support complex step derivative approximation
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 839
- Forks
- 220
- Avg merge
- 2d 4h
- Merged PRs (30d)
- 14
Description
Summary:
This is different from #789, as there is no complex version of var involved.
For a given function f that supports complex number evaluation, calculate its derivative using complex step derivative approximation and generate var using the result.
Description:
Given
std::complex<double> f(const std::complex<double> x) {
/* ... */
}
one can calculate df/dx using complex step derivative approximation, avoiding loss of significance due to cancellation in real-number finite difference. This gives Math support in domains where complex number calculation is common (signal analysis, elector-magnetic solver, etc) and solvers are implemented with complex-number capability.
The connection of CSDA to forward-mode of AD is pointed out in Chap. 10 of GRIEWANK, A. 2000.
Current Version:
v2.18.0
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No source files, tests, or entry points are named. Start by reading issue #789 and the linked complex-step differentiation reference, then trace how the library generates var from derivative information. Done means supporting the described complex-valued function case and generating var from the approximation, with coverage for the new behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- tooling
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100