5 releases

0.1.4 Nov 25, 2024
0.1.3 Nov 25, 2024
0.1.2 Nov 8, 2024
0.1.1 Nov 6, 2024
0.1.0 Nov 1, 2024

#5 in #uncertainty

Download history 101/week @ 2024-10-28 303/week @ 2024-11-04 44/week @ 2024-11-11 9/week @ 2024-11-18 304/week @ 2024-11-25 4/week @ 2024-12-02 144/week @ 2024-12-09

462 downloads per month
Used in 2 crates

MIT license

18KB
380 lines

Cert

Cert is a simple library for handling and manipulating numerical values associated with an uncertainty. It provides a common interface for handling types with relative and absolute uncertainty, and facilitates conversion between the two. It also implements common mathematical operations on types implementing the Uncertainty trait. This allows the values to be operated on, respecting propagation of error into the final result.

Cert exposes two concrete types, a AbsUncertainty and a RelUncertainty, both of which implement the Uncertainty interface. The AbsUncertainty type is used to represent absolute uncertainties, or values where the uncertainty is defined by the standard deviation of the value. The RelUncertainty type is used to represent relative uncertainties, where the uncertainty level is defined by the coefficient of variation, or the ratio of the standard deviation and the value.

Both types are generic over the inner float, which has to implement the Float and Zero traits from the num_traits crate.

Dependencies

~93–310KB