donkeycar-rl-autoresearch/agent/models/eval_gentrack_minimonaco_20...

50 lines
2.8 KiB
Plaintext

[18:49:02] ======================================================================
[18:49:02] Eval: generated-track specialists on mini-monaco (zero-shot)
[18:49:02] Track : donkey-minimonaco-track-v0
[18:49:02] Episodes: 7 x max 2000 steps
[18:49:02] Host : localhost:9091
[18:49:02] Log : /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/eval_gentrack_minimonaco_20260506_184902.log
[18:49:02] ======================================================================
[18:49:02]
[18:49:02] ── exp13-gentrack-v4 ──────────────────────────────────────
[18:49:02] Model: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/exp13-gentrack-v4/best_model.zip
[18:49:31] ep1: 5.5r / 2000s ✅
[18:49:58] ep2: 5.9r / 2000s ✅
[18:50:24] ep3: 5.8r / 2000s ✅
[18:50:51] ep4: 5.6r / 2000s ✅
[18:51:17] ep5: 5.8r / 2000s ✅
[18:51:44] ep6: 5.5r / 2000s ✅
[18:52:10] ep7: 5.5r / 2000s ✅
[18:52:11] SUMMARY: 7/7 full | mean 2000s / 5.7r | ✅ DRIVES
[18:52:13]
[18:52:13] ── wave5-gentrack-only ──────────────────────────────────────
[18:52:13] Model: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/wave5-gentrack-only/model.zip
[18:52:40] ep1: 5.6r / 2000s ✅
[18:53:06] ep2: 5.7r / 2000s ✅
[18:53:33] ep3: 5.9r / 2000s ✅
[18:53:42] ep4: 4.6r / 550s ❌@550
[18:54:09] ep5: 5.9r / 2000s ✅
[18:54:18] ep6: 4.5r / 540s ❌@540
[18:54:45] ep7: 5.8r / 2000s ✅
[18:54:45] SUMMARY: 5/7 full | mean 1584s / 5.4r | ✅ DRIVES
[18:54:47]
[18:54:47] ── wave4-trial-0009 ──────────────────────────────────────
[18:54:47] Model: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/wave4-trial-0009/model.zip
[18:55:01] ep1: 5.3r / 865s ❌@865
[18:55:28] ep2: 6.0r / 2000s ✅
[18:55:42] ep3: 5.0r / 874s ❌@874
[18:55:55] ep4: 5.2r / 858s ❌@858
[18:56:22] ep5: 6.0r / 2000s ✅
[18:56:37] ep6: 5.3r / 845s ❌@845
[18:56:51] ep7: 5.1r / 890s ❌@890
[18:56:52] SUMMARY: 2/7 full | mean 1190s / 5.4r | ✅ DRIVES
[18:56:54]
[18:56:54] ======================================================================
[18:56:54] FINAL RESULTS
[18:56:54] ======================================================================
[18:56:54] exp13-gentrack-v4 7/7 full mean 2000s / 5.7r ✅ DRIVES
[18:56:54] wave5-gentrack-only 5/7 full mean 1584s / 5.4r ✅ DRIVES
[18:56:54] wave4-trial-0009 2/7 full mean 1190s / 5.4r ✅ DRIVES
[18:56:54]
Log saved: /home/paulh/projects/donkeycar-rl-autoresearch/agent/models/eval_gentrack_minimonaco_20260506_184902.log