Closed LeeChunHei closed 2 years ago
what pytorch version are you using?
what pytorch version are you using?
Oh, I'm using 1.10.2 I should use 1.11 right?
But in the install.md. We are asked to create conda env with python 3.6 And it looks like conda install pytorch will install 1.10.2 in python 3.6
Yes, you should use pytorch>=1.11. You can either checkout to the branch pytorch<1.11, or you can try to install the pytorch 1.11 by: conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch Try if this installation works
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
Package pytorch conflicts for: torchvision==0.12.0 -> pytorch==1.11.0 torchaudio==0.11.0 -> pytorch==1.11.0The following specifications were found to be incompatible with your system:
Your installed version is: 2.35
The problem here is python 3.6 is installed. I think the INSTALL.md should update to correct install instruction, for example "conda create -n alphapose python=3.7 -y" ?
Oh I see. Thanks for reporting this issue!
Hi, I'm trying to install the latest AlphaPose in Linux. The distribution I use is Pop!_OS 22.04 LTS. The GPU I'm using is GTX1080. When I ran python setup.py build develop, the following error message was shown.
detector/nms/src/nms_kernel.cu:5:10: fatal error: ATen/ceil_div.h: No such file or directory 5 | #include <ATen/ceil_div.h> | ^
~~~~ compilation terminated. error: command '/usr/lib/cuda/bin/nvcc' failed with exit status 1May I ask how to fix this problem?