Upgrade Steps
- The requirements can be seen in the file requirements.txt. In addition, you will need pyOpt. This is not written in the requirement as python would install pyopt instead. To install pyOpt, you may use the following command:
pip install git+https://github.com/madebr/pyOpt - Linux users have the possibility to create a virtual environment to install the packages. This ensure package compatibility.
Improvements
New graphical user interface.
- It is possible to preview the result before launching the optimization. This makes errors easier to spot.
- For more clarity, everything happens in a single window with several tabs which represent the steps (initialisation, parameters, optimization). There is also a dedicated space for progression and error display.
- It is possible to go back and change parameters and files: the user can correct errors or try different parameters without launching the software several times.
New Features
- Debye oscillator model is implemented.
- It is possible to take into account the scattering (Rayleigh)
- The user can change change optimization parameters (swarmsize and number of iterations). They also get an output file with informations about the optimization and the history of parameters values.