Problem: The current implementation of the cholesky decomposition is far to slow as
We use a self-implementation
We are working with [Q]s which have a high precision at the cost of runtime
Therefore I looked at ways to overcome these issues
Using FLINTs function -> sadly not supported in FFI
Use simplify to get to a certain precision -> too slow...
[x] Use [f64] (double-precision) between conversions
The latest of these options proved to be the best. As we can not access FLINTs function, I basically reformulated their implementation into Rust and used the same format.
for/ of Component.
Testing
[x] I added basic working examples (possibly in doc-comment)
Due to the precision - it does not work for large integers
[x] I triggered all possible errors in my test in every possible way
[x] I included tests for all reasonable edge cases
Checklist:
[x] I have performed a self-review of my own code
[x] The code provides good readability and maintainability s.t. it fulfills best practices like talking code, modularity, ...
[x] The chosen implementation is not more complex than it has to be
[x] My code should work as intended and no side effects occur (e.g. memory leaks)
Description
Problem: The current implementation of the cholesky decomposition is far to slow as
Q]s which have a high precision at the cost of runtimeTherefore I looked at ways to overcome these issues
f64] (double-precision) between conversions The latest of these options proved to be the best. As we can not access FLINTs function, I basically reformulated their implementation into Rust and used the same format.for/ of
Component.Testing
Checklist: