We introduce a new, open-source, Python module for the acquisition and processing of archival data from many X-ray telescopes - Democratising Archival X-ray Astronomy (hereafter referred to as DAXA). Our software is built to increase access to, and use of, large archives of X-ray astronomy data; providing a unified, easy-to-use, Python interface to the disparate archives and processing tools. We provide this interface for the majority of X-ray telescopes launched within the last 30 years. This module enables much greater access to X-ray data for non-specialists, while preserving low-level control of processing for X-ray experts. It is useful for identifying relevant observations of a single object of interest but it excels at creating multi-mission datasets for serendipitous or targeted studies of large samples of X-ray emitting objects. The management and organization of datasets is also made easier; DAXA archives can be version controlled and updated if new data become available. Once relevant observations are identified, the raw data can be downloaded (and optionally processed) through DAXA, or pre-processed event lists, images, and exposure maps can be downloaded if they are available. X-ray observations are perfectly suited to serendipitous discoveries and archival analyses, and with a decade-long `X-ray desert' potentially on the horizon archival data will take on even greater importance; enhanced access to those archives will be vital to the continuation of X-ray astronomy.

pysymlog provides utilities for binning, normalizing colors, wrangling tick marks, and other tasks, in symmetric logarithm space. For numbers spanning positive and negative values, the code works in log scale with a transition through zero, down to some threshold. This is useful for representing data that span many scales such as standard log-space that include values of zero or even negative values. pysymlog provides convenient functions for creating 1D and 2D histograms and symmetric log bins, generating logspace-like arrays through zero and managing matplotlib major and minor ticks in symlog space, as well as bringing symmetric log scaling functionality to plotly.

This paper introduces RadioSunPy, an open-source Python package developed for accessing, visualizing, and analyzing multi-band radio observations of the Sun from the RATAN-600 solar complex. The advancement of observational technologies and software for processing and visualizing spectro-polarimetric microwave data obtained with the RATAN-600 radio telescope opens new opportunities for studying the physical characteristics of solar plasma at the levels of the chromosphere and corona. These levels remain some difficult to detect in the ultraviolet and X-ray ranges. The development of these methods allows for more precise investigation of the fine structure and dynamics of the solar atmosphere, thereby deepening our understanding of the processes occurring in these layers. The obtained data also can be utilized for diagnosing solar plasma and forecasting solar activity. However, using RATAN-600 data requires extensive data processing and familiarity with the RATAN-600. The package offers comprehensive data processing functionalities, including direct access to raw data, essential processing steps such as calibration and quiet Sun normalization, and tools for analyzing solar activity. This includes automatic detection of local sources, identifying them with NOAA (National Oceanic and Atmospheric Administration) active regions, and further determining parameters for local sources and active regions. By streamlining data processing workflows, RadioSunPy enables researchers to investigate the fine structure and dynamics of the solar atmosphere more efficiently, contributing to advancements in solar physics and space weather forecasting.

ysoisochrone is a Python3 package that handles the isochrones for young stellar objects (YSOs), and utilize isochrones to derive the stellar mass and ages. Our primary method is a Bayesian inference approach, and the Python code builds on the IDL version developed in Pascucci et al. (2016). The code estimates the stellar masses, ages, and associated uncertainties by comparing their stellar effective temperature, bolometric luminosity, and their uncertainties with different stellar evolutionary models, including those specifically developed for YSOs. User-developed evolutionary tracks can also be utilized when provided in the specific format described in the code documentation.

The kete tools are intended to enable the simulation of all-sky surveys of solar system objects. This includes multi-body physics orbital dynamics, thermal and optical modeling of the objects, as well as field of view and light delay corrections. These tools in conjunction with the Minor Planet Centers (MPC) database of known asteroids can be used to not only plan surveys but can also be used to predict what objects are visible for existing or past surveys.

