config.json

{
    "type": "pi0fast",
    "n_obs_steps": 1,
    "normalization_mapping": {
        "VISUAL": "IDENTITY",
        "STATE": "MEAN_STD",
        "ACTION": "MEAN_STD"
    },
    "input_features": {
        "observation.image": {
            "type": "VISUAL",
            "shape": [
                3,
                256,
                256
            ]
        },
        "observation.image2": {
            "type": "VISUAL",
            "shape": [
                3,
                256,
                256
            ]
        },
        "observation.image3": {
            "type": "VISUAL",
            "shape": [
                3,
                256,
                256
            ]
        },
        "observation.state": {
            "type": "STATE",
            "shape": [
                8
            ]
        }
    },
    "output_features": {
        "action": {
            "type": "ACTION",
            "shape": [
                7
            ]
        }
    },
    "chunk_size": 10,
    "n_action_steps": 5,
    "max_state_dim": 32,
    "max_action_dim": 32,
    "resize_imgs_with_padding": [
        224,
        224
    ],
    "interpolate_like_pi": false,
    "empty_cameras": 0,
    "adapt_to_pi_aloha": false,
    "use_delta_joint_actions_aloha": false,
    "tokenizer_max_length": 48,
    "proj_width": 1024,
    "max_decoding_steps": 256,
    "fast_skip_tokens": 128,
    "max_input_seq_len": 256,
    "use_cache": true,
    "freeze_vision_encoder": true,
    "freeze_lm_head": true,
    "optimizer_lr": 0.0001,
    "optimizer_betas": [
        0.9,
        0.95
    ],
    "optimizer_eps": 1e-08,
    "optimizer_weight_decay": 1e-5,
    "scheduler_warmup_steps": 1000,
    "scheduler_decay_steps": 30000,
    "scheduler_decay_lr": 2.5e-06
}