{ "dataset": { "repo_id": "cetiennec/yellow_on_purple_super", "repo_type": "dataset", "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": null, "use_imagenet_stats": true, "video_backend": "torchcodec", "return_uint8": false, "depth_output_unit": "mm", "streaming": false, "eval_split": 0.2 }, "env": null, "policy": { "type": "flux3", "n_obs_steps": 1, "input_features": { "observation.images.scene": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.images.wrist": { "type": "VISUAL", "shape": [ 3, 256, 256 ] }, "observation.state": { "type": "STATE", "shape": [ 6 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 6 ] } }, "device": "cuda", "use_amp": false, "use_peft": true, "push_to_hub": false, "repo_id": null, "private": null, "tags": null, "license": null, "pretrained_path": "/fsx/caroline_pascal/models/so101", "pretrained_revision": null, "conditioning": "history", "packer": null, "condition_on_past_actions": false, "action_representation": "delta", "delta_absolute_dims": [ -1 ], "text_fixed_length": 320, "video_position_fps": 24.0, "history_snapshots": 1, "normalization_stats": null, "normalization_clip": 6.0, "separate_timesteps": true, "video_logit_mean": 1.08, "video_logit_std": 1.0, "conditioning_noise_max": 0.2, "loss_reduction": "modalities", "action_channel_weights": [ 1.0, 1.0, 1.0, 1.0, 1.0, 2.0 ], "gradient_checkpointing": true, "chunk_size": 32, "n_action_steps": 30, "fps": 30.0, "camera_layout": "side_by_side", "camera_keys": [ "observation.images.scene", "observation.images.wrist" ], "canvas_hw": [ 256, 512 ], "action_modality": "action", "action_scale": 2.0, "gripper_flip_dims": [], "use_relative_actions": false, "relative_exclude_joints": [ "gripper" ], "action_feature_names": [ "shoulder_pan.pos", "shoulder_lift.pos", "elbow_flex.pos", "wrist_flex.pos", "wrist_roll.pos", "gripper.pos" ], "normalization_mapping": { "VISUAL": "IDENTITY", "STATE": "IDENTITY", "ACTION": "IDENTITY" }, "trunk_weights": null, "video_vae_id": "/fsx/caroline_pascal/models/so101/components/video_vae.safetensors", "text_encoder_id": "/fsx/caroline_pascal/models/so101/components/text_encoder", "dit_config": null, "content_streams": [ "video", "video_cond" ], "attn_mode": "torch", "dtype": "bfloat16", "compile_model": false, "sampler": "cosmos_unipc", "num_inference_steps": 4, "guidance_scale": 3.0, "guidance_scale_action": null, "sampler_shift": 5.0, "inference_seed": 0, "head_init_seed": 0, "train_timestep_width": 0.75, "train_timestep_shift": 42.0, "action_loss_weight": 0.5, "video_loss_weight": 1.0, "caption_dropout": 0.1, "augment": false, "optimizer_lr": 0.0001, "optimizer_lr_heads_multiplier": 5.0, "optimizer_betas": [ 0.9, 0.999 ], "optimizer_eps": 1e-08, "optimizer_weight_decay": 0.0, "optimizer_grad_clip_norm": 1.0, "scheduler_freeze_backbone_steps": 0, "scheduler_warmup_steps": 0, "scheduler_warmup_steps_heads": 0, "scheduler_decay_steps": 0, "scheduler_cooldown_steps": 0 }, "reward_model": null, "output_dir": "/fsx/caroline_pascal/outputs/yellow_on_purple_lora_4gpu", "job_name": "flux3", "resume": false, "seed": 42, "cudnn_deterministic": false, "num_workers": 8, "batch_size": 2, "prefetch_factor": 4, "persistent_workers": true, "dataloader_multiprocessing_context": "spawn", "steps": 10000, "env_eval_freq": 0, "log_freq": 20, "eval_steps": 500, "max_eval_samples": 0, "tolerance_s": 0.0001, "save_checkpoint": true, "save_freq": 500, "checkpoint_format": "safetensors", "use_policy_training_preset": true, "optimizer": { "type": "adamw", "lr": 0.0001, "weight_decay": 0.0, "grad_clip_norm": 1.0, "betas": [ 0.9, 0.999 ], "eps": 1e-08 }, "scheduler": { "type": "frozen_warmup_constant", "num_warmup_steps": 0, "freeze_steps": 0, "warmup_steps_heads": 0, "decay_steps": 0, "cooldown_steps": 0 }, "parallelism": { "dp_replicate": 4, "dp_shard": 1, "context_parallel": { "ring_degree": 1, "ulysses_degree": 1 }, "cfg_parallel": 1 }, "accelerator": { "mixed_precision": "bf16", "gradient_accumulation": { "steps": 1 }, "fsdp": { "reshard_after_forward": true, "wrap_modules": null, "min_num_params": null, "cpu_offload": false, "ignored_modules": null }, "ddp": { "find_unused_parameters": true, "gradient_as_bucket_view": false, "static_graph": false }, "compile": { "enabled": false, "backend": "inductor", "mode": null, "regional": true }, "activation_checkpointing": { "mode": "none" } }, "eval": { "n_episodes": 50, "batch_size": 50, "use_async_envs": true, "recording": false, "recording_repo_id": null, "recording_private": false }, "ema": { "enable": true, "decay": 0.999, "update_after_step": 0, "inv_gamma": 1.0, "power": 0.75, "min_decay": 0.0, "max_decay": 0.9999, "use_for_eval": true }, "wandb": { "enable": true, "disable_artifact": false, "project": "flux3-so101", "entity": null, "notes": null, "run_id": "yrahmxpq", "mode": null, "add_tags": true }, "peft": { "target_modules": null, "full_training_modules": null, "method_type": "LORA", "init_type": null, "r": 32, "lora_alpha": 32 }, "job": { "target": null, "image": "huggingface/lerobot-gpu:latest", "timeout": "2d", "detach": false, "tags": [] }, "save_checkpoint_to_hub": false, "sample_weighting": null, "rename_map": { "observation.images.on_top": "observation.images.scene", "observation.images.camera_on_follower": "observation.images.wrist" }, "checkpoint_path": null }