How to properly benchmark and test A-SABR? #23
Description
Problem
After making changes, I want to make sure they did not negatively impact the API or the performances.
Currently, as far as I understand, Criterion 0.3 is used to run benchmarks and the examples/ are used with cargo run to make sure the program behaves.
I did not find instructions or recommended parameters, or a way to automate these tests.
Proposed solution
We could devise a solution to run tests and benchmarks in an environment that's as reproducible and clean of noise as possible, and does not take too long.
To run it, we could then use GitHub Actions and/or Nix or other tools.
We could also introduce unit tests in the codebase?