{ "dataset": { "type": "qwena1", "repo_id": "EBench/long_horizon_v0.2_v30/bottle EBench/long_horizon_v0.2_v30/detergent EBench/long_horizon_v0.2_v30/dish EBench/long_horizon_v0.2_v30/dishwasher EBench/long_horizon_v0.2_v30/fruit EBench/long_horizon_v0.2_v30/make_sandwich EBench/long_horizon_v0.2_v30/microwave EBench/long_horizon_v0.2_v30/pen EBench/long_horizon_v0.2_v30/shop EBench/simple_pnp_v0.2_v30/task1 EBench/simple_pnp_v0.2_v30/task10 EBench/simple_pnp_v0.2_v30/task2 EBench/simple_pnp_v0.2_v30/task3 EBench/simple_pnp_v0.2_v30/task4 EBench/simple_pnp_v0.2_v30/task5 EBench/simple_pnp_v0.2_v30/task6 EBench/simple_pnp_v0.2_v30/task7 EBench/simple_pnp_v0.2_v30/task8 EBench/simple_pnp_v0.2_v30/task9 EBench/teleop_tasks_v0.2_v30/collect_coffee_beans EBench/teleop_tasks_v0.2_v30/flip_cup_collect_cookies EBench/teleop_tasks_v0.2_v30/frame_against_pen_holder EBench/teleop_tasks_v0.2_v30/install_gear EBench/teleop_tasks_v0.2_v30/peg_in_hole EBench/teleop_tasks_v0.2_v30/put_glass_in_glassbox EBench/teleop_tasks_v0.2_v30/tighten_nut", "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 ] } }, "affine": { "weight": 1.0, "type": "RandomAffine", "kwargs": { "degrees": [ -5.0, 5.0 ], "translate": [ 0.05, 0.05 ] } } } }, "revision": null, "use_imagenet_stats": true, "use_external_stats": true, "external_stats_path": "/mnt/inspurfs/efm-data/huggingface/lerobot/stats/lift2/delta/generalist/stats.json", "weight_rules_path": null, "video_backend": "torchcodec", "streaming": false, "dist_loading": true, "buffer_size": 1024, "action_mode": "delta", "repack_transforms": { "inputs": [], "outputs": [] }, "data_transforms": { "inputs": [ { "type": "delta_action", "mask": null, "mapping": {}, "robot_type": null }, { "type": "resize_with_pad", "height": 224, "width": 224, "mode": "bilinear", "mapping": {} }, { "type": "remap_image_key", "mapping": {} }, { "type": "qwena1_processor", "pretrained_model_name_or_path": "Qwen/Qwen3-VL-2B-Instruct", "max_length": 48, "task_key": "task", "padding_side": "right", "padding": "max_length", "truncation": true, "spatial_merge_size": 2, "vision_start_token_id": 151652, "vision_end_token_id": 151653, "image_token_id": 151655, "process": null }, { "type": "normalize", "selected_keys": null, "mode": "mean_std", "norm_stats": {} }, { "type": "compose_fields", "mapping": {} }, { "type": "pad_state_and_action", "max_state_dim": 32, "max_action_dim": 32 }, { "type": "unify_qwena1_inputs" } ], "outputs": [] }, "model_transforms": { "inputs": [], "outputs": [] }, "height": 224, "width": 224, "max_state_dim": 32, "max_action_dim": 32 }, "policy": { "type": "qwena1", "n_obs_steps": 1, "input_features": { "observation.state": { "type": "STATE", "shape": [ 32 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 32 ] } }, "device": "cuda", "use_amp": false, "push_to_hub": false, "repo_id": "lerobot_lab/qwena1", "private": null, "tags": null, "license": null, "pretrained_path": "/mnt/inspurfs/efm-data/mahaoxiang/outputs/qwena1/2026_04_21_21_29_36-qwena1-a1-finetune-ebench-generalist-20w-128-wo-basestate/checkpoints/150000/pretrained_model", "qwen3_vl_variant": "qwen3_vl_28l", "action_expert_variant": "qwen3_28l", "dtype": "bfloat16", "chunk_size": 50, "n_action_steps": 50, "max_state_dim": 32, "max_action_dim": 32, "num_inference_steps": 10, "time_sampling_beta_alpha": 1.5, "time_sampling_beta_beta": 1.0, "time_sampling_scale": 0.999, "time_sampling_offset": 0.001, "min_period": 0.004, "max_period": 4.0, "image_resolution": [ 224, 224 ], "empty_cameras": 0, "normalization_mapping": { "VISUAL": "IDENTITY", "STATE": "IDENTITY", "ACTION": "IDENTITY" }, "gradient_checkpointing": false, "compile_model": false, "compile_mode": "max-autotune", "optimizer_lr": 5e-05, "optimizer_betas": [ 0.9, 0.95 ], "optimizer_eps": 1e-08, "optimizer_weight_decay": 0.01, "optimizer_grad_clip_norm": 1.0, "scheduler_warmup_steps": 2000, "scheduler_decay_steps": 200000, "scheduler_decay_lr": 5e-06, "tokenizer_max_length": 48, "freeze_vision_encoder": false, "train_expert_only": false, "train_vlm_only": false, "scale_factor": 8, "lambda_gen": 0.01 }, "output_dir": "/mnt/inspurfs/efm-data/mahaoxiang/outputs/qwena1/2026_04_24_15_24_11-qwena1-a1-finetune-ebench-generalist-20w-128-wo-basestate", "job_name": "2026_04_24_15_24_11-qwena1-a1-finetune-ebench-generalist-20w-128-wo-basestate", "resume": true, "seed": 42, "num_workers": 8, "batch_size": 8, "steps": 200000, "eval_freq": 20000, "log_freq": 200, "save_checkpoint": true, "save_freq": 50000, "use_policy_training_preset": true, "optimizer": { "type": "adamw", "lr": 5e-05, "weight_decay": 0.01, "grad_clip_norm": 1.0, "betas": [ 0.9, 0.95 ], "eps": 1e-08 }, "scheduler": { "type": "cosine_decay_with_warmup", "num_warmup_steps": 2000, "num_decay_steps": 200000, "peak_lr": 5e-05, "decay_lr": 5e-06 }, "eval": { "n_episodes": 50, "batch_size": 50, "use_async_envs": false }, "wandb": { "enable": false, "disable_artifact": false, "project": "lerobot_lab_qwena1", "entity": null, "notes": null, "run_id": null, "mode": null }, "checkpoint_path": "/mnt/inspurfs/efm-data/mahaoxiang/outputs/qwena1/2026_04_21_21_29_36-qwena1-a1-finetune-ebench-generalist-20w-128-wo-basestate/checkpoints/150000", "rename_map": {} }