Machine Learning

Comparison of PSF reconstruction showing a cutout of a bright point source and four recovery methods at 4x resolution as well as their residuals.

📄 Pixellated Posterior Sampling of Point Spread Functions in Astronomical Images

A framework for fully probabilistic, pixel-level inference of the point spread function directly from astronomical images, enabling principled uncertainty quantification in downstream photometric analyses.

📝 Simulation Based Inference

An introduction to likelihood-free inference: what it is, why you’d use it, and how the four main approaches (NLE, NPE, NJE, NRE) work.

A multi-band, multi-object fitting example with AstroPhot. Fitting seven bands from UV to IR, showing the data, model, and residuals.

💻 AstroPhot: Fitting Everything Everywhere All at Once in Astronomical Images

AstroPhot is a Python package for simultaneous multi-band, multi-object photometric fitting of astronomical images using automatic differentiation and GPU acceleration.

AstroPhot

AstroPhot is a Python package for simultaneous multi-band, multi-object photometric fitting of astronomical images using automatic differentiation and GPU acceleration.