{ "dataset": { "repo_id": "fecasado/Ncubes-to-Nbaskets-320x240", "root": null, "episodes": null, "image_transforms": { "enable": true, "max_num_transforms": 1, "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, "video_backend": "torchcodec", "return_uint8": false, "streaming": false }, "env": null, "policy": { "type": "gaze_flow_matching", "n_obs_steps": 1, "input_features": { "observation.images.right": { "type": "VISUAL", "shape": [ 240, 320, 3 ] }, "observation.images.left": { "type": "VISUAL", "shape": [ 240, 320, 3 ] }, "observation.images.user": { "type": "VISUAL", "shape": [ 240, 320, 3 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 26 ] } }, "device": "cuda", "use_amp": false, "use_peft": false, "push_to_hub": true, "repo_id": "fecasado/gfm-cubes-26ap1", "private": null, "tags": null, "license": null, "pretrained_path": null, "normalization_mapping": { "VISUAL": "MEAN_STD", "STATE": "MEAN_STD", "ACTION": "MEAN_STD" }, "gaze_feature_dim": 3, "gaze_state_indices": [ -3, -2, -1 ], "gaze_temporal_window": 1, "vision_backbone": "resnet18", "vit_token_size": 16, "vit_foveation_strides": [ 1, 2, 6 ], "vit_foveation_grid_sizes": [ 2, 3, 3 ], "vit_model_size": "small", "vit_pretrained_repo_id": "iantc104/mae_vitb_foveated_vit", "vit_freeze": true, "vit_backbone": "vit_foveated", "gaze_cross_camera": false, "gaze_cross_camera_method_train": "roma", "gaze_cross_camera_method_inference": "roma", "gaze_cross_camera_match_every": 5, "gaze_cross_camera_timeout": 0.05, "gaze_cross_camera_cache": null, "obs_step_size": 1, "n_action_steps": 2, "horizon": 16, "n_sampling_steps": 8, "use_temporal_ensemble": true, "temporal_ensemble_coeff": -0.01, "dim_model": 512, "n_heads": 8, "mlp_ratio": 4.0, "dropout": 0.1, "dit_n_layers": 8, "dit_time_dim": 256, "pool_n_queries": 16, "pool_n_layers": 4, "use_gaze": true, "use_gaze_as_action": false, "gaze_model_repo_id": "", "gaze_noise": 0.01, "image_norm_mode": "mean_std", "state_norm_mode": "min_max", "action_norm_mode": "min_max", "image_to_gaze_key": {}, "optimizer_lr": 0.0001, "optimizer_lr_backbone": 0.0001, "optimizer_betas": [ 0.95, 0.999 ], "optimizer_eps": 1e-08, "optimizer_weight_decay": 1e-06, "scheduler_name": "cosine", "scheduler_warmup_steps": 500, "use_ema": true, "ema_decay": 0.99, "input_shape": [ 240, 320 ], "visualize": true }, "reward_model": null, "output_dir": "outputs/train/gfm-cubes-26ap1", "job_name": "gfm-cubes-26ap1", "resume": false, "seed": 1000, "cudnn_deterministic": false, "num_workers": 15, "batch_size": 128, "prefetch_factor": 4, "persistent_workers": true, "steps": 100000, "eval_freq": 20000, "log_freq": 200, "tolerance_s": 0.0001, "save_checkpoint": true, "save_freq": 10000, "use_policy_training_preset": true, "optimizer": { "type": "adamw", "lr": 0.0001, "weight_decay": 1e-06, "grad_clip_norm": 10.0, "betas": [ 0.95, 0.999 ], "eps": 1e-08 }, "scheduler": null, "eval": { "n_episodes": 50, "batch_size": 11, "use_async_envs": true }, "wandb": { "enable": true, "disable_artifact": true, "project": "lerobot", "entity": null, "notes": null, "run_id": "9qhibdvs", "mode": null, "add_tags": true }, "peft": null, "sample_weighting": null, "rename_map": {}, "checkpoint_path": null }