194 lines
25 KiB
Plaintext
194 lines
25 KiB
Plaintext
[2026-04-13 13:36:24] ============================================================
|
|
[2026-04-13 13:36:24] [AutoResearch] Phase 1 — Real PPO Training + GP+UCB Optimization
|
|
[2026-04-13 13:36:24] [AutoResearch] Max trials: 50 | kappa: 2.0 | push every: 10
|
|
[2026-04-13 13:36:24] [AutoResearch] Results: /home/paulh/projects/donkeycar-rl-autoresearch/agent/outerloop-results/autoresearch_results_phase1.jsonl
|
|
[2026-04-13 13:36:24] [AutoResearch] Champion: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/champion
|
|
[2026-04-13 13:36:24] ============================================================
|
|
[2026-04-13 13:36:24] [AutoResearch] Loaded 0 existing Phase 1 results.
|
|
[2026-04-13 13:36:24] [AutoResearch] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:36:24]
|
|
[AutoResearch] ========== Trial 1/50 ==========
|
|
[2026-04-13 13:36:24] [AutoResearch] Only 0 results — using random proposal.
|
|
[2026-04-13 13:36:24] [AutoResearch] Proposed: {'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:36:26] [AutoResearch] Launching trial 1: {'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:41:13] [AutoResearch] Trial 1 finished in 286.6s, returncode=0
|
|
[2026-04-13 13:41:13] [AutoResearch] Trial 1: mean_reward=14.3331 std_reward=0.7924
|
|
[2026-04-13 13:41:13] [AutoResearch] === Trial 1 Summary ===
|
|
[2026-04-13 13:41:13] Total Phase 1 runs: 1
|
|
[2026-04-13 13:41:13] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:41:13] Top 5:
|
|
[2026-04-13 13:41:13] mean_reward=14.3331 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:41:15]
|
|
[AutoResearch] ========== Trial 2/50 ==========
|
|
[2026-04-13 13:41:15] [AutoResearch] Only 1 results — using random proposal.
|
|
[2026-04-13 13:41:15] [AutoResearch] Proposed: {'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:41:17] [AutoResearch] Launching trial 2: {'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:44:39] [AutoResearch] Trial 2 finished in 201.7s, returncode=0
|
|
[2026-04-13 13:44:39] [AutoResearch] Trial 2: mean_reward=14.6781 std_reward=0.0047
|
|
[2026-04-13 13:44:39] [AutoResearch] === Trial 2 Summary ===
|
|
[2026-04-13 13:44:39] Total Phase 1 runs: 2
|
|
[2026-04-13 13:44:39] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:44:39] Top 5:
|
|
[2026-04-13 13:44:39] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:44:39] mean_reward=14.3331 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:44:41]
|
|
[AutoResearch] ========== Trial 3/50 ==========
|
|
[2026-04-13 13:44:41] [AutoResearch] Only 2 results — using random proposal.
|
|
[2026-04-13 13:44:41] [AutoResearch] Proposed: {'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:44:43] [AutoResearch] Launching trial 3: {'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:22] [AutoResearch] Trial 3 finished in 99.2s, returncode=0
|
|
[2026-04-13 13:46:22] [AutoResearch] Trial 3: mean_reward=15.0946 std_reward=0.0381
|
|
[2026-04-13 13:46:22] [AutoResearch] === Trial 3 Summary ===
|
|
[2026-04-13 13:46:22] Total Phase 1 runs: 3
|
|
[2026-04-13 13:46:22] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:22] Top 5:
|
|
[2026-04-13 13:46:22] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:22] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:22] mean_reward=14.3331 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:24]
|
|
[AutoResearch] ========== Trial 4/50 ==========
|
|
[2026-04-13 13:46:24] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:46:24] UCB=2.3657 mu=0.6683 sigma=0.8487 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497}
|
|
[2026-04-13 13:46:24] UCB=2.3642 mu=0.6129 sigma=0.8757 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.003872092322417417, 'timesteps': 1454}
|
|
[2026-04-13 13:46:24] UCB=2.3627 mu=0.6363 sigma=0.8632 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.002284233345527573, 'timesteps': 2792}
|
|
[2026-04-13 13:46:24] UCB=2.3611 mu=0.6142 sigma=0.8735 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0024387325888159195, 'timesteps': 1898}
|
|
[2026-04-13 13:46:24] UCB=2.3610 mu=0.6522 sigma=0.8544 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0007105241846548975, 'timesteps': 1492}
