GMENoiseReduce
Python implementation of the Generalized Maximum Entropy white noise elimination technique discussed in https://pubs.aip.org/aip/jap/article/132/7/074903/2837401/Eliminating-white-noise-in-spectra-A-generalized
Installation
-
Requires Numpy, no other dependencies
pip install GMENoiseReduce
Usage
from GMENoiseReduce import GME
x,y = data
smoothed-yvals = GME.smooth(x,y)
Advanced Usage
The full function takes in additional arguments if the curve is not ideal
smoothed-yvals = GME.smooth(x,y, int order, int noise_threshold, int offset)
Despite this method needing zero information about the original system, the solutions provided are currently not always stable.
- order : The order of the CME (Corrected Maximum Entropy) calculations, defaults to 22
- noise_threshold : The white noise coefficient cut-off, defaults to 10
- offset : The empirical offset to the R-matrix zero coefficient, defaults to 2