density spectrum. The output of the analysis is a cospectrum, or a power density spectrum, that can be fitted with XSPEC (ascl:9910.005) or ISIS (ascl:1302.002). The software also calculates time lags. Though written for NuSTAR data, MaLTPyNT can also perform standard spectral analysis on X-ray data from other satellite such as XMM-Newton and RXTE. This code has been merged into HENDRICS (ascl:1805.019).

MAKEE (MAuna Kea Echelle Extraction) reduces data from the HIRES and ESI instruments at Keck Observatory. It is optimized for the spectral extraction of single, unresolved point sources and is designed to run non-interactively using a set of default parameters. Taking the raw HIRES FITS files as input, the code determines the position (or trace) of each echelle order, defines the object and background extraction boundaries, optimally extracts a spectrum for each order, and computes wavelength calibrations. MAKEE produces FITS format "spectral images" (each row is a separate echelle order spectrum) and the data values are in arbitrary (relative) flux units. MAKEE will reduce data from all HIRES formats, including the single CCD format, the single CCD with Red and UV cross dispersers, and the current 3 CCD system. It can handle a variety of pixel binnings, including 1x1, 1x2, 1x4 (column x row).

MakeCloud makes turbulent giant molecular cloud (GMC) initial conditions for GIZMO (ascl:1410.003). It generates turbulent velocity fields on the fly and stores that data in a user-specified path for efficiency. The code is flexible, allowing the user control through various parameters, including the radius of the cloud, number of gas particles, type of initial turbulent velocity (Gaussian or full), and magnetic energy as a fraction of the binding energy, among other options. With an additional file, it can also create glassy initial conditions.

MAH calculates the posterior distribution of the "minimum atmospheric height" (MAH) of an exoplanet by inputting the joint posterior distribution of the mass and radius. The code collapses the two dimensions of mass and radius into a one dimensional term that most directly speaks to whether the planet has an atmosphere or not. The joint mass-radius posteriors derived from a fit of some exoplanet data (likely using MCMC) can be used by MAH to evaluate the posterior distribution of R_MAH, from which the significance of a non-zero R_MAH (i.e. an atmosphere is present) is calculated.

Magritte performs 3D radiative transfer modeling; though focused on astrophysics and cosmology, the techniques can also be applied more generally. The code uses a deterministic ray-tracer with a formal solver that currently focuses on line radiative transfer. Magritte can either be used as a C++ library or as a Python package.

MAGRATHEA solves planet interiors and considers the case of fully differentiated interiors. The code integrates the hydrostatic equation in order to determine the correct planet radius given the mass in each layer. The code returns the pressure, temperature, density, phase, and radius at steps of enclosed mass. The code support four layers: core, mantle, hydrosphere, and atmosphere. Each layer has a phase diagram with equations of state chosen for each phase.

MAGRATHEA (Multi-processor Adaptive Grid Refinement Analysis for THEoretical Astrophysics) is a foundational cosmological library and a relativistic raytracing code. Classical linear algebra libraries come with their own operations and can be difficult to leverage for new data types. Instead of providing basic types, MAGRATHEA provides tools to generate base types such as scalar quantities, points, vectors, or tensors.

Magrathea-Pathfinder propagates photons within cosmological simulations to construct observables. This high-performance framework uses a 3D Adaptive-Mesh Refinement and is built on top of the MAGRATHEA metalibrary (ascl:2203.023).

MAGPy-RV (Modelling stellar Activity with Gaussian Processes in Radial Velocity) models data with Gaussian Process regression and affine invariant Monte Carlo Markov Chain parameter searching. Developed to model intrinsic, quasi-periodic variations induced by the host star in radial velocity (RV) surveys for the detection of exoplanets and the accurate measurements of their orbital parameters and masses, it now includes a variety of kernels and models and can be applied to any timeseries analysis. MAGPy-RV includes publication level plotting, efficient posterior extraction, and export-ready LaTeX results tables. It also handles multiple datasets at once and can model offsets and systematics from multiple instruments. MAGPy-RV requires no external dependencies besides basic python libraries and corner (ascl:1702.002).

MAGPHYS is a self-contained, user-friendly model package to interpret observed spectral energy distributions of galaxies in terms of galaxy-wide physical parameters pertaining to the stars and the interstellar medium. MAGPHYS is optimized to derive statistical constraints of fundamental parameters related to star formation activity and dust content (e.g. star formation rate, stellar mass, dust attenuation, dust temperatures) of large samples of galaxies using a wide range of multi-wavelength observations. A Bayesian approach is used to interpret the SEDs all the way from the ultraviolet/optical to the far-infrared.

Magnetron, written in Python, decomposes magnetar bursts into a superposition of small spike-like features with a simple functional form, where the number of model components is itself part of the inference problem. Markov Chain Monte Carlo (MCMC) sampling and reversible jumps between models with different numbers of parameters are used to characterize the posterior distributions of the model parameters and the number of components per burst.

Magnetizer computes time and radial dependent magnetic fields for a sample of galaxies in the output of a semi-analytic model of galaxy formation. The magnetic field is obtained by numerically solving the galactic dynamo equations throughout history of each galaxy. Stokes parameters and Faraday rotation measure can also be computed along a random line-of-sight for each galaxy.