The primary goal for kete is to enable a set of tools that can operate on the entire MPC catalog at once, without having to do queries on specific objects. It has been used to simulate over 10 years of survey time for the NEO Surveyor mission using 10 million main-belt and near-Earth asteroids.

Astrophysical turbulent flows display an intrinsically multi-scale nature, making their numerical simulation and the subsequent analyses of simulated data a complex problem. In particular, two fundamental steps in the study of turbulent velocity fields are the Helmholtz-Hodge decomposition (compressive+solenoidal; HHD) and the Reynolds decomposition (bulk+turbulent; RD). These problems are relatively simple to perform numerically for uniformly-sampled data, such as the one emerging from Eulerian, fix-grid simulations; but their computation is remarkably more complex in the case of non-uniformly sampled data, such as the one stemming from particle-based or meshless simulations. In this paper, we describe, implement and test vortex-p, a publicly available tool evolved from the vortex code, to perform both these decompositions upon the velocity fields of particle-based simulations, either from smoothed particle hydrodynamics (SPH), moving-mesh or meshless codes. The algorithm relies on the creation of an ad-hoc adaptive mesh refinement (AMR) set of grids, on which the input velocity field is represented. HHD is then addressed by means of elliptic solvers, while for the RD we adapt an iterative, multi-scale filter. We perform a series of idealised tests to assess the accuracy, convergence and scaling of the code. Finally, we present some applications of the code to various SPH and meshless finite-mass (MFM) simulations of galaxy clusters performed with OpenGadget3, with different resolutions and physics, to showcase the capabilities of the code.

GalCraft creates mock integral-field spectroscopic (IFS) observations of the Milky Way and other hydrodynamical/N-body simulations. It conducts all the procedures from inputting data and spectral templates to the output of IFS data cubes in FITS format. The produced mock data cubes can be analyzed in the same way as real IFS observations by many methods, particularly codes like Voronoi binning (ascl:1211.006), pPXF (ascl:1210.002), line-strength indices, or a combination of them (e.g., the GIST pipeline, ascl:1907.025). The code is implemented using Python-native parallelization. GalCraft will be particularly useful for directly comparing the Milky Way with other MW-like galaxies in terms of kinematics and stellar population parameters and ultimately linking the Galactic and extragalactic to study galaxy evolution.

pyRRG measures the 2nd and 4th order moments using a TinyTim model to correct for PSF distortions. The code is invariant to the number exposures and orientation of the drizzle images. pyRRG uses a machine learning algorithm to automatically classify stars and galaxies; this can also be done manually if greater accuracy is needed.

Padé simulates protoplanetary disk hydrodynamics in cylindrical coordinates. Written in Fortran90, it is a finite-difference code and the compact 4th-order standard Padé scheme is used for spatial differencing. Padé differentiation is known to have spectral-like resolving power. The z direction can be periodic or non-periodic. The 4th order Runge-Kutta is used for time advancement. Padé implements a version of the FARGO technique to eliminate the time-step restriction imposed by Keplerian advection, and capturing of shocks that are not too strong can be done by using artificial bulk viscosity.

PySR performs Symbolic Regression; it uses machine learning to find an interpretable symbolic expression that optimizes some objective. Over a period of several years, PySR has been engineered from the ground up to be (1) as high-performance as possible, (2) as configurable as possible, and (3) easy to use. PySR is developed alongside the Julia library SymbolicRegression.jl, which forms the powerful search engine of PySR. Symbolic regression works best on low-dimensional datasets, but one can also extend these approaches to higher-dimensional spaces by using "Symbolic Distillation" of Neural Networks. Here, one essentially uses symbolic regression to convert a neural net to an analytic equation. Thus, these tools simultaneously present an explicit and powerful way to interpret deep neural networks.

WISE2MBH uses infrared cataloged data from the Wide-field Infrared Survey Explorer (WISE) to estimate the mass of supermassive black holes (SMBH). It implements a Monte Carlo approach for error propagation, considering mean photometric errors from WISE magnitudes, errors in fits of scaling relations used and scatter of those relations, if available.

