223 lines
28 KiB
Plaintext
223 lines
28 KiB
Plaintext
[2026-04-13 12:26:21] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:26:21] UCB=2.7567 mu=1.2278 sigma=0.7644 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.002270622623224986, 'timesteps': 3888}
|
|
[2026-04-13 12:26:21] UCB=2.7300 mu=1.1710 sigma=0.7795 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.002011397993568161, 'timesteps': 4033}
|
|
[2026-04-13 12:26:21] UCB=2.6457 mu=1.4878 sigma=0.5790 params={'n_steer': 9, 'n_throttle': 2, 'learning_rate': 0.00219005726516088, 'timesteps': 4774}
|
|
[2026-04-13 12:26:21] UCB=2.6320 mu=1.1819 sigma=0.7250 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0020813954690263674, 'timesteps': 4022}
|
|
[2026-04-13 12:26:21] UCB=2.5412 mu=1.2499 sigma=0.6457 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0025942479713410636, 'timesteps': 4135}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=50.0000 params={'n_steer': 5}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=80.0000 params={'n_steer': 7}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 0: mean_reward=50.0000 params={'r': 50}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=80.0000 params={'r': 80}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 3: mean_reward=90.0000 params={'r': 90}
|
|
[2026-04-13 12:26:21] [Champion] 🏆 NEW BEST! Trial 5: mean_reward=75.0000 params={'n_steer': 8}
|
|
[2026-04-13 12:26:21] [AutoResearch] Only 1 results — using random proposal.
|
|
[2026-04-13 12:27:28] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:27:28] UCB=2.7567 mu=1.2278 sigma=0.7644 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.002270622623224986, 'timesteps': 3888}
|
|
[2026-04-13 12:27:28] UCB=2.7300 mu=1.1710 sigma=0.7795 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.002011397993568161, 'timesteps': 4033}
|
|
[2026-04-13 12:27:28] UCB=2.6457 mu=1.4878 sigma=0.5790 params={'n_steer': 9, 'n_throttle': 2, 'learning_rate': 0.00219005726516088, 'timesteps': 4774}
|
|
[2026-04-13 12:27:28] UCB=2.6320 mu=1.1819 sigma=0.7250 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0020813954690263674, 'timesteps': 4022}
|
|
[2026-04-13 12:27:28] UCB=2.5412 mu=1.2499 sigma=0.6457 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0025942479713410636, 'timesteps': 4135}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=50.0000 params={'n_steer': 5}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=80.0000 params={'n_steer': 7}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 0: mean_reward=50.0000 params={'r': 50}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=80.0000 params={'r': 80}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 3: mean_reward=90.0000 params={'r': 90}
|
|
[2026-04-13 12:27:28] [Champion] 🏆 NEW BEST! Trial 5: mean_reward=75.0000 params={'n_steer': 8}
|
|
[2026-04-13 12:27:28] [AutoResearch] Only 1 results — using random proposal.
|
|
[2026-04-13 12:27:56] ============================================================
|
|
[2026-04-13 12:27:56] [AutoResearch] Phase 1 — Real PPO Training + GP+UCB Optimization
|
|
[2026-04-13 12:27:56] [AutoResearch] Max trials: 50 | kappa: 2.0 | push every: 10
|
|
[2026-04-13 12:27:56] [AutoResearch] Results: /home/paulh/projects/donkeycar-rl-autoresearch/agent/outerloop-results/autoresearch_results_phase1.jsonl
|
|
[2026-04-13 12:27:56] [AutoResearch] Champion: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/champion
|
|
[2026-04-13 12:27:56] ============================================================
|
|
[2026-04-13 12:27:56] [AutoResearch] Loaded 0 existing Phase 1 results.
|
|
[2026-04-13 12:27:56] [AutoResearch] No champion yet.
|
|
[2026-04-13 12:27:56]
|
|
[AutoResearch] ========== Trial 1/50 ==========
|
|
[2026-04-13 12:27:56] [AutoResearch] Only 0 results — using random proposal.