Large-scale coherent magnetic fields are observed in galaxies and clusters, but their ultimate origin remains a mystery. We reconsider the prospects for primordial magnetogenesis by a cosmic string network. We show that the magnetic flux produced by long strings has been overestimated in the past, and give improved estimates. We also compute the fields created by the loop population, and find that it gives the dominant contribution to the total magnetic field strength on present-day galactic scales. We present numerical results obtained by evolving semi-analytic models of string networks (including both one-scale and velocity-dependent one-scale models) in a Lambda-CDM cosmology, including the forces and torques on loops from Hubble redshifting, dynamical friction, and gravitational wave emission. Our predictions include the magnetic field strength as a function of correlation length, as well as the volume covered by magnetic fields. We conclude that string networks could account for magnetic fields on galactic scales, but only if coupled with an efficient dynamo amplification mechanism.

MAGNETAR is a set of tools for the study of the magnetic field in simulations of MHD turbulence and polarization observations. It calculates the histogram of relative orientation between density structure in the magnetic field in data cubes from simulations of MHD turbulence and observations of polarization using the method of histogram of relative orientations (HRO).

MAGIX provides an interface between existing codes and an iterating engine that minimizes deviations of the model results from available observational data; it constrains the values of the model parameters and provides corresponding error estimates. Many models (and, in principle, not only astrophysical models) can be plugged into MAGIX to explore their parameter space and find the set of parameter values that best fits observational/experimental data. MAGIX complies with the data structures and reduction tools of Atacama Large Millimeter Array (ALMA), but can be used with other astronomical and with non-astronomical data.

The R suite magicaxis makes useful and pretty plots for scientific plotting and includes functions for base plotting, with particular emphasis on pretty axis labelling in a number of circumstances that are often used in scientific plotting. It also includes functions for generating images and contours that reflect the 2D quantile levels of the data designed particularly for output of MCMC posteriors where visualizing the location of the 68% and 95% 2D quantiles for covariant parameters is a necessary part of the post MCMC analysis, can generate low and high error bars, and allows clipping of values, rejection of bad values, and log stretching.

MagIC simulates fluid dynamics in a spherical shell. It solves for the Navier-Stokes equation including Coriolis force, optionally coupled with an induction equation for Magneto-Hydro Dynamics (MHD), a temperature (or entropy) equation and an equation for chemical composition under both the anelastic and the Boussinesq approximations. MagIC uses either Chebyshev polynomials or finite differences in the radial direction and spherical harmonic decomposition in the azimuthal and latitudinal directions. The time-stepping scheme relies on a semi-implicit Crank-Nicolson for the linear terms of the MHD equations and a Adams-Bashforth scheme for the non-linear terms and the Coriolis force.

The MAGIC (Microlensing Analysis Guided by Intelligent Computation) PyTorch framework efficiently and accurately infers the microlensing parameters of binary events with realistic data quality. The code divides binary microlensing parameters into two groups, which are inferred separately with different neural networks. The neural controlled differential equation handles light curves with irregular sampling and large data gaps. MAGIC can achieve fractional uncertainties of a few percent on the binary mass ratio and separation, and can locate the degenerate solutions even when large data gaps are introduced. As irregular samplings are common in astronomical surveys, this code may be useful for other time series studies.

MAGI (MAny-component Galaxy Initializer) generates initial conditions for numerical simulations of galaxies that resemble observed galaxies and are dynamically stable for time-scales longer than their characteristic dynamical times, taking into account galaxy bulges, discs, and haloes. MAGI adopts a distribution-function-based method and supports various kinds of density models, including custom-tabulated inputs and the presence of more than one disc, and is fast and easy to use.

MAESTROeX solves the equations of low Mach number hydrodynamics for stratified atmospheres or stars with a general equation of state. It includes reactions and thermal diffusion and can be used on anything from a single core to 100,000s of processor cores with MPI + OpenMP. MAESTROeX maintains the accuracy of its predecessor MAESTRO (ascl:1010.044) while taking advantage of a simplified temporal integration scheme and leveraging the AMReX software framework for block-structured adaptive mesh refinement (AMR) applications.

MAESTRO, a low Mach number stellar hydrodynamics code, simulates long-time, low-speed flows that would be prohibitively expensive to model using traditional compressible codes. MAESTRO is based on an equation set derived using low Mach number asymptotics; this equation set does not explicitly track acoustic waves and thus allows a significant increase in the time step. MAESTRO is suitable for two- and three-dimensional local atmospheric flows as well as three-dimensional full-star flows, and adaptive mesh refinement (AMR) has been incorporated into the code. The expansion of the base state for full-star flows using a novel mapping technique between the one-dimensional base state and the Cartesian grid is also available. NOTE: MAESTRO is no longer being actively developed. Users should switch to MAESTROeX (ascl:1908.019) to take advantage of the latest capabilities.

maelstrom models binary orbits through the phase modulation technique. This set of custom PyMC3 models and solvers fit each individual datapoint in the time series by forward modeling the time delay onto the light curve. This approach fully captures variations in a light curve caused by an orbital companion.

MADYS (Manifold Age Determination for Young Stars) determines the age and mass of young stellar and substellar objects. The code automatically retrieves and cross-matches photometry from several catalogs, estimates interstellar extinction, and derives age and mass estimates for individual objects through isochronal fitting. MADYS harmonizes the heterogeneity of publicly-available isochrone grids and the user can choose amongst several models, some of which have customizable astrophysical parameters. Particular attention has been dedicated to the categorization of these models, labeled through a four-level taxonomical classification.

