docs(searches): add multi-start gradient searches to the modeling searches guide#278
Merged
Merged
Conversation
…rches guide Add a MultiStartAdam section to scripts/guides/modeling/searches.py documenting af.MultiStartAdam (with MultiStartADABelief / MultiStartLion as alternatives) — the JAX multi-start gradient MAP optimizer that works on the complex lens parameter spaces where single-start LBFGS gets stuck. Config-only, mirroring the LBFGS section; regenerate the paired notebook. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Phase 3 of the multi-start gradient MAP search promotion (PyAutoFit#1369). Adds a MultiStartAdam section to
scripts/guides/modeling/searches.pydocumentingaf.MultiStartAdam(withMultiStartADABelief/MultiStartLionnoted as drop-in optax-rule alternatives), positioned after LBFGS. Framed against the caveat the guide already states — that optimizers struggle on complex lens parameter spaces — since multi-start Adam is the optimizer that works there (its wide population of parallel starts escapes the local maxima that trap single-start LBFGS; benchmark-proven on the HST MGE lens likelihood).This guide is a configuration reference (it constructs search objects to document the API/settings; it does not run fits), so the addition mirrors how LBFGS is shown — a config block + prose. Contents updated; paired notebook regenerated.
Test Plan
searches.pyparses;af.MultiStartAdam(path_prefix=..., name="MultiStartAdam", n_starts=50, n_steps=500, learning_rate=0.01)constructspy_to_notebook+ colab-injection transform; diff scoped tosearches.py+searches.ipynbResolves #277.
Generated by the PyAutoLabs agent workflow.