BUG: nan returned by np.linalg.det while it should be 0 on arm64 mac #22025
Description
Describe the issue:
np.linalg.det on a 4x4 real matrix returns nan, while it should be 0.
If the same code example is run on an intel64 linux machine, the determinant is zero for both matrices.
This happens both with conda-forge and pypi numpy, on both python 3.9 and 3.10.
Reproduce the code example:
import numpy as np
np.linalg.det(np.array([[1.0000000000000000e+000, 0.0000000000000000e+000,
0.0000000000000000e+000, 0.0000000000000000e+000],
[0.0000000000000000e+000, 1.0000000000000000e+000,
3.8307904270117927e-146, 1.4674955295685193e-291],
[0.0000000000000000e+000, 3.8307904270117927e-146,
1.4674955295685193e-291, 0.0000000000000000e+000],
[0.0000000000000000e+000, 1.4674955295685193e-291,
0.0000000000000000e+000, 0.0000000000000000e+000]]))
/opt/homebrew/Caskroom/miniforge/base/envs/numpy-test/lib/python3.10/site-packages/numpy/linalg/linalg.py:2154: RuntimeWarning: overflow encountered in det
r = _umath_linalg.det(a, signature=signature)
/opt/homebrew/Caskroom/miniforge/base/envs/numpy-test/lib/python3.10/site-packages/numpy/linalg/linalg.py:2154: RuntimeWarning: invalid value encountered in det
r = _umath_linalg.det(a, signature=signature)
nanThis one instead works fine:
np.linalg.det(np.array([[1.0000000000000000e+000, 0.0000000000000000e+000,
0.0000000000000000e+000, 0.0000000000000000e+000],
[0.0000000000000000e+000, 1.0000000000000000e+000,
3.8307904315347111e-146, 1.4674955330337903e-291],
[0.0000000000000000e+000, 3.8307904315347111e-146,
1.4674955330337899e-291, 0.0000000000000000e+000],
[0.0000000000000000e+000, 1.4674955330337903e-291,
0.0000000000000000e+000, 0.0000000000000000e+000]]))I'm also attaching a .zip file with the matrices saved as .npy files in case there are numerical precision problems:
numpy_nan.zip.
test_nan.npy gives the nan while for test_ok.npy the determinant is correct.
Error message:
No response
NumPy/Python version information:
1.23.1 3.10.0 | packaged by conda-forge | (default, Nov 20 2021, 02:27:15) [Clang 11.1.0 ]
Output of np.show_config():
>>> np.show_config()
openblas64__info:
libraries = ['openblas64_', 'openblas64_']
library_dirs = ['/usr/local/lib']
language = c
define_macros = [('HAVE_CBLAS', None), ('BLAS_SYMBOL_SUFFIX', '64_'), ('HAVE_BLAS_ILP64', None)]
runtime_library_dirs = ['/usr/local/lib']
blas_ilp64_opt_info:
libraries = ['openblas64_', 'openblas64_']
library_dirs = ['/usr/local/lib']
language = c
define_macros = [('HAVE_CBLAS', None), ('BLAS_SYMBOL_SUFFIX', '64_'), ('HAVE_BLAS_ILP64', None)]
runtime_library_dirs = ['/usr/local/lib']
openblas64__lapack_info:
libraries = ['openblas64_', 'openblas64_']
library_dirs = ['/usr/local/lib']
language = c
define_macros = [('HAVE_CBLAS', None), ('BLAS_SYMBOL_SUFFIX', '64_'), ('HAVE_BLAS_ILP64', None), ('HAVE_LAPACKE', None)]
runtime_library_dirs = ['/usr/local/lib']
lapack_ilp64_opt_info:
libraries = ['openblas64_', 'openblas64_']
library_dirs = ['/usr/local/lib']
language = c
define_macros = [('HAVE_CBLAS', None), ('BLAS_SYMBOL_SUFFIX', '64_'), ('HAVE_BLAS_ILP64', None), ('HAVE_LAPACKE', None)]
runtime_library_dirs = ['/usr/local/lib']
Supported SIMD extensions in this NumPy install:
baseline = NEON,NEON_FP16,NEON_VFPV4,ASIMD
found = ASIMDHP
not found = ASIMDDP,ASIMDFHM