MadVoro constructs three‑dimensional Voronoi diagrams in distributed‐memory systems using a parallel algorithm built on Delaunay triangulation. The library decomposes a point set across multiple processors, constructs local Delaunay triangulations, and then couples them via distributed communication to form a global Voronoi tessellation. It exposes an API that gives access to cells, faces, vertices, centroids, and cell volumes, enabling detailed mesh analysis at scale. MadVoro supports load balancing and data redistribution across hosts to optimize performance in high‑throughput simulations.

MADmap produces maximum-likelihood images of the sky from time-ordered data which include correlated noise, such as those gathered by Cosmic Microwave Background (CMB) experiments. It works efficiently on platforms ranging from small workstations to the most massively parallel supercomputers. Map-making is a critical step in the analysis of all CMB data sets, and the maximum-likelihood approach is the most accurate and widely applicable algorithm; however, it is a computationally challenging task. This challenge will only increase with the next generation of ground-based, balloon-borne and satellite CMB polarization experiments. The faintness of the B-mode signal that these experiments seek to measure requires them to gather enormous data sets. MADmap has the ability to address problems typically encountered in the analysis of realistic CMB data sets. The massively parallel and distributed implementation is detailed and scaling complexities are given for the resources required. MADmap is capable of analyzing the largest data sets now being collected on computing resources currently available.

MADLens produces non-Gaussian cosmic shear maps at arbitrary source redshifts. A MADLens simulation with only 256^3 particles produces convergence maps whose power agree with theoretical lensing power spectra up to scales of L=10000. The code is based on a highly parallelizable particle-mesh algorithm and employs a sub-evolution scheme in the lensing projection and a machine-learning inspired sharpening step to achieve these high accuracies.

MADHAT (Model-Agnostic Dark Halo Analysis Tool) analyzes gamma-ray emission from dwarf satellite galaxies and dwarf galaxy candidates due to dark matter annihilation, dark matter decay, or other nonstandard or unknown astrophysics. The tool is data-driven and model-independent, and provides statistical upper bounds on the number of observed photons in excess of the number expected using a stacked analysis of any selected set of dwarf targets. MADHAT also calculates the resulting bounds on the properties of dark matter under any assumptions the user makes regarding dark sector particle physics or astrophysics.

MadDM computes dark matter relic abundance and dark matter nucleus scattering rates in a generic model. The code is based on the existing MadGraph 5 architecture and as such is easily integrable into any MadGraph collider study. A simple Python interface offers a level of user-friendliness characteristic of MadGraph 5 without sacrificing functionality. MadDM is able to calculate the dark matter relic abundance in models which include a multi-component dark sector, resonance annihilation channels and co-annihilations. The direct detection module of MadDM calculates spin independent / spin dependent dark matter-nucleon cross sections and differential recoil rates as a function of recoil energy, angle and time. The code provides a simplified simulation of detector effects for a wide range of target materials and volumes.

MADCUBA analyzes astronomical datacubes and multiple spectra from various astronomical facilities, including ALMA, Herschel, VLA, IRAM 30m, APEX, GBT, and others. These telescopes, and in particular ALMA, generate extremely large datacubes (spatial, spectral and polarization). This software combines a user-friendly interface and powerful data analysis system to derive the physical conditions of molecular gas, its chemical complexity and the kinematics from datacubes. Built using the ImageJ (ascl:1206.013) infrastructure, MADCUBA visualizes astronomical datacubes with thousands on spectral channels, and datasets with thousands of spectra; it also identifies molecular species using publicly available molecular catalogs. It can automatically derive the physical parameters of the molecular species: column density, excitation temperature, velocity and linewidths and provides the best non-linear least-squared fit using the Levenberg-Marquardt algorithm, among other tasks.

MADCOW is a set of parallelized programs written in ANSI C and Fortran 77 that perform a maximum likelihood analysis of visibility data from interferometers observing the cosmic microwave background (CMB) radiation. This software has been used to produce power spectra of the CMB with the Very Small Array (VSA) telescope.

Photometric rotational modulations due to starspots remain the most common and accessible way to study stellar activity. Modelling rotational modulations allows one to invert the observations into several basic parameters, such as the rotation period, spot coverage, stellar inclination and differential rotation rate. The most widely used analytic model for this inversion comes from Budding (1977) and Dorren (1987), who considered circular, grey starspots for a linearly limb darkened star. That model is extended to be more suitable in the analysis of high precision photometry such as that by Kepler. Macula, a Fortran 90 code, provides several improvements, such as non-linear limb darkening of the star and spot, a single-domain analytic function, partial derivatives for all input parameters, temporal partial derivatives, diluted light compensation, instrumental offset normalisations, differential rotation, starspot evolution and predictions of transit depth variations due to unocculted spots. The inclusion of non-linear limb darkening means macula has a maximum photometric error an order-of-magnitude less than that of Dorren (1987) for Sun-like stars observed in the Kepler-bandpass. The code executes three orders-of-magnitude faster than comparable numerical codes making it well-suited for inference problems.

This repository implements an optimized XGBoost-based framework for photometric classification of Type Ia supernovae, addressing class imbalance through PR-AUC and F1-score prioritization. The approach is designed for scalability in large-scale astronomical surveys such as LSST and ensures improved classification robustness compared to traditional metrics like ROC-AUC.