|
|
[2026-04-13 12:27:56] [AutoResearch] Proposed: {'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:27:58] [AutoResearch] Launching trial 1: {'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:31:19] [AutoResearch] Trial 1 finished in 200.9s, returncode=0
|
|
[2026-04-13 12:31:19] [AutoResearch] Trial 1: mean_reward=270.5642 std_reward=0.1427
|
|
[2026-04-13 12:31:19] [Champion] 🏆 NEW BEST! Trial 1: mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:31:19] [AutoResearch] === Trial 1 Summary ===
|
|
[2026-04-13 12:31:19] Total Phase 1 runs: 1
|
|
[2026-04-13 12:31:19] Champion: trial=1 mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:31:19] Top 5:
|
|
[2026-04-13 12:31:19] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:31:21]
|
|
[AutoResearch] ========== Trial 2/50 ==========
|
|
[2026-04-13 12:31:21] [AutoResearch] Only 1 results — using random proposal.
|
|
[2026-04-13 12:31:21] [AutoResearch] Proposed: {'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:31:23] [AutoResearch] Launching trial 2: {'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:33:09] [AutoResearch] Trial 2 finished in 106.4s, returncode=0
|
|
[2026-04-13 12:33:09] [AutoResearch] Trial 2: mean_reward=15.537 std_reward=0.0257
|
|
[2026-04-13 12:33:09] [AutoResearch] === Trial 2 Summary ===
|
|
[2026-04-13 12:33:09] Total Phase 1 runs: 2
|
|
[2026-04-13 12:33:09] Champion: trial=1 mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:33:09] Top 5:
|
|
[2026-04-13 12:33:09] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:33:09] mean_reward=15.5370 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:33:11]
|
|
[AutoResearch] ========== Trial 3/50 ==========
|
|
[2026-04-13 12:33:11] [AutoResearch] Only 2 results — using random proposal.
|
|
[2026-04-13 12:33:11] [AutoResearch] Proposed: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.002684399373384131, 'timesteps': 1755, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:33:13] [AutoResearch] Launching trial 3: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.002684399373384131, 'timesteps': 1755, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:55] [AutoResearch] Trial 3 finished in 101.3s, returncode=0
|
|
[2026-04-13 12:34:55] [AutoResearch] Trial 3: mean_reward=15.0687 std_reward=0.0219
|
|
[2026-04-13 12:34:55] [AutoResearch] === Trial 3 Summary ===
|
|
[2026-04-13 12:34:55] Total Phase 1 runs: 3
|
|
[2026-04-13 12:34:55] Champion: trial=1 mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:55] Top 5:
|
|
[2026-04-13 12:34:55] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:55] mean_reward=15.5370 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:55] mean_reward=15.0687 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.002684399373384131, 'timesteps': 1755, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:57]
|
|
[AutoResearch] ========== Trial 4/50 ==========
|
|
[2026-04-13 12:34:57] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:34:57] UCB=2.4445 mu=0.7362 sigma=0.8541 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279}
|
|
[2026-04-13 12:34:57] UCB=2.4375 mu=0.9279 sigma=0.7548 params={'n_steer': 9, 'n_throttle': 4, 'learning_rate': 0.0005153901257954376, 'timesteps': 3901}
|
|
[2026-04-13 12:34:57] UCB=2.4265 mu=0.9512 sigma=0.7376 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.00041200493694593554, 'timesteps': 3979}
|
|
[2026-04-13 12:34:57] UCB=2.3771 mu=0.5079 sigma=0.9346 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.000984749088576101, 'timesteps': 4646}
|
|
[2026-04-13 12:34:57] UCB=2.3748 mu=0.6941 sigma=0.8403 params={'n_steer': 8, 'n_throttle': 2, 'learning_rate': 0.0010994124879848107, 'timesteps': 3297}
|
|
[2026-04-13 12:34:57] [AutoResearch] Proposed: {'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:34:59] [AutoResearch] Launching trial 4: {'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] [AutoResearch] Trial 4 finished in 297.3s, returncode=0
|
|
[2026-04-13 12:39:56] [AutoResearch] Trial 4: mean_reward=627.6915 std_reward=2.3536
|
|
[2026-04-13 12:39:56] [Champion] 🏆 NEW BEST! Trial 4: mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] [AutoResearch] === Trial 4 Summary ===
|
|
[2026-04-13 12:39:56] Total Phase 1 runs: 4
|
|
