README.md

Create 3D memory efficient Fourier Transformations with Python to analyze scattering data.

1. SAXSsimulations folder contains:
   • files to simulate 3D density of sphere/ hard sphere / cylinder in the unit box
   • read-in some density containing file
   • function to fast fourier transform a 3D density structure with different methods
   • rebinning
   • comparing simulation to SasView simulation
   • example of usage is in the main directory -> creating_forms.ipynb
   • example how to read in a density file -> real_density.ipynb
2. ML_model folder contains machine learning models tested in the thesis (requires FrEIA):
   • utils.py file contains the class wrapper to read data, normalize it, train etc.
   • losses.py contains definitions of the loss functions
   • monitoring.py contains files to output loss decay information
   • visualizations.py contains methods to visualize the result of prediction
   • Notebooks with respective architectures and their training results
3. tests contains scripts and csv files to reproduce the results section of the thesis with all the necessary plotting functions
4. create_data.py creates training set with SasView
5. envoronment file