PyExoCross, a Python adaptation of ExoCross (ascl:1803.014), post-processes molecular line lists generated by ExoMol, HITRAN, and HITEMP and other similar initiatives. It generates absorption and emission spectra and other properties, including partition functions, specific heats, and cooling functions, based on molecular line lists. The code also calculates cross sections with four line profiles: Doppler, Gaussian, Lorentzian, and Voigt. PyExoCross can convert data format between ExoMol and HITRAN, and supports importing and exporting line lists in the ExoMol and HITRAN/HITEMP formats.

GASTLI (GAS gianT modeL for Interiors) calculates the interior structure models for gas giants exoplanets. The code computes mass-radius curves, thermal evolution curves, and interior composition retrievals to fit a interior structure model to your mass, radius, age, and if available, atmospheric metallicity data. GASTLI can also plot the results, including internal and atmospheric profiles, a pressure-temperature diagram, mass-radius relations, and thermal evolution curves.

symbolic_pofk provides simple Python functions and a Fortran90 routine for precise symbolic emulations of the linear and non-linear matter power spectra and for the conversion σ 8 ↔ A s as a function of cosmology. These can be easily copied, pasted, and modified to other languages. Outside of a tested k range, the fit includes baryons by default; however, this can be switched off.

planetMagFields accesses and analyzes information about magnetic fields of planets in our solar system and visualizes them in both 2D and 3D. The code provides access to properties of a planet, such as dipole tilt, Gauss coefficients, and computed radial magnetic field at surface, and has methods to plot the field and write a vts file for 3D visualization. planetMagFields can be used to produce both 2D and 3D visualizations of a planetary field; it also provides the option of potential extrapolation.

The software framework AMReX is designed for building massively parallel block-structured adaptive mesh refinement (AMR) applications. Key features of AMReX include C++ and Fortran interfaces; 1-, 2- and 3-D support; and support for cell-centered, face-centered, edge-centered, and nodal data. The framework also supports hyperbolic, parabolic, and elliptic solves on hierarchical adaptive grid structure, optional subcycling in time for time-dependent PDEs, and parallelization via flat MPI, OpenMP, hybrid MPI/OpenMP, or MPI/MPI, and parallel I/O. AMReX supports the plotfile format with AmrVis, VisIt (ascl:1103.007), ParaView (ascl:1103.014), and yt (ascl:1011.022).

ClassiPyGRB downloads, processes, visualizes, and classifies GRBs in the Swift/BAT database. Users can query light curves for any GRB and use tools to preprocess the data, including noise/duration reduction and interpolation. The package provides a set of facilities and tutorials for classifying GRBs based on their light curves using a method based on a dimensionality reduction of the data using t-Distributed Stochastic Neighbour Embedding (TSNE); results are visualized using a Graphical User Interface (GUI). ClassiPyGRB also plots and animates the results of the TSNE analysis for a deeper hyperparameter grid search.

BeyonCE (Beyond Common Eclipsers) explores the large parameter space of eclipsing disc systems. The fitting code reduces the parameter space encompassed by the transit of circumsecondary disc (CSD) systems with azimuthally symmetric, non-uniform optical-depth profiles to constrain the size and orientation of discs with a complex sub-structure. BeyonCE does this by rejecting disc geometries that do not reproduce the measured gradients within their light curves.

resonances identifies mean-motion resonances of small bodies. It uses the REBOUND integrator (ascl:1110.016) and automatically identifies two-body and three-body mean-motion resonance in the Solar system. The package can be used for other possible planetary systems, including exoplanets. resonances accurately differentiates different types of resonances (pure, transient, uncertain) and provides an interface for mass tasks, such as finding resonant areas in a planetary system. The software can also plot time series and periodograms.

cloudyfsps is a Python interface between FSPS (ascl:1010.043) and Cloudy (ascl:9910.001). It compiles FSPS models for use as ionizing sources (Stellar SED grids) within Cloudy and generates Cloudy input files, single-parameter or grids of parameters. It runs Cloudy models in parallel and formats the output, which is nebular continuum and nebular line emission, for FSPS input and for explorative manipulation and plotting within Python. cloudyfsps includes pre-packaged plots for BPT diagrams (NII, SII, OI, OII) with observed data from HII regions and SDSS galaxies, and also provides comparisons with MAPPINGS III (ascl:1306.008) models.

