functional: add value_and_grad and auxiliary value/grad args#51
Merged
Conversation
Widen Functional.value/grad to accept *args/**kwargs (auxiliary parameters such as data, temperature, or a penalty weight) and add value_and_grad with a base default that delegates to value + grad, so subclasses can override with a single-pass evaluator. Extras are threaded through __call__ and the check-free cores; checked_method already forwards them and validates only x. Groundwork for the 0.4.2 optimizer-loop work (fused value_and_grad) and TraceFunctional. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Pavlo3P
added a commit
that referenced
this pull request
Jul 1, 2026
minimize_optax now runs the whole optimization inside
jax.jit(jax.lax.while_loop(...)): the fused F.value_and_grad is evaluated once
per iteration and cached, so the stopping test (grad_norm <= tol) and the
progress log reuse cached values with no recomputation and no per-iteration host
sync. It returns a rich OptaxResult and supports on-device convergence +
finiteness stop, four-column progress logging (iteration / value / delta F /
grad_norm) via jax.debug.print, a preallocated history buffer with post-hoc
progress_callback replay, and an optional project retraction hook.
BREAKING: the old fixed-`steps` eager loop is removed. New signature:
minimize_optax(F, x0, opt, *, max_iter=1000, tol=1e-6, project=None,
verbose=1, log_every=50, history_every=None,
record_history=True, progress_callback=None) -> OptaxResult
Call sites and tests migrated (steps -> max_iter; returns OptaxResult).
Builds on Functional.value_and_grad (#51).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This was referenced Jul 1, 2026
Merged
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
Widen
Functional.value/gradto accept*args, **kwargs(auxiliary parameters such as data, temperature, or a penalty weight) and add avalue_and_gradmethod.Changes
value(self, x, *args, **kwargs)/grad(self, x, *args, **kwargs); extras threaded through__call__,_value_core,_grad_core.value_and_grad(self, x, *args, **kwargs) -> (value, grad)— base default delegates tovalue+grad; subclasses may override with a single-pass AD.checked_methodunchanged — already forwards extras and validates onlyx.Testing
pytest tests/functional tests/optimize -q→ 375 passed.🤖 Generated with Claude Code