| config: {'dataset': {'repo_id': 'daniecraig/libero_90_no_noops_lerobot_v30', 'root': None, 'episodes': [3092, 3093, 3094, 3095, 3096, 3097, 3098, 3099, 3100, 3101, 3102, 3103, 3104, 3105, 3106, 3107, 3108, 3109, 3110, 3111, 3112, 3113, 3114, 3115, 3116, 3117, 3118, 3119, 3120, 3121, 3122, 3123, 3124, 3125, 3126, 3127, 3128, 3129, 3130, 3131, 3132, 3133, 3134, 3135, 3136], 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': 'main', 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'streaming': False}, 'env': {'type': 'libero', 'task': 'libero_90', 'fps': 30, 'features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [7]}, 'pixels/agentview_image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [360, 360, 3]}, 'pixels/robot0_eye_in_hand_image': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [360, 360, 3]}, 'robot_state/eef/pos': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [3]}, 'robot_state/eef/quat': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [4]}, 'robot_state/eef/mat': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [3, 3]}, 'robot_state/gripper/qpos': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [2]}, 'robot_state/gripper/qvel': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [2]}, 'robot_state/joints/pos': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [7]}, 'robot_state/joints/vel': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [7]}}, 'features_map': {'action': 'action', 'robot_state/eef/pos': 'observation.state.eef_pos', 'robot_state/eef/quat': 'observation.state.eef_quat', 'robot_state/eef/mat': 'observation.state.eef_mat', 'robot_state/gripper/qpos': 'observation.state.gripper_qpos', 'robot_state/gripper/qvel': 'observation.state.gripper_qvel', 'robot_state/joints/pos': 'observation.state.joint_pos', 'robot_state/joints/vel': 'observation.state.joint_vel', 'pixels/agentview_image': 'observation.images.image', 'pixels/robot0_eye_in_hand_image': 'observation.images.image2'}, 'max_parallel_tasks': 1, 'disable_env_checker': True, 'task_ids': [48], 'episode_length': None, 'obs_type': 'pixels_agent_pos', 'render_mode': 'rgb_array', 'camera_name': 'agentview_image,robot0_eye_in_hand_image', 'init_states': True, 'camera_name_mapping': None, 'observation_height': 360, 'observation_width': 360, 'control_mode': 'relative'}, 'policy': {'type': 'smolvla', 'n_obs_steps': 1, 'input_features': None, 'output_features': None, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': True, 'repo_id': 'ardalon/sequential_t1_task48_fft', 'private': None, 'tags': None, 'license': None, 'pretrained_path': '/content/smolvla_base', 'chunk_size': 50, 'n_action_steps': 50, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.MEAN_STD: 'MEAN_STD'>, 'ACTION': <NormalizationMode.MEAN_STD: 'MEAN_STD'>}, 'max_state_dim': 32, 'max_action_dim': 32, 'resize_imgs_with_padding': [512, 512], 'empty_cameras': 0, 'adapt_to_pi_aloha': False, 'use_delta_joint_actions_aloha': False, 'tokenizer_max_length': 48, 'num_steps': 10, 'use_cache': True, 'freeze_vision_encoder': True, 'train_expert_only': True, 'train_state_proj': True, 'optimizer_lr': 2e-05, 'optimizer_betas': [0.9, 0.95], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 0.01, 'optimizer_grad_clip_norm': 10.0, 'scheduler_warmup_steps': 1000, 'scheduler_decay_steps': 30000, 'scheduler_decay_lr': 2.5e-06, 'vlm_model_name': 'HuggingFaceTB/SmolVLM2-500M-Video-Instruct', 'load_vlm_weights': True, 'add_image_special_tokens': False, 'attention_mode': 'cross_attn', 'prefix_length': 0, 'pad_language_to': 'max_length', 'num_expert_layers': 0, 'num_vlm_layers': 16, 'self_attn_every_n_layers': 2, 'expert_width_multiplier': 0.75, 'min_period': 0.004, 'max_period': 4.0, 'rtc_config': None, 'compile_model': False, 'compile_mode': 'max-autotune'}, 'output_dir': 'outputs/train/sequential_t1_task48_fft', 'job_name': 'sequential_t1_task48_fft', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 16, 'steps': 3000, 'eval_freq': 0, 'log_freq': 1, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 500, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 2e-05, 'weight_decay': 0.01, 'grad_clip_norm': 10.0, 'betas': [0.9, 0.95], 'eps': 1e-08}, 'scheduler': {'type': 'cosine_decay_with_warmup', 'num_warmup_steps': 1000, 'num_decay_steps': 30000, 'peak_lr': 2e-05, 'decay_lr': 2.5e-06}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot-smolvla', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'use_rabc': False, 'rabc_progress_path': None, 'rabc_kappa': 0.01, 'rabc_epsilon': 1e-06, 'rabc_head_mode': 'sparse', 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}} |
