scikit-learn 1.2中导入sklearn.model_selection报错求助

问题详情 #

使用scikit-learn 1.2版本，执行以下导入代码时触发AttributeError：

from sklearn.model_selection import train_test_split

完整错误栈：

Output exceeds the size limit. Open the full output data in a text editor
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[13], line 1
----> 1 from sklearn.model_selection import train_test_split

File /opt/homebrew/anaconda3/lib/python3.9/site-packages/sklearn/model_selection/__init__.py:23
     20 from ._split import train_test_split
     21 from ._split import check_cv
---> 23 from ._validation import cross_val_score
     24 from ._validation import cross_val_predict
     25 from ._validation import cross_validate

File /opt/homebrew/anaconda3/lib/python3.9/site-packages/sklearn/model_selection/_validation.py:32
     30 from ..utils.fixes import delayed
     31 from ..utils.metaestimators import _safe_split
---> 32 from ..metrics import check_scoring
     33 from ..metrics._scorer import _check_multimetric_scoring, _MultimetricScorer
     34 from ..exceptions import FitFailedWarning

File /opt/homebrew/anaconda3/lib/python3.9/site-packages/sklearn/metrics/__init__.py:41
     37 from ._classification import multilabel_confusion_matrix
     39 from ._dist_metrics import DistanceMetric
---> 41 from . import cluster
     42 from .cluster import adjusted_mutual_info_score
     43 from .cluster import adjusted_rand_score
...
     19 )

File sklearn/metrics/_pairwise_distances_reduction/_base.pyx:1, in init sklearn.metrics._pairwise_distances_reduction._base()

AttributeError: module 'sklearn.metrics._dist_metrics' has no attribute 'DistanceMetric32'

解决步骤 #

这个错误源于scikit-learn的Cython编译模块缺失或安装不完整，按以下方法修复：

• 重新安装指定版本：先卸载现有包，再重新安装scikit-learn 1.2.0，确保安装过程无中断

  pip uninstall -y scikit-learn
  pip install scikit-learn==1.2.0
  

• Anaconda环境用户用conda安装：避免pip与conda的依赖冲突，用conda重新安装

  conda remove scikit-learn
  conda install scikit-learn=1.2.0
  

• 清理安装缓存：如果重新安装后仍报错，清理缓存再重试
  • pip缓存清理：pip cache purge
  • conda缓存清理：conda clean -a
• 验证安装有效性：安装完成后运行以下代码确认

  import sklearn
  print(sklearn.__version__)
  from sklearn.model_selection import train_test_split
  

内容的提问来源于stack exchange，提问作者saad