The MATLAB Astronomy and Astrophysics Toolbox (MAAT) is a collection of software tools and modular functions for astronomy and astrophysics written in the MATLAB environment. It includes over 700 MATLAB functions and a few tens of data files and astronomical catalogs. The scripts cover a wide range of subjects including: astronomical image processing, ds9 control, astronomical spectra, optics and diffraction phenomena, catalog retrieval and searches, celestial maps and projections, Solar System ephemerides, planar and spherical geometry, time and coordinates conversion and manipulation, cosmology, gravitational lensing, function fitting, general utilities, plotting utilities, statistics, and time series analysis.

m2mcluster performs made-to-measure modeling of star clusters, and can fit target observations of a Galactic globular cluster's 3D density profile and individual kinematic properties, including proper motion velocity dispersion, and line of sight velocity dispersion. The code uses AMUSE (ascl:1107.007) to model the gravitational <i>N</i>-body evolution of the system between time steps; GalPy (ascl:1411.008) is also required.

M_SMiLe computes an approximation of the probability of magnification for a lens system consisting of microlensing by compact objects within a galaxy cluster. It specifically focuses on the scenario where the galaxy cluster is strongly lensing a background galaxy and the compact objects, such as stars, are sensitive to this microlensing effect. The microlenses responsible for this effect are stars and stellar remnants, though exotic objects such as compact dark matter candidates (including PBHs and axion mini-halos) can contribute to this effect.

M_-M_K- converts absolute 2MASS Ks-band magnitude (or a distance and a Ks-band magnitude) into an estimate of the stellar mass using the empirical relation derived from the resolved photometry and orbits of astrometric binaries. The code requires scalar values for K, distance, and corresponding uncertainties. M_-M_K- outputs errors based on the relationship's scatter and errors in the provided distance and Ks magnitude.

LZIFU (LaZy-IFU) is an emission line fitting pipeline for integral field spectroscopy (IFS) data. Written in IDL, the pipeline turns IFS data to 2D emission line flux and kinematic maps for further analysis. LZIFU has been applied and tested extensively to various IFS data, including the SAMI Galaxy Survey, the Wide-Field Spectrograph (WiFeS), the CALIFA survey, the S7 survey and the MUSE instrument on the VLT.

LyaCoLoRe uses CoLoRe (ascl:2111.009) simulations to generate simulated Lyman alpha forest spectra. The code takes the output files from CoLoRe as an input, carries out several stages of processing, and produces realistic skewers of transmitted flux fraction as an output. The repository includes tools to tune the parameters within LyaCoLoRe's transformation, and to measure the 1D power spectrum of output skewers quickly.

Long Wavelength Propagation Capability (LWPC), written as a collection of separate programs that perform unique actions, generates geographical maps of signal availability for coverage analysis. The program makes it easy to set up these displays by automating most of the required steps. The user specifies the transmitter location and frequency, the orientation of the transmitting and receiving antennae, and the boundaries of the operating area. The program automatically selects paths along geographic bearing angles to ensure that the operating area is fully covered. The diurnal conditions and other relevant geophysical parameters are then determined along each path. After the mode parameters along each path are determined, the signal strength along each path is computed. The signal strength along the paths is then interpolated onto a grid overlying the operating area. The final grid of signal strength values is used to display the signal-strength in a geographic display. The LWPC uses character strings to control programs and to specify options. The control strings have the same meaning and use among all the programs.

Lux models astronomical data with noisy labels using a multi-output, latent-variable framework. It simultaneously infers latent parameters and predicts multiple observed properties while accounting for measurement uncertainties and label noise. The code generates synthetic observations for model validation and supports probabilistic analyses of stellar and galactic datasets. Lux enables flexible model training and evaluation and handles heterogeneous datasets efficiently.

LUNA generates dynamically accurate lightcurves from a planet-moon pair, analytically accounting for shadow overlaps, stellar limb darkening, and planet-moon dynamical motion. The code takes transit timing/duration variations and ingress/egress asymmetries into consideration not only for the planet, but also the moon. LUNA was designed to be analytical and dynamical and to incorporate limb darkening (including non-linear laws) and account for all orbital elements, including eccentricity and longitude of the ascending node. Because the software is precise and analytic, LUNA is a highly potent tool for exomoon detection.

LumFunc is a numerical code to model the Luminosity Function based on central galaxy luminosity-halo mass and total galaxy luminosity-halo mass relations. The code can handle rest b_J-band (2dFGRS), r'-band (SDSS), and K-band luminosities, and any redshift with redshift dependences specified by the user. It separates the luminosity function (LF) to conditional luminosity functions, LF as a function of halo mass, and also to galaxy types. By specifying a narrow mass range, the code will return the conditional luminosity functions. The code returns luminosity functions for galaxy types as well (broadly divided to early-type and late-type). The code also models the cluster luminosity function, either mass averaged or for individual clusters.

luas builds Gaussian processes (GPs) primarily for two-dimensional data sets. It uses different optimizations to make the application of GPs to 2D data sets possible within a reasonable timeframe. The code is implemented using Jax (ascl:2111.002), which helps calculate derivatives of the log-likelihood as well as permitting the code to be easily run on either CPU or GPU. luas can be used with popular inference frameworks such as NumPyro (ascl:2505.005) and PyMC. The package makes it easier to account for systematics correlated across two dimensions in data sets, in addition to being helpful for any other applications (<i>e.g.</i>, interpolation).

