Paul Huliganga 4f77b8a468 fix: always save and return the BEST model, not the last one ... This was the root cause of losing good models during training. The model could learn to lap at step 30k then drift to a worse policy by step 90k, and we only ever saved the final weights. Changes to train_multitrack(): - Tracks best_segment_reward across all segments - Saves best_model.zip whenever a new high score is achieved - At end of training, RELOADS best_model.zip before returning so the caller always gets the best policy found, not the drift Both files saved per trial: model.zip <- latest checkpoint (crash recovery) best_model.zip <- best policy seen during training (used for eval) Agent: pi Tests: 102 passed Tests-Added: 0 TypeScript: N/A		2026-04-17 14:45:37 -04:00
..
model-000	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
model-001	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
model-002	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
model-003	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
autoresearch_log.txt	AUTORESEARCH: 300 total trials complete - best mean_reward=141.85 at n_steer=8, n_throttle=5, lr=0.00202	2026-04-13 01:56:06 -04:00
autoresearch_phase1_log.txt	milestone: Phase 1 complete — genuine driving confirmed; launch Phase 2 corner learning	2026-04-13 19:33:06 -04:00
autoresearch_phase1_log_CORRUPTED_circular_driving.txt	fix: path-efficiency reward (v3) defeats circular driving exploit	2026-04-13 13:36:17 -04:00
autoresearch_phase1_log_CORRUPTED_reward_hacking.txt	fix: hack-proof reward shaping + reward hacking detection + research log	2026-04-13 12:27:48 -04:00
autoresearch_phase2_log.txt	fix: always save and return the BEST model, not the last one	2026-04-17 14:45:37 -04:00
autoresearch_phase3_log.txt	fix: always save and return the BEST model, not the last one	2026-04-17 14:45:37 -04:00
autoresearch_phase4_log.txt	feat: v5 reward — speed × CTE-quality, drop efficiency term	2026-04-17 13:25:38 -04:00
autoresearch_results.jsonl	AUTORESEARCH: 300 total trials complete - best mean_reward=141.85 at n_steer=8, n_throttle=5, lr=0.00202	2026-04-13 01:56:06 -04:00
autoresearch_results_phase1.jsonl	autoresearch: phase1 trial 50 results	2026-04-13 19:17:56 -04:00
autoresearch_results_phase1_CORRUPTED_circular_driving.jsonl	fix: path-efficiency reward (v3) defeats circular driving exploit	2026-04-13 13:36:17 -04:00
autoresearch_results_phase1_CORRUPTED_reward_hacking.jsonl	fix: hack-proof reward shaping + reward hacking detection + research log	2026-04-13 12:27:48 -04:00
autoresearch_results_phase2.jsonl	autoresearch: phase1 trial 20 results	2026-04-14 04:35:45 -04:00
autoresearch_results_phase3.jsonl	Wave 4: scratch training on generated_track + mountain_track, zero-shot mini_monaco	2026-04-14 22:40:38 -04:00
autoresearch_results_phase3_CONTAMINATED_v2.jsonl	fix: complete LR override — must patch lr_schedule, not just param_groups	2026-04-14 21:27:43 -04:00
autoresearch_results_phase3_CONTAMINATED_wrong_lr.jsonl	fix: LR override was not reaching the optimizer — all trials ran at 0.000225	2026-04-14 20:37:48 -04:00
autoresearch_results_phase4.jsonl	wave3: autoresearch trial 25 results	2026-04-16 20:01:51 -04:00
clean_sweep_results.jsonl	AUTORESEARCH: Full Karpathy-style GP+UCB meta-controller, clean base data, fixed all paths, ready to run	2026-04-13 00:52:00 -04:00
eval_summary.jsonl	fix: track switching via unwrapped viewer.exit_scene() — automatic scene changes work	2026-04-14 10:04:15 -04:00
multitrack_results.jsonl	results: complete multi-track generalization baseline — 1/10 tracks drivable pre-Wave3	2026-04-14 11:31:08 -04:00
nohup_outerloop.log	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
outer_monitor.log	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00
sweep_results.jsonl	Initial commit: stable RL sweep runner, legacy and new scripts, full docs included	2026-04-12 22:57:50 -04:00