[2026-04-13 12:39:56] Champion: trial=4 mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] Top 5:
|
|
[2026-04-13 12:39:56] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] mean_reward=15.5370 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:56] mean_reward=15.0687 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.002684399373384131, 'timesteps': 1755, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:39:58]
|
|
[AutoResearch] ========== Trial 5/50 ==========
|
|
[2026-04-13 12:39:58] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:39:58] UCB=2.6036 mu=0.9596 sigma=0.8220 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787}
|
|
[2026-04-13 12:39:58] UCB=2.6006 mu=1.3117 sigma=0.6445 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0007973784600824807, 'timesteps': 4846}
|
|
[2026-04-13 12:39:58] UCB=2.5701 mu=1.4143 sigma=0.5779 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0014200634295944165, 'timesteps': 4764}
|
|
[2026-04-13 12:39:58] UCB=2.5683 mu=1.3628 sigma=0.6028 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0016809469142980496, 'timesteps': 4540}
|
|
[2026-04-13 12:39:58] UCB=2.5337 mu=1.3776 sigma=0.5780 params={'n_steer': 9, 'n_throttle': 2, 'learning_rate': 0.0008758657417255942, 'timesteps': 4642}
|
|
[2026-04-13 12:39:58] [AutoResearch] Proposed: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:40:00] [AutoResearch] Launching trial 5: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0006801262090358742, 'timesteps': 4787, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:42] [AutoResearch] Trial 5 finished in 342.4s, returncode=0
|
|
[2026-04-13 12:45:42] [AutoResearch] Trial 5: mean_reward=4582.7984 std_reward=0.4849
|
|
[2026-04-13 12:45:43] [Champion] 🏆 NEW BEST! 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 12:45:43] [AutoResearch] === Trial 5 Summary ===
|
|
[2026-04-13 12:45:43] Total Phase 1 runs: 5
|
|
[2026-04-13 12:45:43] 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 12:45:43] Top 5:
|
|
[2026-04-13 12:45:43] 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 12:45:43] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:43] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:43] mean_reward=15.5370 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:43] mean_reward=15.0687 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.002684399373384131, 'timesteps': 1755, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:45]
|
|
[AutoResearch] ========== Trial 6/50 ==========
|
|
[2026-04-13 12:45:45] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:45:45] UCB=3.0560 mu=1.6412 sigma=0.7074 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929}
|
|
[2026-04-13 12:45:45] UCB=2.9399 mu=2.0094 sigma=0.4652 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0006569923016159191, 'timesteps': 4926}
|
|
[2026-04-13 12:45:45] UCB=2.9261 mu=1.2265 sigma=0.8498 params={'n_steer': 5, 'n_throttle': 3, 'learning_rate': 0.0011640583789804349, 'timesteps': 4808}
|
|
[2026-04-13 12:45:45] UCB=2.9161 mu=1.3094 sigma=0.8034 params={'n_steer': 6, 'n_throttle': 2, 'learning_rate': 0.000817395398052526, 'timesteps': 4272}
|
|
[2026-04-13 12:45:45] UCB=2.9056 mu=1.2637 sigma=0.8210 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0018216283484187063, 'timesteps': 4962}
|
|
[2026-04-13 12:45:45] [AutoResearch] Proposed: {'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:45:47] [AutoResearch] Launching trial 6: {'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:14] [AutoResearch] Trial 6 finished in 326.9s, returncode=0
|
|
[2026-04-13 12:51:14] [AutoResearch] Trial 6: mean_reward=454.064 std_reward=2.7272
|
|
[2026-04-13 12:51:14] [AutoResearch] === Trial 6 Summary ===
|
|
[2026-04-13 12:51:14] Total Phase 1 runs: 6
|
|
[2026-04-13 12:51:14] 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 12:51:14] Top 5:
|
|
[2026-04-13 12:51:14] 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 12:51:14] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:14] mean_reward=454.0640 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:14] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:14] mean_reward=15.5370 params={'n_steer': 6, 'n_throttle': 4, 'learning_rate': 0.0021131130619403186, 'timesteps': 1622, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:16]