LtU-ILI (Learning the Universe Implicit Likelihood Inference) performs machine learning parameter inference. Given labeled training data or a stochastic simulator, the LtU-ILI piepline automatically trains state-of-the-art neural networks to learn the data-parameter relationship and produces robust, well-calibrated posterior inference. The package comes with a wide range of customizable complexity, including posterior-, likelihood-, and ratio-estimation methods for ILI, including sequential learning analogs, and various neural density estimators, including mixture density networks, conditional normalizing flows, and ResNet-like ratio classifiers. It offers fully-customizable, exotic embedding networks, including CNNs and Graph Neural Networks, and a unified interface for multiple ILI backends such as sbi, pydelfi, and lampe. LtU-ILI also handles multiple marginal and multivariate posterior coverage metrics, and offers Jupyter and command-line interfaces and a parallelizable configuration framework for efficient hyperparameter tuning and production runs.

LTS_LINEFIT and LTS_PLANEFIT are IDL programs to robustly fit lines and planes to data with intrinsic scatter. The code combines the Least Trimmed Squares (LTS) robust technique, proposed by Rousseeuw (1984) and optimized in Rousseeuw & Driessen (2006), into a least-squares fitting algorithm which allows for intrinsic scatter. This method makes the fit converge to the correct solution even in the presence of a large number of catastrophic outliers, where the much simpler σ-clipping approach can converge to the wrong solution. The code is also available in Python as ltsfit.

LTL provides dynamic arrays of up to 7-dimensions, subarrays and slicing, support for fixed-size vectors and matrices including basic linear algebra operations, expression templates-based evaluation, and I/O facilities for ascii and FITS format files. Utility classes for command-line processing and configuration-file processing are provided as well.

LTdwarfIndices studies spectral indices to determine whether one or more brown dwarfs are photometric variable candidates. For a single brown dwarf, it analyzes a given set of indices and outputs the number of graphs the object appears in in the variable area, whether it is a variable or non-variable candidate, and, optionally, an index-index or histogram plot. Using another code module, LTdwarftIndices can also analyze a set of sample indices for many brown dwarfs.

LSSGALPY provides visualization tools to compare the 3D positions of a sample (or samples) of isolated systems with respect to the locations of the large-scale structures galaxies in their local and/or large scale environments. The interactive tools use different projections in the 3D space (right ascension, declination, and redshift) to study the relation of the galaxies with the LSS. The tools permit visualization of the locations of the galaxies for different values of redshifts and redshift ranges; the relationship of isolated galaxies, isolated pairs, and isolated triplets to the galaxies in the LSS can be visualized for different values of the declinations and declination ranges.

LSDCat is a conceptually simple but robust and efficient detection package for emission lines in wide-field integral-field spectroscopic datacubes. The detection utilizes a 3D matched-filtering approach for compact single emission line objects. Furthermore, the software measures fluxes and extents of detected lines. LSDCat is implemented in Python, with a focus on fast processing of large data-volumes.

The Large Survey Database (LSD) is a Python framework and DBMS for distributed storage, cross-matching and querying of large survey catalogs (>10^9 rows, &gt;1 TB). The primary driver behind its development is the analysis of Pan-STARRS PS1 data. It is specifically optimized for fast queries and parallel sweeps of positionally and temporally indexed datasets. It transparently scales to more than &gt;10^2 nodes, and can be made to function in "shared nothing" architectures.

LSCS (Lightweight Space Coronagraph Simulator) simulates realistic high-contrast space imaging instruments in their linear regime of small wavefront perturbations about the nominal dark hole. The code can be used for testing high-order wavefront sensing and control as well as post-processing algorithms. It models broadband images with sensor noise, wavefront drift, actuators drift, and residual effects from low-order wavefront sensing, and supports a model of the Roman Space Telescope Hybrid Lyot Coronagraph based on its FALCO (ascl:2304.004, ascl:2304.005) model. The LSCS package provides an example of dark hole maintenance using an Extended Kalman Filter and Electric Field Conjugation.

LSC (LINEAR Supervised Classification) trains a number of classifiers, including random forest and K-nearest neighbor, to classify variable stars and compares the results to determine which classifier is most successful. Written in R, the package includes anomaly detection code for testing the application of the selected classifier to new data, thus enabling the creation of highly reliable data sets of classified variable stars.

LRGS (Linear Regression by Gibbs Sampling) implements a Gibbs sampler to solve the problem of multivariate linear regression with uncertainties in all measured quantities and intrinsic scatter. LRGS extends an algorithm by Kelly (2007) that used Gibbs sampling for performing linear regression in fairly general cases in two ways: generalizing the procedure for multiple response variables, and modeling the prior distribution of covariates using a Dirichlet process.

This software computes likelihoods for the Luminous Red Galaxies (LRG) data from the Sloan Digital Sky Survey (SDSS). It includes a patch to the existing <a href="http://ascl.net/1102.026">CAMB</a> software (ascl:1102.026; the February 2009 release) to calculate the theoretical LRG halo power spectrum for various models. The code is written in Fortran 90 and has been tested with the Intel Fortran 90 and GFortran compilers.

