Lightweight automatic differentiation and error propagation library
-
Updated
May 12, 2024 - C++
Error (or uncertainty) propagation is the practice of analyzing and accounting for the effect of numeric quantities' uncertainties on the results of functions that involve them.
When variables used in a function or mathematical operation have errors (due to measurement uncertainties, random fluctuations, sample variance, etc.), error propagation can be used to determine the resulting error of the function's output.
Lightweight automatic differentiation and error propagation library
Set of R functions designed to perform and report uncertainty propagation
This code performs generalized Brownian dynamics (GBD) simulations of a microparticle embedded in a viscoelastic fluid and calculates and propagates statistical and other sources of error in passive microrheology
Uncertainty propagation in C++
Simulation Uncertainty Quantification Querying
A lightweight Python utility providing values with associated uncertainty
Linear uncertainty propagation using JAX transformations
Rigorous moment propagation with partial information about moments and dependencies in Julia
Interface between Turing.jl and MonteCarloMeasurements.jl
Uncertainty-Aware Multidimensional Scaling (Reference Implementation)
Simple dead reckoning example in one dimension
Arbitrary Polynomial Chaos Toolkit
This code base is intended to serve as a starting point for interested researchers or practitioners to extend or apply the uncertainty propagation portion of the author's Master's thesis " GUM-compliant neural-network robustness verification".
Contains a method in Python to compute the uncertainty of an inventory (such as a pollutant inventory or a greenhouse gas inventory), based on input uncertainty for specific sub-categories of this inventory.
Example application to estimate wave energy spectrum from accelerometer measurements taken on board a vessel.
Allows to deal with power series which coefficients contain uncertainties
Data for the fourth Makridakis Competition
Experiments with incorporating uncertainty information in back prop. algorithm