Loading dataset...
CachedTrajectoryDataset: 256 demos, 8192 samples
  Source: {'cache_root': '/data/libero/ood_objpos_v3', 'benchmark': 'libero_spatial', 'task_ids': [0], 'frame_stride': 3}
  Total: 8192 samples
✓ Train: 7783 samples
✓ Val: 409 samples

Initializing model (type=para)...
Loading DINOv2 model...
✓ DINOv2 backbone is trainable
✓ DINO embedding dim: 384
✓ Learnable start keypoint embedding (dim=384)
✓ Feature convs: 3× Conv2d(3×3) at pred_size=64
✓ Volume   head → (B, 4, 32, 64, 64)
✓ Gripper  head → (B, 4, 2, 64, 64)  [1×1 conv, CE 2-class]
✓ Rotation head → (B, 4, 3, 32, 64, 64)  [1×1 conv, CE]
Trainable parameters: 32,880,520 / 32,880,520 (100.00%)

Computing dataset stats from random subset: 500/8192 samples (seed=42)
Epochs:   0%|          | 2/9999 [04:59<416:01:38, 149.81s/it]  
✓ Saved stats cache: /data/cameron/para_normalized_losses/libero/checkpoints/para_scratch_256demo/dataset_stats.json
✓ Height range from dataset: [0.917348, 1.177162] m
✓ Gripper range from dataset: [-1.000000, 1.000000]
✓ Rotation range (delta rotvec): ['-0.093', '-0.097', '-0.002'] .. ['0.043', '0.220', '0.025']
✓ Reference rotation: ['0.9996', '0.0013', '-0.0283', '-0.0012']
✓ Position range from dataset: ['-0.413', '-0.366', '0.917'] .. ['0.126', '0.273', '1.177']

Starting training for 9999 epochs...
✓ Rotation loss SKIPPED
✓ EMA loss weights initialized (vol=11.8, rot=3.5, grip=0.69)

============================================================
Epoch 0/9999
============================================================
                                                                                                          
Train Loss: 0.4085 (Volume: 0.2136, Gripper: 0.1950, Rotation: 0.0000)
Val - Loss: 0.6046, Volume: 3.7111, Pixel Error: 10.65px, Height Error: 18.898mm, Gripper: 0.2029
✓ Saved best model (val_loss=0.6046)

============================================================
Epoch 1/9999
============================================================

  Saved step checkpoint: step_1000.pth
Train Loss: 0.7874 (Volume: 0.3946, Gripper: 0.3929, Rotation: 0.0000)
Val - Loss: 0.5072, Volume: 4.0395, Pixel Error: 15.45px, Height Error: 29.130mm, Gripper: 0.3313
✓ Saved best model (val_loss=0.5072)

============================================================
Epoch 2/9999
============================================================

  Saved step checkpoint: step_2000.pth
Train Loss: 0.2475 (Volume: 0.1240, Gripper: 0.1235, Rotation: 0.0000)
Val - Loss: 0.3040, Volume: 3.0182, Pixel Error: 10.78px, Height Error: 18.553mm, Gripper: 0.1650
✓ Saved best model (val_loss=0.3040)

============================================================
Epoch 3/9999
============================================================

  Saved step checkpoint: step_3000.pth
Train Loss: 0.1536 (Volume: 0.0773, Gripper: 0.0764, Rotation: 0.0000)
Val - Loss: 0.1820, Volume: 2.3826, Pixel Error: 8.66px, Height Error: 13.225mm, Gripper: 0.0880
✓ Saved best model (val_loss=0.1820)

⏱ Time limit reached (10.2 / 10 min). Stopping.