Stardust extracts galaxy properties by fitting their multiwavelength data to a set of linearly combined templates. This Python package brings three different families of templates together: 1.) UV+Optical emission from dust unobscured stellar light; 2.) AGN heated dust in the MIR; and 3.) IR dust reprocessed stellar light in the NIR-FIR. Stardust's template fitting does not rely on energy balance. As a result, the total luminosity of dust obscured and dust unobscured stellar light do not rely on each other, and it is possible to fit objects such as SMGs where the energy balance approach might not be applicable.

PICASSO (Python Inpainter for Cosmological and AStrophysical SOurces) provides a suite of inpainting methodologies to reconstruct holes on images (128x128 pixels) extracted from a HEALPIX map. Three inpainting techniques are included; these are divided into two main groups: diffusive-based methods (Nearest-Neighbors), and learning-based methods that rely on training DCNNs to fill the missing pixels with the predictions learned from a training data-set (Deep-Prior and Generative Adversarial Networks). PICASSO also provides scripts for projecting from full sky HEALPIX maps to flat thumbnails images, performing inpainting on GPUs and parallel inpainting on multiple processes, and for projecting from flat images to HEALPIX. Pretrained models are also included.

MCMole3D (Monte-Carlo MOlecular Line Emission) simulates the 3D molecular cloud emission in the Milky Way. In particular, it can simulate both the unpolarized and polarized emission coming from the first rotational line of Carbon Monoxide (CO, J=1-0). MCMole3D seeks to compare the simulated emission with that observed by full sky surveys from the Planck satellite.

FGCluster runs spectral clustering onto Healpix maps for parametric foreground removal, using a map encoding the feature to cluster as inputs. Pixel similarity is given by the geometrical affinity of each pixel in the sphere. FGCluster can also take an uncertainty map as an input, in which case the adjacency is modified in such a way that the pixel similarity accounts also for the statistical significance given by the pixel values in a map and the uncertainties.

SUSHI (Semi-blind Unmixing with Sparsity for hyperspectral images) performs non-stationary unmixing of hyperspectral images. The typical use case is to map the physical parameters such as temperature and redshift from a model with multiple components using data from hyperspectral images. Applying a spatial regularization provides more robust results on voxels with low signal to noise ratio. The code has been used on X-ray astronomy but the method can be applied to any integral field unit (IFU) data cubes.

UltraDark.jl simulates cosmological scalar fields. Written in Julia, it is inspired by PyUltraLight (ascl:1810.009) and designed to be simple to use and extend. It solves a non-interacting scalar field Gross-Pitaevskii equation coupled to Poisson's equation for gravitational potential. The scalar field describes scalar dark matter in models including ultralight dark matter, fuzzy dark matter, axion-like particles and the like. It also describes an inflaton field in the reheating epoch of the early universe.

Written in Python, DarsakX is used to design and analyze the imaging performance of a multi-shell X-ray telescope with an optical configuration similar to Wolter-1 optics for astronomical purposes. It can also assess the impact of figure error on the telescope's imaging performance and optimize the optical design to improve angular resolution for wide-field telescopes. By default, DarsakX uses DarpanX (ascl:2101.015) to calculate the mirror's reflectivity.

SAQQARA analyzes stochastic gravitational wave background signals. This Simulation-based Inference (SBI) library is built on top of the swyft code (ascl:2302.016), which implements neural ratio estimation to efficiently access marginal posteriors for all parameters of interest. Simulation-based inference combined with implicit marginalization (over nuisance parameters) has been shown to be well suited for SGWB data analysis.

