Ubuntu 24.04下conda/pip安装auto-sklearn失败及TPOT导入问题求助

环境说明 #

• 当前使用系统：Ubuntu 24.04
• Windows系统下也遇到相同的auto-sklearn安装问题
• TPOT已安装成功，但无法正常导入

pip安装auto-sklearn时的错误信息 #

File "<string>", line 293, in setup_package
ModuleNotFoundError: No module named 'numpy.distutils'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

conda安装auto-sklearn时的错误信息 #

conda install conda-forge::auto-sklearn
Channels:
 - defaults
 - conda-forge
Platform: linux-64
Collecting package metadata (repodata.json): done
Solving environment: / warning  libmamba Added empty dependency for problem type SOLVER_RULE_UPDATE
failed

LibMambaUnsatisfiableError: Encountered problems while solving:
  - package auto-sklearn-0.12.5-pyhd8ed1ab_0 requires pyrfr >=0.8.1,<0.9, but none of the providers can be installed

Could not solve for environment specs
The following packages are incompatible
├─ auto-sklearn is installable and it requires
│  └─ pyrfr >=0.8.1,<0.9  with the potential options
│     ├─ pyrfr [0.8.1|0.8.2] would require
│     │  └─ python >=3.6,<3.7.0a0 , which can be installed;
│     ├─ pyrfr [0.8.1|0.8.2|0.8.3] would require
│     │  └─ python >=3.7,<3.8.0a0 , which can be installed;
│     ├─ pyrfr [0.8.1|0.8.2|0.8.3] would require
│     │  └─ python >=3.8,<3.9.0a0 , which can be installed;
│     ├─ pyrfr [0.8.1|0.8.2|0.8.3] would require
│     │  └─ python >=3.9,<3.10.0a0 , which can be installed;
│     ├─ pyrfr [0.8.2|0.8.3] would require
│     │  └─ python >=3.10,<3.11.0a0 , which can be installed;
│     └─ pyrfr 0.8.3 would require
│        └─ python >=3.11,<3.12.0a0 , which can be installed;
└─ pin-1 is not installable because it requires
   └─ python 3.12.* , which conflicts with any installable versions previously reported.

解决方法 #

修复pip安装的numpy.distutils缺失问题 #

1. 安装保留numpy.distutils的旧版numpy：

pip install numpy==1.23.5

2. 重新安装auto-sklearn：

pip install auto-sklearn

解决conda安装的Python版本冲突 #

Ubuntu 24.04默认Python为3.12，但auto-sklearn 0.12.5依赖的pyrfr最高仅支持Python 3.11，需创建指定版本的conda环境：

1. 创建并激活Python 3.11环境：

conda create -n autosklearn_env python=3.11
conda activate autosklearn_env

2. 在该环境内安装auto-sklearn：

conda install -c conda-forge auto-sklearn

修复TPOT无法导入的问题 #

TPOT安装成功但无法导入多为依赖版本不兼容，建议在上述conda环境中重新安装：

conda install -c conda-forge tpot

或通过pip安装兼容版本：

pip install tpot==0.12.1

内容的提问来源于stack exchange，提问作者Tom Tom