The Limited Post-Newtonian N-body code (LPNN) simulates post-Newtonian interactions between a massive object and many low-mass objects. The interaction between one massive object and low-mass objects is calculated by post-Newtonian approximation, and the interaction between low-mass objects is calculated by Newtonian gravity. This code is based on the sticky9 code, and can be accelerated with the use of GPU in a CUDA (version 4.2 or earlier) environment.

LPF (Live Pulse Finder) provides real-time automated analysis of the radio image data stream at multiple frequencies. The fully automated GPU-based machine-learning backed pipeline performs source detection, association, flux measurement and physical parameter inference. At the end of the pipeline, an alert of a significant detection of a transient event can be sent out and the data saved for further investigation.

LP-VIcode computes variational chaos indicators (CIs) quickly and easily. The following CIs are included: <ul><li>Lyapunov Indicators, also known as Lyapunov Characteristic Exponents, Lyapunov Characteristic Numbers or Finite Time Lyapunov Characteristic Numbers (LIs) <li>Mean Exponential Growth factor of Nearby Orbits (MEGNO) <li>Slope Estimation of the largest Lyapunov Characteristic Exponent (SElLCE) <li>Smaller ALignment Index (SALI) <li>Generalized ALignment Index (GALI) <li>Fast Lyapunov Indicator (FLI) <li>Orthogonal Fast Lyapunov Indicator (OFLI) <li>Spectral Distance (SD) <li>dynamical Spectra of Stretching Numbers (SSNs) <li>Relative Lyapunov Indicator (RLI) </ul>

We present a set of low resolution empirical SED templates for AGNs and galaxies in the wavelength range from 0.03 to 30 microns based on the multi-wavelength photometric observations of the NOAO Deep-Wide Field Survey Bootes field and the spectroscopic observations of the AGN and Galaxy Evolution Survey. Our training sample is comprised of 14448 galaxies in the redshift range 0&lt;~z&lt;~1 and 5347 likely AGNs in the range 0&lt;~z&lt;~5.58. We use our templates to determine photometric redshifts for galaxies and AGNs. While they are relatively accurate for galaxies, their accuracies for AGNs are a strong function of the luminosity ratio between the AGN and galaxy components. Somewhat surprisingly, the relative luminosities of the AGN and its host are well determined even when the photometric redshift is significantly in error. We also use our templates to study the mid-IR AGN selection criteria developed by Stern et al.(2005) and Lacy et al.(2004). We find that the Stern et al.(2005) criteria suffers from significant incompleteness when there is a strong host galaxy component and at z =~ 4.5, when the broad Halpha emission line is redshifted into the [3.6] band, but that it is little contaminated by low and intermediate redshift galaxies. The Lacy et al.(2004) criterion is not affected by incompleteness at z =~ 4.5 and is somewhat less affected by strong galaxy host components, but is heavily contaminated by low redshift star forming galaxies. Finally, we use our templates to predict the color-color distribution of sources in the upcoming WISE mission and define a color criterion to select AGNs analogous to those developed for IRAC photometry. We estimate that in between 640,000 and 1,700,000 AGNs will be identified by these criteria, but will have serious completeness problems for z &gt;~ 3.4.

LOTUS (non-LTE Optimization Tool Utilized for the derivation of atmospheric Stellar parameters) derives stellar parameters via Equivalent Width (EW) method with the assumption of 1D non-local thermodynamic equilibrium. It mainly applies on the spectroscopic data from high resolution spectral survey. It can provide extremely accurate measurement of stellar parameters compared with non-spectroscopic analysis from benchmark stars. LOTUS provides a fast optimizer for obtaining stellar parameters based on Differential Evolution algorithm, well constrained uncertainty of derived stellar parameters from slice-sampling MCMC from PyMC3 (ascl:1610.016), and can interpolate the Curve of Growth from theoretical EW grid under the assumptions of LTE and Non-LTE. It also visualizes excitation and ionization balance when at the optimal combination of stellar parameters.

LOSSCONE computes the rates of capture of stars by supermassive black holes. It uses a stationary and time-dependent solutions for the Fokker-Planck equation describing the evolution of the distribution function of stars due to two-body relaxation, and works for arbitrary spherical and axisymmetric galactic models that are provided by the user in the form of M(r), the cumulative mass as a function of radius.

LOSP is a FORTRAN77 numerical package that computes the orbital parameters of spectroscopic binaries. The package deals with SB1 and SB2 systems and is able to adjust either circular or eccentric orbits through a weighted fit.

LoSoTo (LOFAR Solution Tool) performs a variety of operations on H5parm data, which is based on the HDF5 format; it isolates direction independent systematic effects and can therefore be transferred to the target field. Subsets of data can be selected for each operation using lists of axes values, regular expressions, or intervals. The LoSoTo package stores solutions in arrays organized in a hierarchical fashion; this provides flexibility and preserves performance. The code can, for example, extract Faraday rotation from RR/LL phase solutions or a rotation matrix, clip solutions around the median, and calculate the ionospheric structure function. LoSoTo includes an outlier flagging procedure, normalizes solutions to a given value, and offers an advanced plotting routine, and many other operations.

LORENE (Langage Objet pour la RElativité NumériquE) solves various problems arising in numerical relativity, and more generally in computational astrophysics. It is a set of C++ classes and provides tools to solve partial differential equations by means of multi-domain spectral methods. LORENE classes implement basic structures such as arrays and matrices, but also abstract mathematical objects, such as tensors, and astrophysical objects, such as stars and black holes.