21cmFirstCLASS extends 21cmFAST (ascl:1102.023) and interfaces with CLASS (ascl:1106.020) to generate initial conditions at recombination that are consistent with the input cosmological model. These initial conditions can be set during the time of recombination, allowing one to compute the 21cm signal (and its spatial fluctuations) throughout the dark ages, as well as in the proceeding cosmic dawn and reionization epochs, just as in the standard 21cmFAST. 21cmFirstCLASS tracks both the CDM density field δ_c as well as the baryons density field δ_b. In addition, the user interface in 21cmFirstCLASS has been improved and allows one to easily plot the 21cm power spectrum while including noise from the output of 21cmSense (ascl:1609.013).

GRBoondi simulates generalized Proca fields on arbitrary analytic fixed backgrounds; it is based on the publicly available 3+1D numerical relativity code GRChombo (ascl:2306.039). GRBoondi reduces the prerequisite knowledge of numerical relativity and GRChombo in the numerical studies of generalized Proca theories. The main steps to perform a study are inputting the additions to the equations of motion beyond the base Proca theory; GRBoondi can then automatically incorporate the higher-order terms in the simulation. The code is written entirely in C++14 and uses hybrid MPI/OpenMP parallelism. GRBoondi inherits all of the capabilities of the main GRChombo code, which makes use of the Chombo library (ascl:1202.008) for adaptive mesh refinement.

RadioSED uses nested sampling to perform a Bayesian analysis of radio SEDs constructed from radio flux density measurements obtained as part of large area surveys (or in some limited cases, as part of targeted followup campaigns). It is a pure Python implementation, and is essentially a wrapper around Bilby (ascl:1901.011), the Bayesian inference library. RadioSED uses dynesty (ascl:1809.013) to perform the sampling steps, though other samplers could also be used. Users can make use of a pre-defined set of models and surveys from which to draw flux density measurements, or they can define their own models and provide their own input flux density measurements. All flux density measurements are referenced against the RACS-LOW survey, and source names and IDs from the survey catalogue are used as identifiers.

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.

BELTCROSS2 calculates the closest approaches of asteroid to the mean orbits of meteoroid streams. It is especially useful to check if an asteroid, which was observed to become active, passed through a meteoroid stream, and through which stream, a short time before the beginning of the activity. The basic characteristics of the closest encounter of the asteroid with the stream are provided by BELTCROSS2.

Cue interprets nebular emission across a wide range of ionizing conditions of galaxies. The software, based on Cloudy (ascl:9910.001), emulates a neural net. It does not require a specific ionizing spectrum as a source, instead approximating the ionizing spectrum with a 4-part piece-wise power-law. Along with the flexible ionizing spectra, Cue allows freedom in [O/H], [N/O], [C/O], gas density, and total ionizing photon budget.

HaloFlow uses a machine learning approach to infer Mh and stellar mass, M∗, using grizy band magnitudes, morphological properties quantifying characteristic size, concentration, and asymmetry, total measured satellite luminosity, and number of satellites.

LADDER (Learning Algorithm for Deep Distance Estimation and Reconstruction) reconstructs the “cosmic distance ladder” by analyzing sequential cosmological data; it can also be applied to other sequential datasets with associated covariance information. It uses the apparent magnitude data from the Pantheon Type Ia supernovae compilation, fully incorporating covariance information to accurately predict mean values and uncertainties. It offers model-independent consistency checks for datasets such as Baryon Acoustic Oscillations (BAO) and can calibrate high-redshift datasets such as Gamma Ray Bursts (GRBs) without assuming any underlying cosmological model. Additionally, LADDER serves as a model-independent mock catalog generator for forecast-based cosmological studies.

