PyLightcurve-torch: a transit modelling package for deep learning applications in PyTorch

Abstract

We present a new open source python package, based on PyLightcurve and PyTorch, tailored for efficient computation and automatic differentiation of exoplanetary transits. The classes and functions implemented are fully vectorised, natively GPU-compatible and differentiable with respect to the stellar and planetary parameters. This makes PyLightcurve-torch suitable for traditional forward computation of transits, but also extends the range of possible applications with inference and optimisation algorithms requiring access to the gradients of the physical model. This endeavour is aimed at fostering the use of deep learning in exoplanets research, motivated by an ever increasing amount of stellar light curves data and various incentives for the improvement of detection and characterisation techniques.

arXiv comment7 pages, 3 figures; submission status updated, fig 2 caption added

References in corpus (14)

• Array Programming with NumPy
• TensorFlow: A system for large-scale machine learning
• The Transiting Exoplanet Survey Satellite
• Deep learning for time series classification: a review
• Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
• The WASP Project and the SuperWASP Cameras
• Deep Learning for Anomaly Detection: A Survey
• Wide-field millimagnitude photometry with HAT: A Tool for Extra-solar Planet Detection
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• The Next Generation Transit Survey (NGTS)
• Searching for Exoplanets Using Artificial Intelligence
• ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks
• The CARMENES search for exoplanets around M dwarfs -- A deep learning approach to determine fundamental parameters of target stars
• Detrending Exoplanetary Transit Light Curves with Long Short-Term Memory Networks

Cited by in corpus (3)

• Efficient Gravitational Wave Template Bank Generation with Differentiable Waveforms
• Unveiling the frontiers of deep learning: innovations shaping diverse domains
• Reconstructing Atmospheric Parameters of Exoplanets Using Deep Learning