Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

TestExp1.test_branch_cut test failure with Intel Fortran

See original GitHub issue

Running Windows Scipy 1.4.1 (compiled with MKL taken from https://www.lfd.uci.edu/~gohlke/pythonlibs/#scipy) standard test suite with Python 3.7.6 x64 results in failure for TestExp1.test_branch_cut test:

_________________________________________________________________________________________________________ TestExp1.test_branch_cut _________________________________________________________________________________________________________

self = <scipy.special.tests.test_exponential_integrals.TestExp1 object at 0x00000296E7953EC8>

    def test_branch_cut(self):
        assert np.isnan(sc.exp1(-1))
>       assert sc.exp1(complex(-1, 0)).imag == (
            -sc.exp1(complex(-1, -0.0)).imag
        )
E       AssertionError: assert -3.141592653589793 == --3.141592653589793
E        +  where -3.141592653589793 = (-1.8951178163559368-3.141592653589793j).imag
E        +    where (-1.8951178163559368-3.141592653589793j) = <ufunc 'exp1'>((-1+0j))
E        +      where <ufunc 'exp1'> = sc.exp1
E        +      and   (-1+0j) = complex(-1, 0)
E        +  and   -3.141592653589793 = (-1.8951178163559368-3.141592653589793j).imag
E        +    where (-1.8951178163559368-3.141592653589793j) = <ufunc 'exp1'>((-1-0j))
E        +      where <ufunc 'exp1'> = sc.exp1
E        +      and   (-1-0j) = complex(-1, -0.0)

self       = <scipy.special.tests.test_exponential_integrals.TestExp1 object at 0x00000296E7953EC8>

C:\Program Files\Python\3.7.6\lib\site-packages\scipy\special\tests\test_exponential_integrals.py:12: AssertionError

On the same PC package manually compiled with MKL on Ubuntu 18.04 (Python 3.6.5) successfully passes all tests.

Issue Analytics

State:
Created 4 years ago
Comments:25 (11 by maintainers)

Top GitHub Comments

1reaction

person142commented, Jan 26, 2020

So I went and checked compiler support, and it seems like we might be ok targeting 2003. Opened this issue: https://github.com/scipy/scipy/issues/11421 so that we can make an official decision. Once that’s settled we can proceed here. 🤞 for 2003 support!

0reactions

h6197627commented, Apr 11, 2021

@rgommers, I made PR https://github.com/numpy/numpy/pull/18753, please help reviewing