{{htmlmetatags>metatag-robots=(index, follow)
metatag-keywords=(anygrasp, 安装部署, MinkowskiEngine, PyTorch安装, gcc9, sklearn报错, Python版本兼容性, numpy版本错误, pointnet2导入错误)
metatag-description=(获取详细的指南，了解如何成功安装和部署AnyGrasp，包括解决MinkowskiEngine、PyTorch版本选择、gcc9编译问题以及常见错误如sklearn、numpy版本错误和pointnet2导入错误的解决方案。)
metatag-media-og:image=(:wiki:anygrasp-install-guide.jpg)
metatag-og:description=(探索如何克服安装AnyGrasp时遇到的各类障碍，从选择合适的Python和PyTorch版本到解决gcc9编译错误及处理依赖库如sklearn、numpy和pointnet2的问题。)
metatag-og:any=(确保您的系统配置满足AnyGrasp的运行要求，并学习如何通过正确的环境设置和依赖管理来避免常见的安装错误。)
}}

====== anygrasp的安装部署 ======

首先，必须安装conda，推荐使用 python10版本。

激活conda环境，然后安装 minkowskiengine [[机器人:minkowskiengine:minkowskiengine安装|minkowskiengine安装]]


根据自己的实际情况安装torch，我这里是4090显卡
<code>
conda install openblas-devel -c anaconda
pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 -f https://download.pytorch.org/whl/torch_stable.html
</code>

后续会用到gcc编译，必须gcc9，不然会有gcc报错，参考 安装 minkowskiengine [[机器人:minkowskiengine:minkowskiengine安装|minkowskiengine安装]]

<code>
/usr/include/c++/11/bits/std_function.h:435:145: error: parameter packs not expanded with ‘...’:
  435 | function(_Functor&& __f)
      | ^
...
error: command '/usr/bin/nvcc' failed with exit code 1
</code>


如果你的网络不佳，这里有加速国内镜像： https://gitee.com/ctbots/mirror-anygrasp_sdk

按照readme安装即可，这里总结一些常见错误：

===== 没有提前申请证书 =====
需要根据https://gitee.com/ctbots/mirror-anygrasp_sdk 的 readme里的说明，有license相关的内容，科学上网，然后填写表格； 会收到

  - 激活码
  - 数据集，分别是checkpoint_detection.tar【290MB】 和 checkpoint_tracking.tar【24MB】

===== pip install 出现sklearn报错 =====

Collecting sklearn (from graspnetAPI->-r requirements.txt (line 5))
  Downloading https://pypi.tuna.tsinghua.edu.cn/packages/46/1c/395a83ee7b2d2ad7a05b453872053d41449564477c81dc356f720b16eac4/sklearn-0.0.post12.tar.gz (2.6 kB)
  Preparing metadata (setup.py) ... error
  error: subprocess-exited-with-error

  × python setup.py egg_info did not run successfully.
  │ exit code: 1
  ╰─> [15 lines of output]
      The 'sklearn' PyPI package is deprecated, use 'scikit-learn'
      rather than 'sklearn' for pip commands.

      Here is how to fix this error in the main use cases:
      - use 'pip install scikit-learn' rather than 'pip install sklearn'
      - replace 'sklearn' by 'scikit-learn' in your pip requirements files
        (requirements.txt, setup.py, setup.cfg, Pipfile, etc ...)
      - if the 'sklearn' package is used by one of your dependencies,
        it would be great if you take some time to track which package uses
        'sklearn' instead of 'scikit-learn' and report it to their issue tracker
      - as a last resort, set the environment variable
        SKLEARN_ALLOW_DEPRECATED_SKLEARN_PACKAGE_INSTALL=True to avoid this error

      More information is available at
      https://github.com/scikit-learn/sklearn-pypi-package
      [end of output]

解决方式，设置环境变量：export SKLEARN_ALLOW_DEPRECATED_SKLEARN_PACKAGE_INSTALL=True

===== gsnet报错 提示python版本3.6之类的和当前python版本不兼容 =====
Traceback (most recent call last):
  File "/home/ctbots/code/mirror-anygrasp_sdk/grasp_detection/demo.py", line 8, in <module>
    from gsnet import AnyGrasp
ImportError: /home/ctbots/code/mirror-anygrasp_sdk/grasp_detection/gsnet.so: undefined symbol: _PyGC_generation0

或者 干脆不显示详细内容，直接显示 python3.6 之类的报错。

解决方式：注意，必须复制2个so到 工作目录，而且版本必须和当前的python对应。 

分别是 gsnet.so 和 lib_cxx.so

分别在 grasp_detection/gsnet_versions/ 目录 和 license_registration/lib_cxx_versions/ 目录；

同时记得把 license 文件夹拷贝到工作目录。

===== 提示numpy版本错误 =====

 pip install "opencv-python<4.11" "numpy==1.23.4" "scipy<1.15.3"

===== pointnet2 提示导入错误 =====

ModuleNotFoundError: No module named 'pointnet2'

解决方案【一定要在当前的conda内，而且是gcc9】：
cd pointnet2
python setup.py install

===== 找不到 libcrypto.so.1.1 =====
从ubuntu的deb包中搜索 libssl1.1_1.1.1f-1ubuntu2.23_amd64.deb， 解压压缩包，把 压缩包里 /usr/lib/x86_64-linux-gnu 的 libcrypto.so.1.1 复制到项目根目录

官方的 https://launchpad.net 如果不科学上网可能无法访问， 可以使用镜像加速 https://748996.xyz/launchpad.net/ubuntu

然后  export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:项目的绝对路径