train_config.json

{
    "dataset": {
        "repo_id": "Beilinghamburger/so100_vla_dataset",
        "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": null,
    "policy": {
        "type": "evo1",
        "n_obs_steps": 1,
        "normalization_mapping": {
            "VISUAL": "IDENTITY",
            "STATE": "MEAN_STD",
            "ACTION": "MEAN_STD"
        },
        "input_features": {
            "observation.state": {
                "type": "STATE",
                "shape": [
                    6
                ]
            },
            "observation.images.front": {
                "type": "VISUAL",
                "shape": [
                    3,
                    480,
                    640
                ]
            },
            "observation.images.wrist": {
                "type": "VISUAL",
                "shape": [
                    3,
                    480,
                    640
                ]
            }
        },
        "output_features": {
            "action": {
                "type": "ACTION",
                "shape": [
                    6
                ]
            }
        },
        "device": "cuda",
        "use_amp": false,
        "push_to_hub": true,
        "repo_id": "Beilinghamburger/evo1_so100_vla",
        "private": null,
        "tags": null,
        "license": null,
        "chunk_size": 50,
        "n_action_steps": 50,
        "max_state_dim": 24,
        "max_action_dim": 24,
        "resize_imgs_with_padding": [
            448,
            448
        ],
        "empty_cameras": 0,
        "vlm_model_name": "OpenGVLab/InternVL3-1B",
        "load_vlm_weights": true,
        "add_image_special_tokens": false,
        "attention_mode": "cross_attn",
        "prefix_length": 0,
        "pad_language_to": "max_length",
        "action_head": "flowmatching",
        "dropout": 0.2,
        "num_layers": 8,
        "per_action_dim": 24,
        "state_dim": 24,
        "action_dim": 1200,
        "use_augmentation": true,
        "binarize_gripper": false,
        "freeze_vision_encoder": false,
        "train_expert_only": false,
        "train_state_proj": true,
        "optimizer_lr": 1e-05,
        "optimizer_betas": [
            0.9,
            0.95
        ],
        "optimizer_eps": 1e-08,
        "optimizer_weight_decay": 0.001,
        "optimizer_grad_clip_norm": 1.0,
        "scheduler_warmup_steps": 1000,
        "scheduler_decay_steps": 30000,
        "scheduler_decay_lr": 2.5e-06,
        "batch_size": 16,
        "max_steps": 30000,
        "log_interval": 10,
        "ckpt_interval": 2500,
        "save_dir": "./checkpoints/evo1",
        "resume": true,
        "resume_path": "/home/usyd/Evo-1-e/so100_evo1/checkpoint",
        "resume_pretrain": true,
        "num_workers": 4,
        "disable_wandb": true,
        "return_cls_only": false,
        "run_name": "Evo1_Default",
        "dataset_type": "lerobot",
        "dataset_config_path": "",
        "data_paths": null,
        "num_steps": 10,
        "min_period": 0.004,
        "max_period": 4.0
    },
    "output_dir": "outputs/train/evo1_so100_vla",
    "job_name": "evo1_so100_vla",
    "resume": false,
    "seed": 1000,
    "num_workers": 4,
    "batch_size": 1,
    "steps": 100000,
    "eval_freq": 20000,
    "log_freq": 200,
    "save_checkpoint": true,
    "save_freq": 20000,
    "use_policy_training_preset": true,
    "optimizer": {
        "type": "adamw",
        "lr": 1e-05,
        "weight_decay": 0.001,
        "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": 1e-05,
        "decay_lr": 2.5e-06
    },
    "eval": {
        "n_episodes": 50,
        "batch_size": 50,
        "use_async_envs": false
    },
    "wandb": {
        "enable": false,
        "disable_artifact": false,
        "project": "lerobot",
        "entity": null,
        "notes": null,
        "run_id": null,
        "mode": null
    }
}