{ "dataset": { "repo_id": "pepijn223/robocasa_pretrain_human300_v4_annotated5", "root": null, "episodes": null, "exclude_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": "main", "use_imagenet_stats": true, "video_backend": "pyav", "return_uint8": false, "depth_output_unit": "mm", "streaming": false, "eval_split": 0.0 }, "env": null, "policy": { "type": "pi052", "n_obs_steps": 1, "input_features": { "observation.images.robot0_agentview_left": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.images.robot0_agentview_right": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.images.robot0_eye_in_hand": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.state": { "type": "STATE", "shape": [ 16 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 12 ] } }, "device": "cuda", "use_amp": false, "use_peft": false, "push_to_hub": true, "repo_id": "pepijn223/pi052_robocasa_joint_flow1_30k_32gpu", "private": null, "tags": null, "license": null, "pretrained_path": "lerobot/pi05_base", "pretrained_revision": null, "paligemma_variant": "gemma_2b", "action_expert_variant": "gemma_300m", "dtype": "bfloat16", "chunk_size": 20, "n_action_steps": 20, "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, "use_relative_actions": false, "relative_exclude_joints": [ "gripper" ], "action_feature_names": null, "rtc_config": null, "image_resolution": [ 224, 224 ], "empty_cameras": 0, "tokenizer_max_length": 128, "normalization_mapping": { "VISUAL": "IDENTITY", "STATE": "QUANTILES", "ACTION": "QUANTILES" }, "gradient_checkpointing": true, "compile_model": true, "compile_mode": "default", "freeze_vision_encoder": false, "train_expert_only": false, "optimizer_lr": 2.5e-05, "optimizer_betas": [ 0.9, 0.95 ], "optimizer_eps": 1e-08, "optimizer_weight_decay": 1e-10, "optimizer_grad_clip_norm": 1.0, "scheduler_warmup_steps": 1000, "scheduler_decay_steps": 30000, "scheduler_decay_lr": 2.5e-06, "recipe_path": "recipes/subtask_joint.yaml", "apply_chat_template": false, "text_loss_weight": 1.0, "flow_loss_weight": 1.0, "unfreeze_lm_head": true, "plan_dropout_prob": 0.0, "memory_dropout_prob": 0.0, "subtask_dropout_prob": 0.0, "enable_fast_action_loss": true, "action_tokenizer_name": "lerobot/fast-action-tokenizer", "max_action_tokens": 384, "fast_skip_tokens": 1152, "fast_action_loss_weight": 1.0, "subtask_replan_steps": 0, "joint_subtask_conditioning": true, "auto_fit_fast_tokenizer": true, "fast_tokenizer_cache_dir": "~/.cache/lerobot/fast_tokenizers", "fast_tokenizer_fit_samples": 8192, "knowledge_insulation": true, "use_flashrt_adarms": false, "use_compiled_text_ce": true, "use_compiled_vision": false, "use_flex_attention": false, "use_manual_attention": false, "manual_attention_scope": "all", "lm_head_lr_scale": 1.0, "backbone_lr_scale": 1.0, "action_expert_lr_scale": 1.0, "flow_num_repeats": 5, "text_ce_z_loss_weight": 0.0001, "use_flashrt_fp8_mlp": false, "optimizer_foreach": false, "optimizer_fused": true }, "reward_model": null, "output_dir": "/fsx/pepijn/outputs/train/pi052-robocasa-joint-flow1-30k_22374833", "job_name": "pi052-robocasa-joint-flow1-30k", "resume": false, "seed": 1000, "cudnn_deterministic": false, "num_workers": 4, "batch_size": 32, "prefetch_factor": 4, "persistent_workers": true, "steps": 30000, "env_eval_freq": 30000, "log_freq": 100, "eval_steps": 0, "max_eval_samples": 0, "tolerance_s": 0.0001, "save_checkpoint": true, "save_freq": 5000, "use_policy_training_preset": true, "optimizer": { "type": "adamw", "lr": 2.5e-05, "weight_decay": 1e-10, "grad_clip_norm": 1.0, "betas": [ 0.9, 0.95 ], "eps": 1e-08, "foreach": false, "fused": true }, "scheduler": { "type": "cosine_decay_with_warmup", "num_warmup_steps": 1000, "num_decay_steps": 30000, "peak_lr": 2.5e-05, "decay_lr": 2.5e-06 }, "eval": { "n_episodes": 50, "batch_size": 50, "use_async_envs": true, "recording": false, "recording_repo_id": null, "recording_private": false }, "wandb": { "enable": true, "disable_artifact": true, "project": "hirobot", "entity": null, "notes": null, "run_id": "2pmqwdxm", "mode": null, "add_tags": true }, "peft": null, "job": { "target": null, "image": "huggingface/lerobot-gpu:latest", "timeout": "2d", "detach": false, "tags": [] }, "save_checkpoint_to_hub": false, "sample_weighting": null, "rename_map": {}, "checkpoint_path": null }