Changelog
Version 0.13.0
Breaking Changes:
Moved math submodule into utils, to avoid name conflict with Python’s math module
Bug fixes:
Fix error in eigenvalue sorting (PropergatorEig and non-isotropic backmaterials)
Pin version of ipywidgets to keep plotly working
Version 0.12.0
Breaking Changes:
Renamed the conversion functions and added more
result.R and .T now return the reflectance/transmittance instead of the respective matrix, which can be accessed with .R_matrix/.T_matrix
Renamed PropagatorExpmScipy to PropagatorExpm
Removed Torch and Tensorflow Solvers
New:
Added a lot of documentation
Added a Bruggeman EMA Material
Support for transmissive Ellipsometry
Bug fixes:
Fix nan values in MaxwellGarnettEMA
Version 0.11.0
Adds reading functionality for NeXus files.
Version 0.10.1
The fitting module is not imported at top-level anymore. It has now to be imported by
elli.fitting.
Version 0.10.0
Dispersions are now addressed by their name only (instead of Dispersion…)
Dispersions are initialized with two distinguished set of parameters for parameters which are set once and parameters which may be set multiple times (for oscillators etc). They can be added by invoking the
addcommand on the respective class.There is a new factory class
DispersionFactoryto get a dispersion from it’s string name, i.e.DispersionFactory.get_dispersion(...)
Version 0.9.2
Fitting decorator buttons (fit, undo, redo)
Data export for decorators
Changed file loading to -pi, pi delta convention
Fixed Jones vector default
Included more documentation
Version 0.9.1
Automated build of Docker images
Benchmarking setup
Added more documentation
Renamed a lot of functions for PEP8 compliance
More bugfixes
Version 0.9.0
Adapted examples to the new codebase
Added a first batch of unit tests
Reimplemented inhomogeneous and mixed materials
Changed the sign convention of the E_rp vector
Improve fitting decorators
Added plotting and fitting for Mueller matrices
Added type hints
Various calculation bugfixes
Version 0.1.0
Initial Version after un-forking
Complete rewrite of module structure
Vectorization of calculations for massive speed-up
Packaging as PyPi package
Added additional dispersion models
Added helper utilities for data-handling