TypeError: dist must be a Distribution instance

[chaerlo]

python3.6 build.py
Traceback (most recent call last):
File "build.py", line 30, in
extra_objects=extra_objects
File "/home/xx/enviorment/python3.6/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 228, in init
super(BuildExtension, self).init(*args, **kwargs)
File "/home/xx/enviorment/python3.6/lib/python3.6/site-packages/setuptools/init.py", line 127, in init
_Command.init(self, dist)
File "/home/xx/enviorment/python3.6/lib/python3.6/distutils/cmd.py", line 57, in init
raise TypeError("dist must be a Distribution instance")
TypeError: dist must be a Distribution instance

[chaerlo]

I am encountered with a problem when building nms.
Really thanks for giving any suggestions or help.

[JarvisKevin]

Maybe you can try PyTorch 0.3

[Swjtu-only]

i meet the same problem.

[Swjtu-only]

when i change my 0.3.1 pytorch version,i solve the problem.

[huangpan2507]

when i change my 0.3.1 pytorch version,i solve the problem.

can u show your environment ？ cuda cudnn torchvison pytorch version？thanks a lot

[huangpan2507]

when i change my 0.3.1 pytorch version,i solve the problem.

when i try to install pytorch ==0.3.1 it show that cuda=8.x version，but my cuda=10.1， it make me upset

[Swjtu-only]

@huangpan2507 Here is my environment lists:
h5py==2.8.0
imageio==2.5.0
imagesize==0.7.1
matplotlib==3.0.3
numpy==1.16.3
opencv-python==3.4.3.18
pandas==0.18.1
Pillow==6.2.0
PyYAML==5.1
scikit-image==0.15.0
scikit-learn==0.19.1
scikit-surprise==1.0.4
scipy==1.2.1
six==1.12.0
sklearn==0.0
tensorboardX==1.4
torch==0.4.0
torchvision==0.2.1
plyfile==0.6

you can write to txt,and then pip install -u *.txt.
Oh,python is 3.5.

[ghost]

How to solve this problem if we do not degrade torch version from 1.2.0 to 0.4.0?

[henbucuoshanghai]

teh same error
/envs/frcnn/lib/python3.6/distutils/cmd.py", line 57, in init
raise TypeError("dist must be a Distribution instance")
TypeError: dist must be a Distribution instance