Building failure.Thanks for your help.

[JimWang151]

Below is the error message:

D:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\setuptools_init_.py:85: _DeprecatedInstaller: setuptools.installer and fetch_build_eggs are deprecated.
!!

    ********************************************************************************
    Requirements should be satisfied by a PEP 517 installer.
    If you are using pip, you can try `pip install --use-pep517`.
    ********************************************************************************

!!
dist.fetch_build_eggs(dist.setup_requires)
D:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\setuptools_distutils\dist.py:268: UserWarning: Unknown distribution option: 'tests_require'
warnings.warn(msg)
running build
running build_py
copying ctc_segmentation\ctc_segmentation.py -> build\lib.win32-cpython-39\ctc_segmentation
copying ctc_segmentation\partitioning.py -> build\lib.win32-cpython-39\ctc_segmentation
copying ctc_segmentation_init_.py -> build\lib.win32-cpython-39\ctc_segmentation
running build_ext
building 'ctc_segmentation.ctc_segmentation_dyn' extension
"d:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.40.33807\bin\HostX86\x86\cl.exe" /c /nologo /O2 /W3 /GL /DNDEBUG /MD -ID:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\numpy_core\include -ID:\ProgramData\anaconda3\envs\anitalker\include -ID:\ProgramData\anaconda3\envs\anitalker\Include "-Id:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.40.33807\include" "-Id:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.40.33807\ATLMFC\include" "-Id:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\um" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\shared" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\winrt" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\cppwinrt" /Tcctc_segmentation/ctc_segmentation_dyn.c /Fobuild\temp.win32-cpython-39\Release\ctc_segmentation/ctc_segmentation_dyn.obj
ctc_segmentation_dyn.c
D:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\numpy_core\include\numpy\npy_1_7_deprecated_api.h(14) : Warning Msg: Using deprecated NumPy API, disable it with #define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
ctc_segmentation/ctc_segmentation_dyn.c(3726): error C2039: "subarray": 不是 "_PyArray_Descr" 的成员
D:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\numpy_core\include\numpy\ndarraytypes.h(618): note: 参见“_PyArray_Descr”的声明
ctc_segmentation/ctc_segmentation_dyn.c(3726): error C2198: “_Py_INCREF”: 用于调用的参数太少
ctc_segmentation/ctc_segmentation_dyn.c(3727): error C2039: "subarray": 不是 "_PyArray_Descr" 的成员
D:\ProgramData\anaconda3\envs\anitalker\lib\site-packages\numpy_core\include\numpy\ndarraytypes.h(618): note: 参见“_PyArray_Descr”的声明
ctc_segmentation/ctc_segmentation_dyn.c(4870): warning C4244: “=”: convert from“double”to“float”，maybe miss data
error: command 'd:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.40.33807\bin\HostX86\x86\cl.exe' failed with exit code 2

[JimWang151]

My python version:3.9.19

Thanks.

[cedilla1312]

@JimWang151 Did you test with different Numpy version? Is your version 1.7.0? I was running ESPnet2 custom script after GPU drivers upgrade and I could not run it with similar error pertaining to NDARRAY after upgrading pytorch,python (to 3.9.19) ,tvision,taudio until I have installed compatible version of Numpy, maybe you could try installing Numpy 1.20.3 as I have in conda. If it won't resolve, try downgrading to your previous version of numpy with pip, then ctc_segmentation_dyn.c needs to be edited - advanced and probably not effective. Did you try pip install --use-pep517 ?

[JimWang151]

@JimWang151 Did you test with different Numpy version? Is your version 1.7.0? I was running ESPnet2 custom script after GPU drivers upgrade and I could not run it with similar error pertaining to NDARRAY after upgrading pytorch,python (to 3.9.19) ,tvision,taudio until I have installed compatible version of Numpy, maybe you could try installing Numpy 1.20.3 as I have in conda. If it won't resolve, try downgrading to your previous version of numpy with pip, then ctc_segmentation_dyn.c needs to be edited - advanced and probably not effective. Did you try pip install --use-pep517 ?

Thanks for your help.

But after i reinstalled Numpy 1.20.3，it does not work.

Can u tell me how to edit 'ctc_segmentation_dyn.c'?

[cedilla1312]

I advise you wait for developer's answer, never built this software.

My further suggestions:
Try to compile it with mingw-w64 (prolly won't do anything), try numpy 1.21;1.24; 1.19.5, the culprit could be Cython works with only old versions of Numpy (specifically its Numpy API) - Cython-Numpy mismatch, pip install --upgrade cython - another option.

Please provide additional info for devs on Numpy, Pytorch/Torch, Cython versions, compiler setup/flags or in other words, steps to repro like here.

Related links:
https://numpy.org/neps/nep-0053-c-abi-evolution.html#impact-on-cython-users
alekseik1/ifg-py@bd18222 - compatibility matrix
cython/cython#6249
cython/cython#6249 (comment) - editing ctc_segmentation_dyn.pyx - adding this line somewhere near the top of the file - probably won't fix and only trigger an error instead of warning
https://numpy.org/neps/nep-0029-deprecation_policy.html#support-table
https://github.com/numpy/numpy/blob/3113704a745e6bb8a3ce7c474ef199ee918e1e81/numpy/_core/code_generators/numpy_api.py
https://numpy.org/doc/stable/reference/c-api/deprecations.html#deprecation-mechanism-npy-no-deprecated-api
h5py/h5py#1538 (comment)

[JimWang151]

I advise you wait for developer's answer, never built this software.

My further suggestions: Try to compile it with mingw-w64 (prolly won't do anything), try numpy 1.21;1.24; 1.19.5, the culprit could be Cython works with only old versions of Numpy (specifically its Numpy API) - Cython-Numpy mismatch, pip install --upgrade cython - another option.

Please provide additional info for devs on Numpy, Pytorch/Torch, Cython versions, compiler setup/flags or in other words, steps to repro like here.

Related links: https://numpy.org/neps/nep-0053-c-abi-evolution.html#impact-on-cython-users alekseik1/ifg-py@bd18222 - compatibility matrix cython/cython#6249 cython/cython#6249 (comment) - editing ctc_segmentation_dyn.pyx - adding this line somewhere near the top of the file - probably won't fix and only trigger an error instead of warning https://numpy.org/neps/nep-0029-deprecation_policy.html#support-table https://github.com/numpy/numpy/blob/3113704a745e6bb8a3ce7c474ef199ee918e1e81/numpy/_core/code_generators/numpy_api.py https://numpy.org/doc/stable/reference/c-api/deprecations.html#deprecation-mechanism-npy-no-deprecated-api h5py/h5py#1538 (comment)

Thanks for you help.I will try your suggestion.

This issue can be closed.Thanks again.