INFO 2026-06-21 16:51:24 ot_train.py:197 {'batch_size': 8, 'checkpoint_path': None, 'cudnn_deterministic': False, 'dataset': {'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'affine': {'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}, 'type': 'RandomAffine', 'weight': 1.0}, 'brightness': {'kwargs': {'brightness': [0.8, 1.2]}, 'type': 'ColorJitter', 'weight': 1.0}, 'contrast': {'kwargs': {'contrast': [0.8, 1.2]}, 'type': 'ColorJitter', 'weight': 1.0}, 'hue': {'kwargs': {'hue': [-0.05, 0.05]}, 'type': 'ColorJitter', 'weight': 1.0}, 'saturation': {'kwargs': {'saturation': [0.5, 1.5]}, 'type': 'ColorJitter', 'weight': 1.0}, 'sharpness': {'kwargs': {'sharpness': [0.5, 1.5]}, 'type': 'SharpnessJitter', 'weight': 1.0}}}, 'repo_id': '3a279/piper_place_block_cam0fix_random500', 'revision': None, 'root': '/workspace/datasets/piper_place_block_cam0fix_random500_lerobot', 'streaming': False, 'use_imagenet_stats': True, 'video_backend': 'torchcodec'}, 'env': None, 'eval': {'batch_size': 50, 'n_episodes': 50, 'use_async_envs': False}, 'eval_freq': 20000, 'job_name': 'piper_place_block_cam0fix_bs8_50k', 'log_freq': 200, 'num_workers': 4, 'optimizer': {'betas': [0.95, 0.999], 'eps': 1e-08, 'grad_clip_norm': 10.0, 'lr': 0.0001, 'type': 'adam', 'weight_decay': 1e-06}, 'output_dir': '/workspace/outputs/piper_place_block_cam0fix_bs8_50k', 'peft': None, 'policy': {'beta_end': 0.02, 'beta_schedule': 'squaredcos_cap_v2', 'beta_start': 0.0001, 'clip_sample': True, 'clip_sample_range': 1.0, 'compile_mode': 'reduce-overhead', 'compile_model': False, 'crop_is_random': True, 'crop_ratio': 1.0, 'crop_shape': None, 'device': 'cuda', 'diffusion_step_embed_dim': 128, 'do_mask_loss_for_padding': False, 'down_dims': [512, 1024, 2048], 'drop_n_last_frames': 7, 'horizon': 16, 'input_features': {}, 'kernel_size': 5, 'license': None, 'n_action_steps': 8, 'n_groups': 8, 'n_obs_steps': 2, 'noise_scheduler_type': 'DDPM', 'normalization_mapping': {'ACTION': , 'STATE': , 'VISUAL': }, 'num_inference_steps': None, 'num_train_timesteps': 100, 'optimizer_betas': [0.95, 0.999], 'optimizer_eps': 1e-08, 'optimizer_lr': 0.0001, 'optimizer_weight_decay': 1e-06, 'output_features': {}, 'prediction_type': 'epsilon', 'pretrained_backbone_weights': None, 'pretrained_path': None, 'private': None, 'push_to_hub': False, 'repo_id': '3a279/piper_place_block_cam0fix_bs8_50k', 'resize_shape': None, 'scheduler_name': 'cosine', 'scheduler_warmup_steps': 500, 'spatial_softmax_num_keypoints': 32, 'tags': None, 'type': 'diffusion', 'use_amp': False, 'use_film_scale_modulation': True, 'use_group_norm': True, 'use_peft': False, 'use_separate_rgb_encoder_per_camera': False, 'vision_backbone': 'resnet18'}, 'rabc_epsilon': 1e-06, 'rabc_head_mode': 'sparse', 'rabc_kappa': 0.01, 'rabc_progress_path': None, 'rename_map': {}, 'resume': False, 'save_checkpoint': True, 'save_freq': 10000, 'scheduler': {'name': 'cosine', 'num_warmup_steps': 500, 'type': 'diffuser'}, 'seed': 1000, 'steps': 50000, 'tolerance_s': 0.0001, 'use_policy_training_preset': True, 'use_rabc': False, 'wandb': {'add_tags': True, 'disable_artifact': False, 'enable': False, 'entity': None, 'mode': None, 'notes': None, 'project': 'lerobot', 'run_id': None}} INFO 2026-06-21 16:51:24 ot_train.py:205 Logs will be saved locally. INFO 2026-06-21 16:51:24 ot_train.py:221 Creating dataset INFO 2026-06-21 16:51:25 eo_utils.py:106 Using video codec: libsvtav1 INFO 2026-06-21 16:51:28 ot_train.py:239 Creating policy INFO 2026-06-21 16:51:32 ot_train.py:294 Creating optimizer and scheduler INFO 2026-06-21 16:51:32 ot_train.py:329 Output dir: /workspace/outputs/piper_place_block_cam0fix_bs8_50k INFO 2026-06-21 16:51:32 ot_train.py:336 cfg.steps=50000 (50K) INFO 2026-06-21 16:51:32 ot_train.py:337 dataset.num_frames=49573 (50K) INFO 2026-06-21 16:51:32 ot_train.py:338 dataset.num_episodes=500 INFO 2026-06-21 16:51:32 ot_train.py:341 Effective batch size: 8 x 1 = 8 INFO 2026-06-21 16:51:32 ot_train.py:342 num_learnable_params=270353703 (270M) INFO 2026-06-21 16:51:32 ot_train.py:343 num_total_params=270353703 (270M) Training: 0%| | 0/50000 [00:00