CANN GR00T N1.6昇腾推理

概述【免费下载链接】cann-recipes-embodied-intelligence本项目针对具身智能业务中的典型模型、加速算法提供基于CANN平台的优化样例项目地址: https://gitcode.com/cann/cann-recipes-embodied-intelligenceGR00T N1.6是2025年12月发布的通用机器人基础模型旨在突破传统VLA模型在长程具身操作任务中的泛化瓶颈解决小样本场景下动作生成的时序一致性问题通过升级 VLM 模块与动作预测范式实现从短程静态桌面操作到动态长程具身任务的能力提升降低通用人形机器人在真实场景落地的工程化门槛。支持的产品型号Atlas A3 系列产品环境准备安装CANN环境 本样例的编译执行依赖CANN开发套件包cann-toolkit与CANN二进制算子包cann-kernels支持的CANN软件版本为CANN 8.3.RC1请从软件包下载地址下载Ascend-cann-toolkit_${version}_linux-${arch}.run与Atlas-A3-cann-kernels_${version}_linux-${arch}.run软件包并参考CANN安装文档进行安装。${version}表示CANN包版本号如8.3.RC1。${arch}表示CPU架构如aarch64、x86_64。克隆仓库 GR00T依赖某些依赖的子模块。克隆时需要包含它们git clone --recurse-submodules https://github.com/NVIDIA/Isaac-GR00T cd Isaac-GR00T git checkout e29d8fc50b0e4745120ae3fb72447986fe638aa6 cd .. git clone https://gitcode.com/cann/cann-recipes-embodied-intelligence.git复制和替换cann-recipes-embodied-intelligence代码仓manipulation/Isaac-GR00T中所有文件到官方代码仓Isaac-GR00T中cp -rf cann-recipes-embodied-intelligence/manipulation/Isaac-GR00T/* Isaac-GR00T cd Isaac-GR00T环境设置本样例提供conda和uv两种环境管理方式进行环境配置。conda管理conda create -n gr00t python3.10 -y conda activate gr00t pip install -r requirements.txt chmod x set_conda_env.sh ./set_conda_env.shuv管理说明解析pyproject.toml中的[tool.uv.extra-build-dependencies]需要 uv v0.8.4。创建环境并安装GR00Tuv sync --python 3.10安装环境所需的ffmpeg和decord库chmod x setup.sh ./setup.sh export LD_LIBRARY_PATH$(pwd)/.venv/lib:$LD_LIBRARY_PATH安装decorator库uv pip install decorator激活uv环境source .venv/bin/activate推理执行数据准备demo_data下提供样例数据集请确保您已经git完整的数据集。您也可以自行准备数据集详情请参阅数据准备指南。快速推理准备好数据后GR00T模型可使用以下脚本生成输出动作如果您使用的是conda进行环境配置使用命令python scripts/deployment/standalone_inference_script.py \ --model-path nvidia/GR00T-N1.6-3B \ --dataset-path demo_data/gr1.PickNPlace \ --embodiment-tag GR1 \ --traj-ids 2 \ --video-backend decord \ --seed 42 \ --action-horizon 8如果您使用uv进行环境管理使用命令uv run python scripts/deployment/standalone_inference_script.py \ --model-path nvidia/GR00T-N1.6-3B \ --dataset-path demo_data/gr1.PickNPlace \ --embodiment-tag GR1 \ --traj-ids 2 \ --video-backend decord \ --seed 42 \ --action-horizon 8运行上述命令后模型开始推理并打印出机器人left_arm、right_arm、left_hand、right_hand、waist等部位的运动指令。Citationinproceedings{gr00tn1_2025, archivePrefix {arxiv}, eprint {2503.14734}, title {{GR00T} {N1}: An Open Foundation Model for Generalist Humanoid Robots}, author {NVIDIA and Johan Bjorck and Fernando Castañeda, Nikita Cherniadev and Xingye Da and Runyu Ding and Linxi Jim Fan and Yu Fang and Dieter Fox and Fengyuan Hu and Spencer Huang and Joel Jang and Zhenyu Jiang and Jan Kautz and Kaushil Kundalia and Lawrence Lao and Zhiqi Li and Zongyu Lin and Kevin Lin and Guilin Liu and Edith Llontop and Loic Magne and Ajay Mandlekar and Avnish Narayan and Soroush Nasiriany and Scott Reed and You Liang Tan and Guanzhi Wang and Zu Wang and Jing Wang and Qi Wang and Jiannan Xiang and Yuqi Xie and Yinzhen Xu and Zhenjia Xu and Seonghyeon Ye and Zhiding Yu and Ao Zhang and Hao Zhang and Yizhou Zhao and Ruijie Zheng and Yuke Zhu}, month {March}, year {2025}, booktitle {ArXiv Preprint}, }附录根目录相关代码目录树检查整体代码目录树经过上述的复制操作GR00T N1.6适配昇腾的根目录中的最终相关代码目录树如下所示. ├── adaptor_patches │ ├── dit_patch.py │ ├── gr00t_n1d6_patch.py │ ├── gr00t_policy_patch.py │ └── modeling_siglip2_patch.py ├── pyproject.toml ├── README.md ├── scripts │ └── deployment │ ├── model_adaptor.py │ └── standalone_inference_script.py ├── requirements.txt ├── setup.sh └── set_conda_env.sh【免费下载链接】cann-recipes-embodied-intelligence本项目针对具身智能业务中的典型模型、加速算法提供基于CANN平台的优化样例项目地址: https://gitcode.com/cann/cann-recipes-embodied-intelligence创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考