uncertainty-framework

Framework for propagating uncertainties through V-FOR-WaTer

Install

pip install uncertainty-framework

Examples

The examples are implemeted as a python module. The main function imports fire to have a nice command line interface. Currently there is only one example.

Variogram estimation

This example illustrates, how the MonteCarlo simulation can be used to simulate measurement uncertainties on a variogram estimation. It replaces the observations by a gaussian distribution around the observation using a scale of 5. The observation value range is [0, 256[. The uncertainty_framework.examples.variograms example has additional dependencies that need to be installed separatly:

pip install scikit-gstat plotly

Then, you can run it through the command line. It is recommended to decrease the default number of iterations for this example.

python -m uncertainty_framework.examples.variograms --num-iter=500 --verbose

It is possbile to increase the used scale for generating new observations and also some of the variogram parameters are exposed:

python -m uncertainty_framework.examples.variograms --num-iter=500 --estimator=cressie --ons-scale=15 --verbose