PR #22548 breaks documentation building #23008
Description
Describe the bug
The changes made in #22548 appear to cause building the example docs with make html to fail (the docs build fine in the previous commit).
Steps/Code to Reproduce
- Setup a development environment for scikit-learn
- Install additional packages to build the documentation
- Navigate to
scikit-learn/docin your development environment and runmake html
Expected Results
Documentation build completes successfully.
Actual Results
Running Sphinx v4.4.0
Configuration error:
There is a programmable error in your configuration file:
Traceback (most recent call last):
File "M:\Anaconda\envs\sklearndev\lib\site-packages\sphinx\config.py", line 340, in eval_config_file
exec(code, namespace)
File "M:\Projects\scikit-learn\doc\conf.py", line 408, in <module>
from sklearn.experimental import enable_iterative_imputer # noqa
File "m:\projects\scikit-learn\sklearn\experimental\enable_iterative_imputer.py", line 15, in <module>
from ..impute._iterative import IterativeImputer
File "m:\projects\scikit-learn\sklearn\impute\__init__.py", line 5, in <module>
from ._knn import KNNImputer
File "m:\projects\scikit-learn\sklearn\impute\_knn.py", line 11, in <module>
from ..neighbors._base import _get_weights
File "m:\projects\scikit-learn\sklearn\neighbors\__init__.py", line 17, in <module>
from ._nca import NeighborhoodComponentsAnalysis
File "m:\projects\scikit-learn\sklearn\neighbors\_nca.py", line 19, in <module>
from ..decomposition import PCA
File "m:\projects\scikit-learn\sklearn\decomposition\__init__.py", line 12, in <module>
from ._sparse_pca import SparsePCA, MiniBatchSparsePCA
File "m:\projects\scikit-learn\sklearn\decomposition\_sparse_pca.py", line 9, in <module>
from ..linear_model import ridge_regression
File "m:\projects\scikit-learn\sklearn\linear_model\__init__.py", line 32, in <module>
from ._glm import PoissonRegressor, GammaRegressor, TweedieRegressor
File "m:\projects\scikit-learn\sklearn\linear_model\_glm\__init__.py", line 3, in <module>
from .glm import (
File "m:\projects\scikit-learn\sklearn\linear_model\_glm\glm.py", line 14, in <module>
from ..._loss.glm_distribution import TweedieDistribution
File "m:\projects\scikit-learn\sklearn\_loss\__init__.py", line 6, in <module>
from .loss import (
File "m:\projects\scikit-learn\sklearn\_loss\loss.py", line 21, in <module>
from ._loss import (
ImportError: cannot import name 'CyHalfTweedieLossIdentity' from 'sklearn._loss._loss' (m:\projects\scikit-learn\sklearn\_loss\_loss.cp39-win_amd64.pyd)
Versions
System:
python: 3.9.7 (default, Sep 16 2021, 16:59:28) [MSC v.1916 64 bit (AMD64)]
executable: M:\Anaconda\envs\sklearndev\python.exe
machine: Windows-10-10.0.19044-SP0
Python dependencies:
sklearn: 1.1.dev0
pip: 21.2.4
setuptools: 58.0.4
numpy: 1.21.5
scipy: 1.7.3
Cython: 0.29.25
pandas: 1.4.1
matplotlib: 3.5.1
joblib: 1.1.0
threadpoolctl: 3.1.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: mkl
prefix: mkl_rt
filepath: M:\Anaconda\envs\sklearndev\Library\bin\mkl_rt.1.dll
version: 2021.4-Product
threading_layer: intel
num_threads: 4
user_api: openmp
internal_api: openmp
prefix: vcomp
filepath: C:\Windows\System32\vcomp140.dll
version: None
num_threads: 8