Conversation
…ateless factory, plain-SGD arm deleted (#308) - sgdStepM: momentumFactor==0 runs one stateless update op per parameter and never reads optim->states (may be NULL) - sgdMCreateOptim: allocates no momentum-state buffers at momentumFactor==0 (optim->states == NULL); freeOptimSgdM handles the no-states case; the ten duplicated per-layer state-building blocks DRYed into one uniform post-loop - delete unreachable sgdStep + SGD enum value (verified unserialized); its three direct test callers (#279 dead-zone fixtures) ported byte-identically Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…GD update ops, FLOAT32 default, fail-fast otherwise (#310) - sgdMCreateOptim gains a by-value arithmetic_t updateMath (7th param, mirrors the layer-side per-op knobs), stored on sgd_t and fed into all three update opSpecs (previously hardcoded {ARITH_FLOAT32, HALF_AWAY}) - anything but ARITH_FLOAT32 fails fast at creation (sgdInit) AND step time (sgdStepM entry; the update kernels raw-cast float*) — both death-tested - mechanical sweep: 48 test + 9 example call sites, zero numeric movement; 7-arg _Static_assert signature contract updated - rounding ownership documented at the #282 seam (prologue = knob, inert for FLOAT32; OUT_WRITE epilogue = target qConfig, the #279 SR escape) - FEATURES.md optimizer section corrected; stale pre-#284 comments swept; mem_instrument guards states==NULL; momentumFactor creation-fixed doc note Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…le-epoch probes first/last train loss are per-EPOCH means; with EPOCHS=1 they are the same number by construction, so the check always warned — bit the CI c-stack-watermark probe, which runs EPOCHS=1 and printed a misleading 'SYM@8 too coarse to descend' on every run (the 50-epoch sweep shows sym8 converging indistinguishably from float). Now prints CONVERGENCE SKIP for single-epoch runs; smoke-verified (EPOCHS=1 -> SKIP line).
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.
No description provided.