fdfitter: Improve inversion performance (Friday afternoon) #558
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Model inversion can be pretty slow and generally, it is cumbersome to have to specify the inversion bounds.
This draft fixes both these issues by introducing a new inversion mode that does simple rootfinding to find the outer extents of the to-be-inverted range. As a fallback method, we fall back to the old method of using
least_squares
to invert the function.Differences with the old method:
Fitter
less general, it these bounds do hold for any sort of F, d model you'd throw at it (positive only).interpolate=True
.