retinaface windows

最新推荐文章于 2025-07-28 21:22:25 发布

原创最新推荐文章于 2025-07-28 21:22:25 发布 · 1.2k 阅读

1 ·

CC 4.0 BY-SA版权

模式识别同时被 2 个专栏收录

86 篇文章

订阅专栏

编程

70 篇文章

订阅专栏

本文解决了在RCNN/Cython中执行pysetup.py build_ext --inplace时遇到的两个常见错误。第一个错误是“AttributeError: 'MSVCCompiler' object has no attribute 'compiler_so'”，解决方案涉及修改setup.py文件并编译*.pyx文件为Windows二进制文件(*.pyd)。第二个错误是“ValueError: buffer dtype mismatch, expected 'int_t' but got 'longlong'”，通过更改cpu_nms.pyx文件中的特定行可以解决。

坑1：在 rcnn/cython 中执行 py setup.py build_ext --inplace，

出现错误 “AttributeError: 'MSVCCompiler' object has no attribute 'compiler_so'”

修改方法：

https://github.com/deepinsight/insightface/issues/679#issuecomment-495870070 ( 感谢 loscheris )

Instead of using python implementation, you can try to compile *.pyx files into windows binaries (*.pyd) :

Replace setup.py in RetinaFace/rcnn/cython dir to:

# setup.py
from distutils.core import setup, Extension
from Cython.Build import cythonize
import numpy

setup(
    ext_modules=cythonize(["bbox.pyx", "anchors.pyx", "cpu_nms.pyx"]),
    include_dirs=[numpy.get_include()]
)

Use cmd to run python setup.py build_ext --inplace in RetinaFace/rcnn/cython dir
Copy the three windows binaries (*.pyd) generated in RetinaFace/rcnn/cython/rcnn/cpython to RetinaFace/rcnn/cython
Run the test code : )

坑2：ValueError: buffer dtype mismatch, expected 'int_t' but got 'long long'

解决方法：https://github.com/CharlesShang/FastMaskRCNN/issues/163#issuecomment-338623985 (感谢 mengcz13)

Fixed by changing line 25 of cpu_num.pyx from

cdef np.ndarray[np.int_t, ndim=1] order = scores.argsort()[::-1]

cdef np.ndarray[np.intp_t, ndim=1] order = scores.argsort()[::-1]

I am having some problems when compiling gpu_nms.pyx on Windows Server, and I am using cpu_nms. Still I don't know what will happen if you use gpu_nms...