Few python scripts for numerical results processing (libs: numpy, sklearn, scipy, matplotlib, TensorFlow v1)
py_scripts/IExpAllN.py
collecting local speckle maximums, parallel screens processing Writing values and its coordinates to csv file
py_scripts/collect_iexp.py
Processing obtained csvs, then for regions of interest we collect statistics from all random seeds and calculate histograms of local maximums
py_scripts/gpu_akde.py
To increase accuracy of histogram we use modified adaptive kernel density estimation (akde) of probability density function (PDF) by gaussian kernels. To speed up this operation TensorFlow library is used
py_scripts/profile_avg_big_mean.py
Script for calculating the dependence of the average intensity on the radius
py_scripts/profile_avg_8192.py
Script for PDF calculation in axial rings, angle averaging
jupyter files
fit_law.ipynb
Few scales comparison of obtained data and its approximations and fits
read_dict.ipynb
Transform data obtained from HPC server calculations to json-like storage
iexp.ipynb
Further HPC data processing