LoRD (Locate Reconnection Distribution) identifies the locations and structures of 3D magnetic reconnection within discrete magnetic field data. The toolkit contains three main functions; the first, ARD (Analyze Reconnection Distribution) locates the grids undergoing reconnection without null points and also recognizes the local configurations of reconnection sites. ANP (Analyze Null Points) locates and classifies the 3D null points, and APNP (Analyze Projected Null Points) analyzes the 2D neutral points projected on a plane near a cell. LoRD is written in Matlab and the toolkit contains demo scripts.

LoLLiPoP is a Planck low-l polarization likelihood based on cross-power-spectra for which the bias is zero when the noise is uncorrelated between maps. It uses a modified approximation to apply to cross-power spectra and is interfaced with the Cobaya (ascl:1910.019) MCMC sampler. Cross-spectra are computed on the CMB maps from Commander component separation applied on each detset-split Planck frequency maps.

Lofti_gaia fits orbital parameters for one wide stellar binary relative to the other, when both objects are resolved in Gaia DR2. It takes as input only the Gaia DR2 source id of the two components, and their masses. It retrieves the relevant parameters from the Gaia archive, computes observational constraints for them, and fits orbital parameters to those measurements. It assumes the two components are bound in an elliptical orbit.

Calibration solutions for the LOFAR radio telescope are stored in a 5-dimensional (time, frequency, station, polarisation and direction in the sky) HDF5 table. H5plot is a GUI application focussing on interactive visual inspection of these calibration solutions.

Locus implements the Locus Algorithm, which maximizes the performance of differential photometry systems by optimizing the number and quality of reference stars in the Field of View with the target.

loci is a shared library for interpolations in up to 4 dimensions. It is written in C and can be used with C/C++, Python and others. In order to calculate the coefficients of the cubic polynom, only local values are used: The data itself and all combinations of first-order derivatives, i.e. in 2D f_x, f_y and f_xy. This is in contrast to splines, where the coefficients are not calculated using derivatives, but non-local data, which can lead to over-smoothing the result.

Lmfit provides a high-level interface to non-linear optimization and curve fitting problems for Python. Lmfit builds on and extends many of the optimization algorithm of scipy.optimize, especially the Levenberg-Marquardt method from optimize.leastsq. Its enhancements to optimization and data fitting problems include using Parameter objects instead of plain floats as variables, the ability to easily change fitting algorithms, and improved estimation of confidence intervals and curve-fitting with the Model class. Lmfit includes many pre-built models for common lineshapes.

LMC is a Markov Chain Monte Carlo engine in Python that implements adaptive Metropolis-Hastings and slice sampling, as well as the affine-invariant method of Goodman & Weare, in a flexible framework. It can be used for simple problems, but the main use case is problems where expensive likelihood evaluations are provided by less flexible third-party software, which benefit from parallelization across many nodes at the sampling level. The parallel/adaptive methods use communication through MPI, or alternatively by writing/reading files, and mostly follow the approaches pioneered by CosmoMC (ascl:1106.025).

LIZARD (Lagrangian Initialization of Zeldovich Amplitudes for Resimulations of Displacements) creates particle initial conditions for cosmological simulations using the Zel'dovich approximation for the matter and velocity power spectrum.

Lizard is an extensible Cyclomatic Complexity Analyzer for imperative programming languages including C/C++/C#, Python, Java, and Javascript. It counts the nloc (lines of code without comments) and CCN (cyclomatic complexity number), and takes a token count of functions and a parameter count of functions. It also does copy-paste detection (code clone detection/code duplicate detection) and many other forms of static code analysis. Lizard is often used in software-related research and calculates how complex the code looks rather than how complex the code really is; thought it's often very hard to get all the included folders and files right when they are complicated, that accuracy is not needed to determine cyclomatic complexity, which can be useful for measuring the maintainability of a software package.

LiveData is a multibeam single-dish data reduction system for bandpass calibration and gridding. It is used for processing Parkes multibeam and Mopra data.

LISACode is a simulator of the LISA mission. Its ambition is to achieve a new degree of sophistication allowing to map, as closely as possible, the impact of the different subsystems on the measurements. Its also a useful tool for generating realistic data including several kind of sources (Massive Black Hole binaries, EMRIs, cosmic string cusp, stochastic background, etc) and for preparing their analysis. It’s fully integrated to the Mock LISA Data Challenge. LISACode is not a detailed simulator at the engineering level but rather a tool whose purpose is to bridge the gap between the basic principles of LISA and a future, sophisticated end-to-end simulator.

The LIghtweight Source finding Algorithms (LiSA) library finds HI sources in next generation radio surveys. LiSA can analyze input data cubes of any size with pipelines that automatically decompose data into different domains for parallel distributed analysis. For source finding, the library contains python modules for wavelet denoising of 3D spatial and spectral data, and robust automatic source finding using null-hypothesis testing. The source-finding algorithms all have options to automatically choose parameters, minimizing the need for manual fine tuning. Finally, LiSA also contains neural network architectures for classification and characterization of 3D spectral data.

LIRA (Low-counts Image Reconstruction and Analysis) deconvolves any unknown sky components, provides a fully Poisson 'goodness-of-fit' for any best-fit model, and quantifies uncertainties on the existence and shape of unknown sky. It does this without resorting to χ2 or rebinning, which can lose high-resolution information. It is written in R and requires the FITSio package.