|
|
[2026-04-13 13:46:24] [AutoResearch] Proposed: {'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:46:26] [AutoResearch] Launching trial 4: {'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:00] [AutoResearch] Trial 4 finished in 273.6s, returncode=0
|
|
[2026-04-13 13:51:00] [AutoResearch] Trial 4: mean_reward=14.6036 std_reward=0.0414
|
|
[2026-04-13 13:51:00] [AutoResearch] === Trial 4 Summary ===
|
|
[2026-04-13 13:51:00] Total Phase 1 runs: 4
|
|
[2026-04-13 13:51:00] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:00] Top 5:
|
|
[2026-04-13 13:51:00] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:00] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:00] mean_reward=14.6036 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:00] mean_reward=14.3331 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:02]
|
|
[AutoResearch] ========== Trial 5/50 ==========
|
|
[2026-04-13 13:51:02] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:51:02] UCB=2.6511 mu=1.1122 sigma=0.7695 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691}
|
|
[2026-04-13 13:51:02] UCB=2.6390 mu=1.1590 sigma=0.7400 params={'n_steer': 9, 'n_throttle': 4, 'learning_rate': 0.0023718639468651766, 'timesteps': 1039}
|
|
[2026-04-13 13:51:02] UCB=2.6371 mu=0.9686 sigma=0.8342 params={'n_steer': 9, 'n_throttle': 5, 'learning_rate': 0.0024714959214247904, 'timesteps': 1444}
|
|
[2026-04-13 13:51:02] UCB=2.6303 mu=1.0400 sigma=0.7951 params={'n_steer': 9, 'n_throttle': 5, 'learning_rate': 0.0028500094580389797, 'timesteps': 1932}
|
|
[2026-04-13 13:51:02] UCB=2.6281 mu=0.9823 sigma=0.8229 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.0018569421425612218, 'timesteps': 1300}
|
|
[2026-04-13 13:51:02] [AutoResearch] Proposed: {'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:51:04] [AutoResearch] Launching trial 5: {'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] [AutoResearch] Trial 5 finished in 114.3s, returncode=0
|
|
[2026-04-13 13:52:58] [AutoResearch] Trial 5: mean_reward=92.4248 std_reward=0.2184
|
|
[2026-04-13 13:52:58] [AutoResearch] === Trial 5 Summary ===
|
|
[2026-04-13 13:52:58] Total Phase 1 runs: 5
|
|
[2026-04-13 13:52:58] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] Top 5:
|
|
[2026-04-13 13:52:58] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] mean_reward=14.6036 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:52:58] mean_reward=14.3331 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0006023644308821473, 'timesteps': 4723, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:53:00]
|
|
[AutoResearch] ========== Trial 6/50 ==========
|
|
[2026-04-13 13:53:00] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:53:00] UCB=2.7680 mu=1.8581 sigma=0.4549 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150}
|
|
[2026-04-13 13:53:00] UCB=2.7571 mu=1.3859 sigma=0.6856 params={'n_steer': 7, 'n_throttle': 5, 'learning_rate': 0.0005903908533825176, 'timesteps': 2343}
|
|
[2026-04-13 13:53:00] UCB=2.6722 mu=1.7592 sigma=0.4565 params={'n_steer': 7, 'n_throttle': 5, 'learning_rate': 0.0007155529793779908, 'timesteps': 1801}
|
|
[2026-04-13 13:53:00] UCB=2.6514 mu=0.8841 sigma=0.8837 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.00022184611967850532, 'timesteps': 1388}
|
|
[2026-04-13 13:53:00] UCB=2.6250 mu=1.2493 sigma=0.6879 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.0007146109608951488, 'timesteps': 1192}
|
|
[2026-04-13 13:53:00] [AutoResearch] Proposed: {'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:53:02] [AutoResearch] Launching trial 6: {'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] [AutoResearch] Trial 6 finished in 109.6s, returncode=0
|
|
[2026-04-13 13:54:52] [AutoResearch] Trial 6: mean_reward=74.2498 std_reward=0.327
|
|
[2026-04-13 13:54:52] [AutoResearch] === Trial 6 Summary ===
|
|
[2026-04-13 13:54:52] Total Phase 1 runs: 6
|
|
[2026-04-13 13:54:52] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] Top 5:
|
|
