lenstronomy release 1.11.1
Simon Birrer
Dear lenstronomy users,
We are happy to announce lenstronomy version 1.11.1 on PyPi and conda-forge, consistent with the current main branch of the github repository. This is a minor release with backwards compatibility, except the psf_error_map definition, and with many new features by many contributors!
Thank you very much for the direct and indirect contributions that make lenstronomy what it is!
We recommend updating your lenstronomy version to be compatible with further development. If you have questions, encounter bugs, have feature requests etc, please let us know and/or join the Slack channel.
Best,
Anowar, Daniel, Simon
Lenstronomy version 1.11.1 release notes:
* psf_error_map definition changed
The previous psf_error_map that was required as input into the PSF() class was defined such that the excess variance of the error is given by sigma2_psf(i,j) = psf_error_map(i,j) * ( * psf_kernel(i,j)^2). This is not an intuitive convention and in the construction of the input value, might lead to ill defined values for numbers where the psf_kernel is effectively zero.
This PR changes the convention to
sigma2_psf(i,j) = psf_error_map(i,j) * ^2.
The changes make the results NOT BACKWARDS COMPATIBLE. Any psf_error_map previously being used needs to be multiplied by the kernel_point_source^2.
* added JWST configurations
By Martin Millon
* minor change in Sersic light profile
Allows non-zero flux values out to 100 times the half-light radius
* simplified LensCosmo class
* NFW c-rho0 inversion extended in range
* added stretch_plot and shear_plot to lens_plot
By Matt Gomer
* minor bug fix for critical_curve_caustic
* enable the change of kwargs_sigma initial guess parameters in FittingSequence
* improve zeus and nautilus sampler implementations
By Natalie Hogg (zeus)
* added EPL_boxydisky lens profile
By Maverick Oh
* added primary beam to image simulation (for interferrometic data)
By Nan Zhang