Sonification extends the Astronify software (ascl:2408.005) to sonify a spatially distributed dataset. The package contains scripts to convert images into scatterplots and sonifications. The reproduce_image.py script takes an image file and reproduces it as a scatterplot by converting the input image to grayscale, extracting pixel values and generating scatter data based on these values, and then plotting the scatter data to create a visual representation of the image. The sonifications script converts the scatterplot data into an audio series and adjusts the note spacing and sonification range to customize an auditory representation. Sonification accepts images in PNG and JPG formats.

Astronify contains tools for sonifying astronomical data, specifically data series. Data series sonification takes a data table and maps one column to time, and one column to pitch. This technique is commonly used to sonify light curves, where observation time is scaled to listening time and flux is mapped to pitch. While Astronify’s sonification uses the columns “time” and “flux” by default, any two columns can be supplied and a sonification created.

Sailfish simulates accreting binary systems, including binary protostars, post-AGN stellar binaries, mass-transferring X-ray binaries, and double black hole systems. The binary components are "on the grid" rather than excised, and are evolved according to the Kepler two-body problem, modified to account for gravitational wave losses or self-consistent forcing from the orbiting gas. The solvers are shock-capturing and are second order accurate in space and time. Gravity is fully Newtonian. Thermodynamics can be treated using a gamma-law equation of state with a blackbody cooling term, or in the locally isothermal approximation, in which the gas temperature is set to a constant times the local free-fall speed. Sailfish is fully Cartesian and has extensive diagnostic capabilities to facilitate accurate calculations of gas-driven orbital evolution or the extraction of electromagnetic disk signatures. The code is extremely efficient, reaching more than one billion zone updates per second on an NVIDIA A100 GPU, enabling extremely high resolution of complex flows around the binary components.

Global mm-VLBI Array (GMVA) observations are accompanied by a lot of metadata (i.e., the so-called 'ANTAB' files) that contain the system temperature (Tsys) and the gain values of the individual GMVA antennas. These data are required for the amplitude calibration of GMVA data which is an essential part in the data reduction. Unfortunately, Tsys measurements in the ANTAB files are not perfect and there are almost always erroneous values in some of the ANTAB files (particularly in the VLBA data). This could lead to incorrect results in the amplitude calibration and thus need to be corrected with proper data inspection/treatment. However, every GMVA station provides the ANTAB file in their own data format which makes the examination tricky. AntabGMVA was designed to resolve these issues and allows GMVA users to manage the GMVA ANTAB files easily and efficiently. Using AntabGMVA, one can perform extraction/inspection/visualization/correction of the Tsys data from the ANTAB files and finally generate one single ANTAB file which includes all the final products.

SHARC (SHArpened Dimensionality Reduction and Classification) performs local gradient clustering-based sharpened dimensionality reduction (SDR) using neural network projections and uses these projections to make classifications. The library also contains functions for finding the optimal SDR parameters and for consolidating classification results obtained through multiple classifiers. It requires pySDR (ascl:2408.002). SHARC provides accurate and physically insightful classification of astronomical objects based on their broadband colors.

pySDR performs local gradient clustering-based sharpened dimensionality reduction (SDR). The library uses the C++ LGCDR_v1 code as its backend.

Sharpened dimensionality reduction (SDR) sharpens original data before dimensionality reduction to create visually segregrated sample clusters. user-guided labeling. Each distinct cluster can then be labeled and used to further analyze an otherwise unlabeled data set. Written in C++, SDR scales well with large high-dimensional data.

Package‑X instantly solves one loop Feynman integrals in full generality. Written in Mathematica and extensively tested and adopted, the package computes dimensionally regulated one-loop integrals with up to four distinct propagators of arbitrarily high rank, calculates traces of Dirac matrices in d dimensions for closed fermion loops, or carries out Dirac algebra for open fermion lines. Package‑X also generates analytic results for any kinematic configuration (e.g., at zero external momentum or physical threshold) for real masses and external invariants, provides analytic expressions for UV-divergent, IR-divergent and finite parts either separately or all together, and computes discontinuities across cuts of one-loop integrals, among other tasks.

