The atmospheric Mass Loss INquiry frameworK (MLink) code performs high-quality interpolation of mass-loss rate values across the grid of upper atmosphere hydrodynamic models published by Kubyhskina et al. (2018) and Kubyshkina & Fossati (2021). MLink uses a Dense Neural Network scheme. The neural network has been trained using the TensorFlow library and saved in the HDF5 format, which is compatible with TensorFlow’s load_model function. Users can load the pre-trained model into their environment to predict outcomes on new datasets with minimal setup. This functionality is included in both command-line and plug-in versions, ensuring users can leverage the predictive capabilities of the model in their preferred workflow.
MLink is available at https://github.com/amitrezaiwf/Mass-loss-rate. See Reza et al. (2025) for further details.
MC3 is a powerful Python Bayesian-statistics tool to perform Levenberg-Marquardt least-squares optimization and Markov-chain Monte Carlo (MCMC) posterior-distribution sampling. MC3 runs from the Shell prompt or through the Python interpreter, supports non-informative or Gaussian priors, and provides correlated-noise estimation with the Time-averaging or the Wavelet-based Likelihood methods.
Find the full MC3 documentation at https://github.com/pcubillos/mc3. See Cubillos et al. (2016) for further details.
This code derives the correction to the stellar activity parameters S and logR' caused by absorption from the interstellar medium.
The code is available in both IDL and Python here. See Fossati et al., submitted for further details.
The Python Radiative Transfer in a Bayesian framework (Pyrat Bay) provides tools for opacity line sampling, planetary atmospheric modeling, spectral synthesis, and Bayesian spectral retrieval of exoplanet atmospheres. Atmospheric profiles can be calculated either in a self-consistent manner assuming radiative and chemical equilibrium, or for retrievals adopting parametric profiles of the temperature and composition (in equilibrium or free chemistry). Pyrat Bay can compute spectra in transmission or emission geometry, thus being suitable for the modeling of transit, eclipse, and phase-curve time series observations.
The Pyrat Bay is an open-source package under a GPL-v2 license, documented athttps://pyratbay.readthedocs.io. See Cubillos & Blecic (2021) for further details.
The Gen TSO package enables the estimation of signal-to-noise ratios for transit and eclipse spectroscopic observations with the James Webb Space Telescope (JWST). Gen TSO provides an interactive graphical interface, that leverages the JWST ETC by combining its noise simulator, Pandeia, with additional exoplanet resources from the NASA Exoplanet Archive, the Gaia DR3 catalog, and the TrExoLiSTS database of JWST programs. Gen TSO allows users to calculate S/Ns for all JWST instruments for the spectroscopic time-series modes currently available. It also allows users to simulate target acquisition on the science targets or, when needed, on nearby stellar targets.
Gen TSO code is documented at https://pcubillos.github.io/gen_tso. See Cubillos (2024) for further details.
The Repack Python package (Cubillos 2017) compresses line-by-line transition opacity databases from Exomol, HITEMP, and Kurucz TiO, enabling faster radiative-transfer computations without loss of information. To do so, Repack preserves the full LBL information of only the strong lines that dominate the spectrum, and compresses the opacity from the weak lines into a continuum opacity. The compressed dataset reproduces ~99% of the original opacity, taking into considertion the temperature dependence of the opacity.
The Repack code is available at https://github.com/pcubillos/repack.