LIRA (LInear Regression in Astronomy) performs Bayesian linear regression that accounts for heteroscedastic errors in both the independent and the dependent variables, intrinsic scatters (in both variables), time evolution of slopes, normalization and scatters, Malmquist and Eddington bias, and break of linearity. The posterior distribution of the regression parameters is sampled with a Gibbs method exploiting the JAGS (ascl:1209.002) library.

lintsampler performs linear interpolant sampling to create a set of sample points from a density function. The code uses the evaluation of the density at the two endpoints of 1D interval, or the four corners of a 2D rectangle, or generally the 2^<i>k</i> vertices of a dimensional hyperbox (or a series of such hyperboxes, <i>e.g.</i>, the cells of a <i>k</i>-dimensional grid) to draw random samples within the hyperbox. lintsampler works by evaluating a given PDF on the nodes of a grid (or grid-like structure, such as a tree); the number of evaluations (and memory occupancy) grows exponentially with the number of dimensions.

LineProf implements a series of line-profile analysis indicators and evaluates its correlation with RV data. It receives as input a list of Cross-Correlation Functions and an optional list of associated RV. It evaluates the line-profile according to the indicators and compares it with the computed RV if no associated RV is provided, or with the provided RV otherwise.

linemake generates formatted and curated atomic and molecular line lists suitable for spectral synthesis work. It is lightweight and easy-to-use. The code requires that the requested beginning and ending wavelengths not bridge the divide between two files of atomic line data; in such cases, run the code twice, once on either side of the divide, to generate the desired lists.

Line-Stacker stacks both 3D cubes or already extracted spectra and is an extension of <a href="https://ascl.net/1912.019">Stacker</a> (ascl:1912.019). It is an ensemble of both CASA tasks and native python tasks. Line-Stacker supports image stacking and some additional tools, allowing further analysis of the stack product, are also included in the module.

The Python code line_selections reads synthetic "full" spectra and elemental spectra, automatically identifies the detectable lines at a given resolution (provided the linelist used to compute the spectra), and returns a table containing various properties of the lines (<i>e.g.</i>, purity, central wavelength, and depth). The code then stores the information in a pandas DataFrame. line_selections demonstrates where chemical information is present in a stellar spectrum, and allows the user to optimize observational strategies, such as choosing resolution and spectra windows, as well as analysis codes with the application of high-quality masks.

Detailed studies of exoplanet and brown dwarf atmospheres rely on precise knowledge of the spectral features of possible atmospheric species. These features are a result of the interaction of different molecules and atoms with the radiation of the host star or the intrinsic thermal radiation of the object. The interaction is described by the opacity of a species, which determines how much light is absorbed at a given wavelength. The strength and shape of these opacities are very dependent on the temperature and pressure of the atmosphere, which is why they have to be calculated for a wide range of conditions. line racer is a Python package that enables computing high-resolution opacities from large molecular line lists in an effective manner. It offers users a wide range of options to customize opacity calculations to their needs and available hardware. The code is designed for efficient parallelization on multi-core and multi-node systems and can produce outputs compatible with popular atmospheric modeling and retrieval tools.

LIMpy models and analyzes multi-line intensity maps of CII (158 µ), OIII (88 µ), and CO (1-0) to CO (13-12) transitions. It can be used as an analytic model for star formation rate, to simulate line intensity maps based on halo catalogs, and to calculate the power spectrum from simulated maps and the cross-correlated signal between two separate lines. Among other things, LIMpy can also create multi-line luminosity models and determine the multi-line intensity power spectrum.

LIMEPY solves distribution function (DF) based lowered isothermal models. It solves Poisson's equation used on input parameters and offers fast solutions for isotropic/anisotropic, single/multi-mass models, normalized DF values, density and velocity moments, projected properties, and generates discrete samples.

LIME solves the molecular and atomic excitation and radiation transfer problem in a molecular gas and predicting emergent spectra. The code works in arbitrary three dimensional geometry using unstructured Delaunay latices for the transport of photons. Various physical models can be used as input, ranging from analytical descriptions over tabulated models to SPH simulations. To generate the Delaunay grid we sample the input model randomly, but weigh the sample probability with the molecular density and other parameters, and thereby we obtain an average grid point separation that scales with the local opacity. Slow convergence of opaque models becomes traceable; when convergence between the level populations, the radiation field, and the point separation has been obtained, the grid is ray-traced to produced images that can readily be compared to observations. LIME is particularly well suited for modeling of ALMA data because of the high dynamic range in scales that can be resolved using this type of grid, and can furthermore deal with overlapping lines of multiple molecular and atomic species.

LimberJack.jl performs cosmological analyses of 2 point auto- and cross-correlation measurements from galaxy clustering, CMB lensing and weak lensing data. Written in Julia, it obtains gradients for its outputs faster than traditional finite difference methods, making the code greatly synergistic with gradient-based sampling methods such as Hamiltonian Monte Carlo. LimberJack.jl can efficiently exploring parameter spaces with hundreds of dimensions.

Limbdark.jl computes analytic transit light-curve models for stars with arbitrary order limb darkening. It provides functions and notebooks to evaluate light curves for specified limb-darkening profiles and transit geometries. The repository includes documentation and example notebooks.