The High-Resolution Imaging and Spectroscopy of Exoplanets (HiRISE) instrument at the Very Large Telescope (VLT) combines the exoplanet imager SPHERE with the high-resolution spectrograph CRIRES using single-mode fibers. HiRISE has been designed to enable the characterization of known, directly-imaged planetary companions in the H band at a spectral resolution on the order of R = λ/∆λ = 140 000. The hipipe package is a custom python pipeline used to reduce the HiRISE data and produce high-level science products that can be used for astrophysical interpretation.

pony3d statistically identifies islands of contiguous emission inside a three-dimensional volume. The primary functionality is the rapid and reliable creation of masks for the deconvolution of radio interferometric radio spectral line emission. It has been designed to run on the output of the wsclean imager (ascl:1408.023) whereby the individual FITS image per frequency plane enables a high degree of parallelism, but can work on any image set providing this criterion is met. Single channel island rejection is offered, along with 3D mask dilation and boxcar averaging. pony3d is also a prototype source-finding and extraction tool.

The photGalIMF code calculates the evolution of stellar mass and luminosity for a galaxy model, based on the PARSEC stellar evolution model (ascl:1502.005). It requires input lists specifying the age, mass, metallicity, and initial mass function (IMF) of single stellar populations. These input parameters can be provided by the companion galaxy chemical simulation code GalIMF (ascl:1903.010), which generates realistic sets of inputs.

ELISA is a Python library designed for efficient spectral modeling and robust statistical inference. With user-friendly interface, ELISA streamlines the spectral analysis workflow.

The modeling framework of ELISA is flexible, allowing users to construct complex models by combining models of ELISA and XSPEC, as well as custom models. Parameters across different model components can also be linked. The models can be fitted to the spectral datasets using either Bayesian or maximum likelihood approaches. For Bayesian fitting, ELISA incorporates advanced Markov Chain Monte Carlo (MCMC) algorithms, including the No-U-Turn Sampler (NUTS), nested sampling, and affine-invariant ensemble sampling, to tackle the posterior sampling problem. For maximum likelihood estimation (MLE), ELISA includes two robust algorithms: the Levenberg-Marquardt algorithm and the Migrad algorithm from Minuit. The computation backend is based on Google's JAX, a high-performance numerical computing library, which can reduce the runtime for fitting procedures like MCMC, thereby enhancing the efficiency of analysis.

After fitting, goodness-of-fit assessment can be done with a single function call, which automatically conducts posterior predictive checks and leave-one-out cross-validation for Bayesian models, or parametric bootstrap for MLE. These methods offer greater accuracy and reliability than traditional fit-statistic/dof measures, and thus better model discovery capability. For comparing multiple candidate models, ELISA provides robust Bayesian tools such as the Widely Applicable Information Criterion (WAIC) and the Leave-One-Out Information Criterion (LOOIC), which are more reliable than AIC or BIC. Thanks to the object-oriented design, collecting the analysis results should be simple. ELISA also provide visualization tools to generate ready-for-publication figures.

ELISA is an open-source project and community contributions are welcome and greatly appreciated.

Heimdall uses direct, tree, and sub-band dedispersion algorithms on massively parallel computing architectures (GPUs) to speed up real-time detection of radio pulsar and other transient events.

Flash-X simulates physical phenomena in several scientific domains, primarily those involving compressible or incompressible reactive flows, using Eulerian adaptive mesh and particle techniques. It derives some of its solvers from and is a descendant of FLASH (ascl:1010.082). Flash-X has a new framework that relies on abstractions and asynchronous communications for performance portability across a range of heterogeneous hardware platforms, including exascale machines. It also includes new physics capabilities, such as the Spark general relativistic magnetohydrodynamics (GRMHD) solver, and supports interoperation with the AMReX mesh framework, the HYPRE linear solver package, and the Thornado neutrino radiation hydrodynamics package, among others.