|
|
[AutoResearch] ========== Trial 7/50 ==========
|
|
[2026-04-13 12:51:16] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:51:16] UCB=3.0459 mu=1.3501 sigma=0.8479 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.002103379184926248, 'timesteps': 4607}
|
|
[2026-04-13 12:51:16] UCB=3.0426 mu=1.8623 sigma=0.5901 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0014223856509359136, 'timesteps': 4157}
|
|
[2026-04-13 12:51:16] UCB=2.9241 mu=1.2240 sigma=0.8500 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0023860629604045636, 'timesteps': 4875}
|
|
[2026-04-13 12:51:16] UCB=2.8724 mu=1.3748 sigma=0.7488 params={'n_steer': 8, 'n_throttle': 2, 'learning_rate': 0.0003305272678070145, 'timesteps': 4528}
|
|
[2026-04-13 12:51:16] UCB=2.8000 mu=1.7430 sigma=0.5285 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0008378766646858452, 'timesteps': 4278}
|
|
[2026-04-13 12:51:16] [AutoResearch] Proposed: {'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.002103379184926248, 'timesteps': 4607, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:51:18] [AutoResearch] Launching trial 7: {'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.002103379184926248, 'timesteps': 4607, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:20] [AutoResearch] Trial 7 finished in 302.9s, returncode=0
|
|
[2026-04-13 12:56:20] [AutoResearch] Trial 7: mean_reward=199.5367 std_reward=2.4906
|
|
[2026-04-13 12:56:20] [AutoResearch] === Trial 7 Summary ===
|
|
[2026-04-13 12:56:20] Total Phase 1 runs: 7
|
|
[2026-04-13 12:56:20] 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 12:56:20] Top 5:
|
|
[2026-04-13 12:56:20] 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 12:56:20] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:20] mean_reward=454.0640 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:20] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:20] mean_reward=199.5367 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.002103379184926248, 'timesteps': 4607, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:22]
|
|
[AutoResearch] ========== Trial 8/50 ==========
|
|
[2026-04-13 12:56:22] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 12:56:22] UCB=3.7155 mu=2.8123 sigma=0.4516 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938}
|
|
[2026-04-13 12:56:22] UCB=3.6165 mu=2.1871 sigma=0.7147 params={'n_steer': 8, 'n_throttle': 2, 'learning_rate': 0.00027774487279327734, 'timesteps': 4741}
|
|
[2026-04-13 12:56:22] UCB=3.5817 mu=2.3074 sigma=0.6372 params={'n_steer': 8, 'n_throttle': 2, 'learning_rate': 0.00042186231182417467, 'timesteps': 4737}
|
|
[2026-04-13 12:56:22] UCB=3.1830 mu=1.7630 sigma=0.7100 params={'n_steer': 8, 'n_throttle': 2, 'learning_rate': 0.0010392862598823948, 'timesteps': 4340}
|
|
[2026-04-13 12:56:22] UCB=2.9801 mu=2.1794 sigma=0.4004 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.001278006004760151, 'timesteps': 4950}
|
|
[2026-04-13 12:56:22] [AutoResearch] Proposed: {'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 12:56:24] [AutoResearch] Launching trial 8: {'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:28] [AutoResearch] Trial 8 finished in 303.7s, returncode=0
|
|
[2026-04-13 13:01:28] [AutoResearch] Trial 8: mean_reward=306.1739 std_reward=13.6044
|
|
[2026-04-13 13:01:28] [AutoResearch] === Trial 8 Summary ===
|
|
[2026-04-13 13:01:28] Total Phase 1 runs: 8
|
|
[2026-04-13 13:01:28] 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 13:01:28] Top 5:
|
|
[2026-04-13 13:01:28] 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 13:01:28] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:28] mean_reward=454.0640 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:28] mean_reward=306.1739 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:28] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:30]
|
|
[AutoResearch] ========== Trial 9/50 ==========
|
|
[2026-04-13 13:01:30] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:01:30] UCB=5.5166 mu=4.1098 sigma=0.7034 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0014813539623020004, 'timesteps': 4054}
|
|
