Lasso error when precompute='auto' #19283
Description
Describe the bug
The docs state that the precompute parameter can be "‘auto’, bool or array-like of shape (n_features, n_features)." However, attempting to use precompute='auto' leads to a ValueError.
Steps/Code to Reproduce
from sklearn.linear_model import Lasso
model = Lasso(precompute='auto')
model.fit([[1, 2], [3, 4]], [[1], [3]])Expected Results
No error, should "just work."
Actual Results
Traceback (most recent call last):
File "/opt/project/vpf/elm/tmp.py", line 4, in <module>
model.fit([[1, 2], [3, 4]], [[1], [3]])
File "/usr/local/lib/python3.8/site-packages/sklearn/linear_model/_coordinate_descent.py", line 757, in fit
raise ValueError('precompute should be one of True, False or'
ValueError: precompute should be one of True, False or array-like. Got 'auto'
Versions
Output from sklearn.show_versions():
System:
python: 3.8.6 (default, Nov 18 2020, 13:49:49) [GCC 8.3.0]
executable: /usr/local/bin/python
machine: Linux-4.19.104-microsoft-standard-x86_64-with-glibc2.2.5
Python dependencies:
pip: 20.3.3
setuptools: 50.3.2
sklearn: 0.24.0
numpy: 1.19.5
scipy: 1.6.0
Cython: None
pandas: None
matplotlib: None
joblib: 1.0.0
threadpoolctl: 2.1.0
Built with OpenMP: True