[2026-04-13 13:54:52] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] mean_reward=74.2498 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:52] mean_reward=14.6036 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0030375027886947775, 'timesteps': 2497, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:54]
|
|
[AutoResearch] ========== Trial 7/50 ==========
|
|
[2026-04-13 13:54:54] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:54:54] UCB=2.5933 mu=0.8355 sigma=0.8789 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081}
|
|
[2026-04-13 13:54:54] UCB=2.5505 mu=0.7947 sigma=0.8779 params={'n_steer': 9, 'n_throttle': 4, 'learning_rate': 0.00014230944957386966, 'timesteps': 2582}
|
|
[2026-04-13 13:54:54] UCB=2.5503 mu=0.8231 sigma=0.8636 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.00023514026888676683, 'timesteps': 3062}
|
|
[2026-04-13 13:54:54] UCB=2.5217 mu=1.2654 sigma=0.6281 params={'n_steer': 7, 'n_throttle': 5, 'learning_rate': 0.0006389418712953596, 'timesteps': 1932}
|
|
[2026-04-13 13:54:54] UCB=2.3784 mu=0.6315 sigma=0.8734 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.00034403242394723434, 'timesteps': 2732}
|
|
[2026-04-13 13:54:54] [AutoResearch] Proposed: {'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:54:56] [AutoResearch] Launching trial 7: {'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] [AutoResearch] Trial 7 finished in 212.1s, returncode=0
|
|
[2026-04-13 13:58:28] [AutoResearch] Trial 7: mean_reward=326.6374 std_reward=2.3715
|
|
[2026-04-13 13:58:28] [AutoResearch] === Trial 7 Summary ===
|
|
[2026-04-13 13:58:28] Total Phase 1 runs: 7
|
|
[2026-04-13 13:58:28] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] Top 5:
|
|
[2026-04-13 13:58:28] mean_reward=326.6374 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] mean_reward=74.2498 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:28] mean_reward=14.6781 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0026097080330405096, 'timesteps': 3663, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:30]
|
|
[AutoResearch] ========== Trial 8/50 ==========
|
|
[2026-04-13 13:58:30] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:58:30] UCB=3.0530 mu=1.8512 sigma=0.6009 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293}
|
|
[2026-04-13 13:58:30] UCB=2.9620 mu=1.3333 sigma=0.8144 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0002690144955499583, 'timesteps': 2184}
|
|
[2026-04-13 13:58:30] UCB=2.9307 mu=1.5416 sigma=0.6945 params={'n_steer': 5, 'n_throttle': 5, 'learning_rate': 0.0007634930463964747, 'timesteps': 2464}
|
|
[2026-04-13 13:58:30] UCB=2.8609 mu=1.2576 sigma=0.8017 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.00030073926504989247, 'timesteps': 2405}
|
|
[2026-04-13 13:58:30] UCB=2.8085 mu=1.2678 sigma=0.7704 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0002460273175097693, 'timesteps': 2273}
|
|
[2026-04-13 13:58:30] [AutoResearch] Proposed: {'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:58:32] [AutoResearch] Launching trial 8: {'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] [AutoResearch] Trial 8 finished in 221.7s, returncode=0
|
|
[2026-04-13 14:02:13] [AutoResearch] Trial 8: mean_reward=492.1545 std_reward=20.4057
|
|
[2026-04-13 14:02:13] [AutoResearch] === Trial 8 Summary ===
|
|
[2026-04-13 14:02:13] Total Phase 1 runs: 8
|
|
[2026-04-13 14:02:13] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] Top 5:
|
|
[2026-04-13 14:02:13] mean_reward=492.1545 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] mean_reward=326.6374 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] mean_reward=74.2498 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:13] mean_reward=15.0946 params={'n_steer': 8, 'n_throttle': 4, 'learning_rate': 0.0022425720960039287, 'timesteps': 1878, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:15]
|
|
[AutoResearch] ========== Trial 9/50 ==========
|
|
[2026-04-13 14:02:15] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 14:02:15] UCB=2.8782 mu=1.2806 sigma=0.7988 params={'n_steer': 4, 'n_throttle': 4, 'learning_rate': 0.0010269810535699494, 'timesteps': 1405}
|
|
[2026-04-13 14:02:15] UCB=2.8699 mu=1.2136 sigma=0.8282 params={'n_steer': 3, 'n_throttle': 4, 'learning_rate': 0.0008951125166902782, 'timesteps': 1867}
