AHM_IL2

这是一个使用 LeRobot 训练的 ACT 模型。

模型信息

• 模型类型: ACT
• 训练框架: LeRobot

使用方法

from lerobot.policies.act.model import ACTPolicy
from lerobot.configs.policies import PreTrainedConfig

# 加载模型
config = PreTrainedConfig(
    repo_id="Junzhe1/AHM_IL2",
    commit="main"  # 或指定特定的checkpoint tag
)
policy = ACTPolicy.from_pretrained(config.repo_id)

训练配置

{
  "type": "act",
  "n_obs_steps": 1,
  "input_features": {
    "observation.state": {
      "type": "STATE",
      "shape": [
        7
      ]
    },
    "observation.images.realsense_wrist": {
      "type": "VISUAL",
      "shape": [
        3,
        480,
        640
      ]
    },
    "observation.images.realsense_front": {
      "type": "VISUAL",
      "shape": [
        3,
        480,
        640
      ]
    }
  },
  "output_features": {
    "action": {
      "type": "ACTION",
      "shape": [
        7
      ]
    }
  },
  "device": "cuda",
  "use_amp": true,
  "push_to_hub": false,
  "repo_id": null,
  "private": null,
  "tags": null,
  "license": null,
  "pretrained_path": null,
  "chunk_size": 100,
  "n_action_steps": 100,
  "normalization_mapping": {
    "VISUAL": "MEAN_STD",
    "STATE": "MEAN_STD",
    "ACTION": "MEAN_STD"
  },
  "vision_backbone": "resnet18",
  "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
  "replace_final_stride_with_dilation": false,
  "pre_norm": false,
  "dim_model": 512,
  "n_heads": 8,
  "dim_feedforward": 3200,
  "feedforward_activation": "relu",
  "n_encoder_layers": 4,
  "n_decoder_layers": 1,
  "use_vae": true,
  "latent_dim": 32,
  "n_vae_encoder_layers": 4,
  "temporal_ensemble_coeff": null,
  "dropout": 0.1,
  "kl_weight": 10.0,
  "optimizer_lr": 1e-05,
  "optimizer_weight_decay": 0.0001,
  "optimizer_lr_backbone": 1e-05
}

引用

如果使用此模型，请引用 LeRobot:

@misc{lerobot2024,
  title={LeRobot: A Versatile Framework for Robot Learning},
  author={LeRobot Team},
  year={2024},
  howpublished={\url{https://github.com/huggingface/lerobot}}
}