{ "dataset": { "repo_id": "HuggingFaceVLA/libero", "root": null, "episodes": null, "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 ] } } } }, "revision": null, "use_imagenet_stats": true, "video_backend": "torchcodec", "streaming": false }, "env": { "type": "libero", "task": "libero_10", "fps": 30, "features": { "action": { "type": "ACTION", "shape": [ 7 ] }, "agent_pos": { "type": "STATE", "shape": [ 8 ] }, "pixels/agentview_image": { "type": "VISUAL", "shape": [ 360, 360, 3 ] }, "pixels/robot0_eye_in_hand_image": { "type": "VISUAL", "shape": [ 360, 360, 3 ] } }, "features_map": { "action": "action", "agent_pos": "observation.state", "pixels/agentview_image": "observation.images.image", "pixels/robot0_eye_in_hand_image": "observation.images.image2" }, "max_parallel_tasks": 1, "disable_env_checker": true, "episode_length": 520, "obs_type": "pixels_agent_pos", "render_mode": "rgb_array", "camera_name": "agentview_image,robot0_eye_in_hand_image", "init_states": true, "camera_name_mapping": null }, "policy": { "type": "dic_vla", "n_obs_steps": 1, "input_features": { "observation.images.image": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.images.image2": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.state": { "type": "STATE", "shape": [ 8 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 7 ] } }, "device": "cuda", "use_amp": false, "push_to_hub": true, "repo_id": "radiuson/Dic-libero-test", "private": null, "tags": null, "license": null, "pretrained_path": null, "chunk_size": 64, "n_action_steps": 64, "normalization_mapping": { "VISUAL": "IDENTITY", "STATE": "MEAN_STD", "ACTION": "MEAN_STD" }, "max_state_dim": 14, "max_action_dim": 14, "resize_imgs_with_padding": [ 512, 512 ], "empty_cameras": 0, "vlm_model_name": "HuggingFaceTB/SmolVLM-256M-Instruct", "load_vlm_weights": false, "num_vlm_layers": 16, "freeze_vision_encoder": true, "train_state_proj": true, "dic_config": "dic-tiny", "dic_base_channels": null, "dic_num_groups": null, "dic_encoder_blocks": null, "dic_bottleneck_blocks": null, "dic_decoder_blocks": null, "dic_dropout": 0.2, "dic_sparse_skip_stride": 1, "num_inference_steps": 4, "min_period": 1.0, "max_period": 10000.0, "condition_pooling": "mean", "condition_dim": null, "optimizer_lr": 0.0001, "optimizer_betas": [ 0.9, 0.95 ], "optimizer_eps": 1e-08, "optimizer_weight_decay": 1e-06, "optimizer_grad_clip_norm": 1.0, "scheduler_warmup_steps": 1000, "scheduler_decay_steps": 30000, "scheduler_decay_lr": 1e-06, "tokenizer_max_length": 48, "add_image_special_tokens": false, "prefix_length": -1, "use_cache": true, "pad_language_to": "longest", "adapt_to_pi_aloha": false, "use_delta_joint_actions_aloha": false }, "output_dir": "test_outputs", "job_name": "libero_dic_vla", "resume": false, "seed": 1000, "num_workers": 4, "batch_size": 1, "steps": 10, "eval_freq": 100000, "log_freq": 200, "save_checkpoint": true, "save_freq": 20000, "use_policy_training_preset": true, "optimizer": { "type": "adamw", "lr": 0.0001, "weight_decay": 1e-06, "grad_clip_norm": 1.0, "betas": [ 0.9, 0.95 ], "eps": 1e-08 }, "scheduler": { "type": "cosine_decay_with_warmup", "num_warmup_steps": 1000, "num_decay_steps": 30000, "peak_lr": 0.0001, "decay_lr": 1e-06 }, "eval": { "n_episodes": 1, "batch_size": 1, "use_async_envs": false }, "wandb": { "enable": false, "disable_artifact": false, "project": "lerobot", "entity": null, "notes": null, "run_id": null, "mode": null } }