kigland 2208e255b8 feat: update optimization defaults for viable CMA-ES convergence
- generations: 20 -> 200 (need at least 200 for high-dim search)
- population_size: 8 -> 64 (was far below CMA-ES recommendation)
- Add pca_dimensions: 128 for dimensionality reduction
- Add use_diagonal_cma: true for sep-CMA-ES variant
- Add cache_fitness: true to avoid redundant API calls

Previous defaults (popsize=8, gen=20) gave only 160 evaluations,
which is ~1% of the theoretical minimum needed for convergence.
2026-03-08 15:06:46 +08:00
2026-03-08 01:43:09 +00:00
2026-03-08 01:57:41 +00:00
2026-03-08 09:03:03 +08:00
2026-03-08 01:53:05 +00:00
2026-03-08 09:03:03 +08:00
2026-03-08 01:45:36 +00:00
S
Description
No description provided
GPL-3.0
813 KiB
Languages
Python 100%