Welcome to dipspeaks’s documentation!

A brief introduction

X-ray binaries are truly fascinating! In these extreme environments, a compact object—either a neutron star or black hole—draws in matter from a companion star, producing intense X-ray emissions. These systems offer a unique window into extreme physics, from the effects of strong gravity and relativistic jets to the presence of intense magnetic fields.

The emitted light travels from the source to the telescope, and during this path trespasses over dense-rarified areas, so-called clumps and the circumstellar environment. The accretion processes are not homogeneous. Some spikes/flares/peaks observed in some light curves were associated to Raighleight-Taylor inestabilities. All these phenomena leaves a footprint in the x-rays lirght curves, in the form of peaks and dips (features).

The main inconvenient is distinguish real features, caused by astrophysical phenomena from random noise. With peaksdips we use an autoencoder approach to flag those events, giving each detection a probability. We also provide some useful analysis tools such as classify the resulting features by prominence and duration using gaussian mixture or calculate overlaps, if we perform the analysis in different energy ranges.

Getting started

Installation

You can install the package directly from PyPI:

pip install dipspeaks

Rationale:

• How dipspeaks builds its synthetic (“noise-only”) light curves
• How the auto-encoder scores dips & peaks
• Probability based on an excess

Example Notebooks:

• Cen X-3 Analysis
• Behaviour of dipspeaks package in a Fake Light Curve

Functions:

• dipspeaks – Key Public Functions

Contribute:

• Contributing to dipspeaks