jgrusewski a0a2a1d736 docs: add hyperopt improvements design and implementation plan ...

A+B approach: smooth penalties + position limits + CUDA cleanup +
search space reduction (45D→25D) + TPE optimizer replacing PSO.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-03 18:17:23 +01:00

..

infra

feat(infra): GPU training launcher with local/cloud routing

2026-02-23 11:19:58 +01:00

plans

docs: add hyperopt improvements design and implementation plan

2026-03-03 18:17:23 +01:00