Closed
Description
Part of #16298.
Describe the bug
When using a sparse container like csr_array for X, LinearRegression even fails to give the same coefficients for unit or no sample weight, and more generally fails the test_linear_regression_sample_weight_consitency checks. In that setting, the underlying solver is scipy.sparse.linalg.lsqr.
Steps/Code to Reproduce
from sklearn.utils.fixes import csr_array
from sklearn.datasets import make_regression
from sklearn.linear_model import LinearRegression
from sklearn.utils._testing import assert_allclose
X, y = make_regression(100, 100, random_state=42)
X = csr_array(X)
reg = LinearRegression(fit_intercept=True)
reg.fit(X, y)
coef1 = reg.coef_
reg.fit(X, y, sample_weight=np.ones_like(y))
coef2 = reg.coef_
assert_allclose(coef1, coef2, rtol=1e-7, atol=1e-9)Expected Results
The assert_allclose should pass.
Actual Results
AssertionError:
Not equal to tolerance rtol=1e-07, atol=1e-09
Mismatched elements: 100 / 100 (100%)
Max absolute difference among violations: 0.00165048
Max relative difference among violations: 0.02621317
ACTUAL: array([-2.450778e-01, 2.917985e+01, 1.678916e+00, 7.534454e+01,
1.241587e+01, 1.076716e+00, -4.975206e-01, -9.262295e-01,
-1.373931e+00, -1.624112e-01, -8.644422e-01, -5.986218e-01,...
DESIRED: array([-2.452359e-01, 2.918078e+01, 1.678681e+00, 7.534410e+01,
1.241459e+01, 1.076624e+00, -4.962305e-01, -9.257701e-01,
-1.373862e+00, -1.622824e-01, -8.652183e-01, -5.981715e-01,...The test also fails for fit_intercept=False. Note that this test and other sample weight consistency checks pass if we do not wrap X in a sparse container.
Versions
System:
python: 3.12.5 | packaged by conda-forge | (main, Aug 8 2024, 18:32:50) [Clang 16.0.6 ]
executable: /Users/abaker/miniforge3/envs/sklearn-dev/bin/python
machine: macOS-14.5-arm64-arm-64bit
Python dependencies:
sklearn: 1.6.dev0
pip: 24.2
setuptools: 73.0.1
numpy: 2.1.0
scipy: 1.14.1
Cython: 3.0.11
pandas: 2.2.2
matplotlib: 3.9.2
joblib: 1.4.2
threadpoolctl: 3.5.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: openblas
num_threads: 8
prefix: libopenblas
filepath: /Users/abaker/miniforge3/envs/sklearn-dev/lib/libopenblas.0.dylib
version: 0.3.27
threading_layer: openmp
architecture: VORTEX
user_api: openmp
internal_api: openmp
num_threads: 8
prefix: libomp
filepath: /Users/abaker/miniforge3/envs/sklearn-dev/lib/libomp.dylib
version: NoneEDIT: discovered while working on #30040 for the case of dense inputs.
Metadata
Metadata
Assignees
Type
Projects
Status
Done