lenstronomy release 0.6.0

[Simon Birrer]

Hi all,

this is to notify about the new lenstronomy release 0.6.0 available now through PyPi.

Frist, don’t worry, all your current use cases are supported. For certain specific use cases, you might need very minor changes. Below you find the release notes and the specific changes that you might be affected.

lenstronomy 0.6.0 provides full support to multi source plane (also in combination of multi lens plane) computation and modeling. There is a highly modernized SimulationAPI module in place that enables you to have a nice and well documented interface to let you create (realistic) (lens) simulations. Supported are also to patch together (real) images and let them play the role of being the lens light or the source light.

Let me know if you have questions, experiencing problems or would like to have a new feature implemented.

Cheers,
Simon

ps: check out the paper by Daniel Gilman:
https://ui.adsabs.harvard.edu//#abs/2019arXiv190111031G/abstract
There he perform ray-tracing computations through several hundreds (or even thousands) of line of sight halos and substructure with the full multi-plane routines at work. If you want to know who to use lenstronomy to do this kind of computations, get in touch with Daniel.

----------------------------------------------

release notes for lenstronomy 0.6.0

Changes:
- surface brightness amplitude conventions have changed. Previously the ‘amp’ parameter were given in counts per pixel. With the latest release the units are in counts per unit square angle (e.g. in square arcseconds). This allows for more straight forward conversion in absolute flux numbers without the knowledge of the pixel scale. Point source amplitudes were already previously handled this way. If you do not used the ‘amp’ parameters to compute fluxes, the change will not impact you.
- SimAPI.configure_data() is now available at lenstronomy.Util.simulation_util as configure_data_simple(). This indicates that this is a simplistic support routine to create keyword arguments used in the lenstronomy Data() class. You can also create your own (with more flexible use cases) Data() instances. Have a look at the lenstronomy.Data.image_data.py file or how in the SimulationAPI this instance is handled.
- SimAPI.configure_psf() is now available at lenstronomy.Util.simulation_util as configure_psf_simple(). This indicates that this is a simplistic support routine to create keyword arguments used in the lenstronomy PSF() class. You can also create your own (with more flexible use cases) PSF() instances. Have a look at the lenstronomy.Data.psf.py file or how in the SimulationAPI this instance is handled.

New Features:
- SimulationAPI Module:
- Notebook in lenstronomy_extension with a use case
- Modular build up of Instrument() and Observation() class instances that can be re-used and in combination of multi-exposure modelling
- automatic astrometric magnitude to lenstronomy brightness conventions. You can use the SimulationAPI by defining your sources in magnitude units conveniently.
- LightModel Module:
- routine total_flux() that enables you to compute the total flux from an object (in counts per unit time). Does support some but not all light profile functions.
- you can optionally define the redshifts of the different source light profiles to simulate compound lenses. If you are interested in that, have a look at the LightModel class or let me know.
- ImSimModule:
- generalized ray-tracing module that can combine multiple lens plane with multiple source plane. This can lead to pretty funny lensing features.