python numpy版本报错： File "*\numpy\__init__.py", line 305, in <module> _win_os_check()

具体代码如下所示：

from numpy import *
import operator

a = random.rand(4, 4)
print(a)

具体报错内容如下所示：

Traceback (most recent call last):
  File "D:/py_prj/rtl_split/venv/Include/test.py", line 1, in <module>
    from numpy import *
  File "*\numpy\__init__.py", line 305, in <module>
    _win_os_check()
  File "*\numpy\__init__.py", line 302, in _win_os_check
    raise RuntimeError(msg.format(__file__)) from None
RuntimeError: The current Numpy installation ('*\\numpy\\__init__.py') fails to pass a sanity check due to a bug in the windows runtime. See this issue for more information: https://tinyurl.com/y3dm3h86

参考资料：https://github.com/numpy/numpy/wiki/FMod-Bug-on-Windows

Sebastian Berg edited this page on 8 Nov · 6 revisions

There is a Bug with fmod on windows, see https://tinyurl.com/y3dm3h86

To prevent unexpected runtime behavior, opencv contains a Check for this:

try:
    a = arange(13 * 13, dtype= float64).reshape(13, 13)
    a = a % 17  # calls fmod
    linalg.eig(a)
except Exception:
    raise RuntimeError("The current Numpy installation (...) fails to pass a sanity check due to a bug in the windows runtime. ...

See Pull Request https://github.com/numpy/numpy/pull/17553

Workaround

This is a windows issue and has to be fixed by Microsoft, however, a fix has not arrived for several weeks now.

• Pinning against NumPy 1.19.3 should help (it uses a newer OpenBLAS version, but this caused other problems). This can be achieved using e.g. with pip install numpy==1.19.3 or similar depending on your setup. To be clear: The only difference between NumPy 1.19.3 and 1.19.4 is the OpenBLAS version it ships.

• 32bit Python does not have these issues

• In principle you could revert the buggy windows update or deactivate the _win_os_check in NumPy (if you are lucky, your code is unaffected by the bug).

原因：是1.19.4版本有问题，需要安装1.19.3版本

更换版本后使用 import numpy

numpy.__version__ 可以查看该版本号