|
|
[2026-04-13 14:02:15] UCB=2.7808 mu=1.0494 sigma=0.8657 params={'n_steer': 3, 'n_throttle': 4, 'learning_rate': 0.0007281373711988931, 'timesteps': 1628}
|
|
[2026-04-13 14:02:15] UCB=2.7699 mu=1.4417 sigma=0.6641 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0012071606973414922, 'timesteps': 1881}
|
|
[2026-04-13 14:02:15] UCB=2.7343 mu=1.6068 sigma=0.5638 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0008505460420664955, 'timesteps': 2487}
|
|
[2026-04-13 14:02:15] [AutoResearch] Proposed: {'n_steer': 4, 'n_throttle': 4, 'learning_rate': 0.0010269810535699494, 'timesteps': 1405, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:02:17] [AutoResearch] Launching trial 9: {'n_steer': 4, 'n_throttle': 4, 'learning_rate': 0.0010269810535699494, 'timesteps': 1405, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] [AutoResearch] Trial 9 finished in 109.5s, returncode=0
|
|
[2026-04-13 14:04:07] [AutoResearch] Trial 9: mean_reward=47.3482 std_reward=0.07
|
|
[2026-04-13 14:04:07] [AutoResearch] === Trial 9 Summary ===
|
|
[2026-04-13 14:04:07] Total Phase 1 runs: 9
|
|
[2026-04-13 14:04:07] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] Top 5:
|
|
[2026-04-13 14:04:07] mean_reward=492.1545 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] mean_reward=326.6374 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] mean_reward=74.2498 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:07] mean_reward=47.3482 params={'n_steer': 4, 'n_throttle': 4, 'learning_rate': 0.0010269810535699494, 'timesteps': 1405, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:09]
|
|
[AutoResearch] ========== Trial 10/50 ==========
|
|
[2026-04-13 14:04:09] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 14:04:09] UCB=3.0861 mu=1.7282 sigma=0.6790 params={'n_steer': 5, 'n_throttle': 3, 'learning_rate': 0.00037737321665256695, 'timesteps': 2717}
|
|
[2026-04-13 14:04:09] UCB=3.0467 mu=2.0076 sigma=0.5196 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.00021279372557134375, 'timesteps': 2438}
|
|
[2026-04-13 14:04:09] UCB=3.0434 mu=2.2259 sigma=0.4088 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.00041913098323291125, 'timesteps': 2524}
|
|
[2026-04-13 14:04:09] UCB=2.9788 mu=1.7738 sigma=0.6025 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.00031041165555133226, 'timesteps': 3122}
|
|
[2026-04-13 14:04:09] UCB=2.9735 mu=1.5955 sigma=0.6890 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0005830228901312196, 'timesteps': 2255}
|
|
[2026-04-13 14:04:09] [AutoResearch] Proposed: {'n_steer': 5, 'n_throttle': 3, 'learning_rate': 0.00037737321665256695, 'timesteps': 2717, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:04:11] [AutoResearch] Launching trial 10: {'n_steer': 5, 'n_throttle': 3, 'learning_rate': 0.00037737321665256695, 'timesteps': 2717, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] [AutoResearch] Trial 10 finished in 226.7s, returncode=0
|
|
[2026-04-13 14:07:58] [AutoResearch] Trial 10: mean_reward=1157.047 std_reward=0.7533
|
|
[2026-04-13 14:07:58] [AutoResearch] === Trial 10 Summary ===
|
|
[2026-04-13 14:07:58] Total Phase 1 runs: 10
|
|
[2026-04-13 14:07:58] Champion: trial=5 mean_reward=4582.7984 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] Top 5:
|
|
[2026-04-13 14:07:58] mean_reward=1157.0470 params={'n_steer': 5, 'n_throttle': 3, 'learning_rate': 0.00037737321665256695, 'timesteps': 2717, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] mean_reward=492.1545 params={'n_steer': 5, 'n_throttle': 4, 'learning_rate': 0.0003046182905194233, 'timesteps': 2293, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] mean_reward=326.6374 params={'n_steer': 6, 'n_throttle': 5, 'learning_rate': 0.0001799978550884136, 'timesteps': 2081, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] mean_reward=92.4248 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0011131823295383878, 'timesteps': 1691, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 14:07:58] mean_reward=74.2498 params={'n_steer': 8, 'n_throttle': 5, 'learning_rate': 0.0010109905842864714, 'timesteps': 1150, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|