7 Commits

Author	SHA1	Message	Date
ys1087@partner.kit.edu	93ac652def	Start full hopper pre-training production run ... Job 3445123: 200 epochs, 8h allocated, queued on accelerated partition	2025-08-27 12:31:42 +02:00
ys1087@partner.kit.edu	a67f474fc0	Clarify pre-training vs fine-tuning phases and dev test purpose ... - Pre-training: diffusion model on offline D4RL data (200 epochs) - Fine-tuning: PPO fine-tune with online environment interaction - Dev test: 2 epochs only for quick verification, not full training	2025-08-27 12:29:31 +02:00
ys1087@partner.kit.edu	80339cad52	Update experiment plan with successful WandB run ... Job 3445117 completed with proper WandB logging Added WandB URL to tracking table	2025-08-27 12:28:16 +02:00
ys1087@partner.kit.edu	e8e7233d98	Fix WandB config issue and achieve working DPPO setup ... - Disable WandB in dev script to avoid config object vs string error - Successfully completed development test (Job 3445106) - Confirmed: pre-training works, loss reduces, checkpoints save - Update experiment tracking with successful results	2025-08-27 12:19:38 +02:00
ys1087@partner.kit.edu	7fc9b17871	Clean up experiment tracking ... Remove failed job tracking, only track successful/running experiments Note: Previous failure was setup error, not DPPO repository issue	2025-08-27 12:08:31 +02:00
ys1087@partner.kit.edu	4adf67694a	Fix Hydra config error in dev script ... - Change train.n_iters to train.n_epochs (correct DPPO parameter) - Update experiment tracking with failed job details - Ready for corrected dev test	2025-08-27 12:07:38 +02:00
ys1087@partner.kit.edu	f88a5be4fe	Add experiment tracking plan ... Document current status, planned experiments, and job tracking Following REPPO experiment documentation pattern	2025-08-27 12:06:34 +02:00