[2026-04-13 13:01:30] UCB=5.3730 mu=4.0927 sigma=0.6401 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.000972771458635757, 'timesteps': 3832}
|
|
[2026-04-13 13:01:30] UCB=5.0616 mu=3.3272 sigma=0.8672 params={'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0021026667911721994, 'timesteps': 4268}
|
|
[2026-04-13 13:01:30] UCB=4.9195 mu=3.4509 sigma=0.7343 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0019399712695168441, 'timesteps': 4823}
|
|
[2026-04-13 13:01:30] UCB=4.7893 mu=3.5895 sigma=0.5999 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0013622938331829832, 'timesteps': 4756}
|
|
[2026-04-13 13:01:30] [AutoResearch] Proposed: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0014813539623020004, 'timesteps': 4054, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:01:32] [AutoResearch] Launching trial 9: {'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0014813539623020004, 'timesteps': 4054, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:16] [AutoResearch] Trial 9 finished in 223.5s, returncode=0
|
|
[2026-04-13 13:05:16] [AutoResearch] Trial 9: mean_reward=15.5625 std_reward=0.0011
|
|
[2026-04-13 13:05:16] [AutoResearch] === Trial 9 Summary ===
|
|
[2026-04-13 13:05:16] Total Phase 1 runs: 9
|
|
[2026-04-13 13:05:16] 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 13:05:16] Top 5:
|
|
[2026-04-13 13:05:16] 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 13:05:16] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:16] mean_reward=454.0640 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:16] mean_reward=306.1739 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:16] mean_reward=270.5642 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.00014623033103219074, 'timesteps': 3139, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:18]
|
|
[AutoResearch] ========== Trial 10/50 ==========
|
|
[2026-04-13 13:05:18] [AutoResearch] GP UCB top-5 candidates:
|
|
[2026-04-13 13:05:18] UCB=3.2755 mu=2.1392 sigma=0.5681 params={'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0010464507674264373, 'timesteps': 4450}
|
|
[2026-04-13 13:05:18] UCB=2.6299 mu=1.2561 sigma=0.6869 params={'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0015652851488979931, 'timesteps': 4530}
|
|
[2026-04-13 13:05:18] UCB=2.4296 mu=1.2377 sigma=0.5960 params={'n_steer': 7, 'n_throttle': 4, 'learning_rate': 0.0005256664341026775, 'timesteps': 4983}
|
|
[2026-04-13 13:05:18] UCB=2.3451 mu=0.9859 sigma=0.6796 params={'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0005171395498457057, 'timesteps': 4013}
|
|
[2026-04-13 13:05:18] UCB=2.3171 mu=1.0474 sigma=0.6348 params={'n_steer': 7, 'n_throttle': 3, 'learning_rate': 0.0021625928193495006, 'timesteps': 4782}
|
|
[2026-04-13 13:05:18] [AutoResearch] Proposed: {'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0010464507674264373, 'timesteps': 4450, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:05:20] [AutoResearch] Launching trial 10: {'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0010464507674264373, 'timesteps': 4450, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:11:06] [AutoResearch] Trial 10 finished in 346.0s, returncode=0
|
|
[2026-04-13 13:11:06] [AutoResearch] Trial 10: mean_reward=682.7352 std_reward=420.9113
|
|
[2026-04-13 13:11:06] [AutoResearch] === Trial 10 Summary ===
|
|
[2026-04-13 13:11:06] Total Phase 1 runs: 10
|
|
[2026-04-13 13:11:06] 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 13:11:06] Top 5:
|
|
[2026-04-13 13:11:06] 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 13:11:06] mean_reward=682.7352 params={'n_steer': 7, 'n_throttle': 2, 'learning_rate': 0.0010464507674264373, 'timesteps': 4450, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:11:06] mean_reward=627.6915 params={'n_steer': 9, 'n_throttle': 3, 'learning_rate': 0.0009549126527603771, 'timesteps': 4279, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:11:06] mean_reward=454.0640 params={'n_steer': 6, 'n_throttle': 3, 'learning_rate': 0.0005165618383365869, 'timesteps': 4929, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|
|
[2026-04-13 13:11:06] mean_reward=306.1739 params={'n_steer': 8, 'n_throttle': 3, 'learning_rate': 0.0003097316245852375, 'timesteps': 4938, 'agent': 'ppo', 'eval_episodes': 3, 'reward_shaping': True}
|