search

SHI-Labs / OneFormer

OneFormer: One Transformer to Rule Universal Image Segmentation, arxiv 2022 / CVPR 2023
https://praeclarumjj3.github.io/oneformer
MIT License
1.41k stars 128 forks source link

Facing issue with make.sh #59

Closed Mayank187 closed 1 year ago

Mayank187 commented 1 year ago

While running the make.sh in oneformer/modeling/pixel_decoder/ops

sh make.sh

I am getting following error

RuntimeError: 
The detected CUDA version (10.1) mismatches the version that was used to compile
PyTorch (11.3). Please make sure to use the same CUDA versions.

How to solve it?

praeclarumjj3 commented 1 year ago

Hi @Mayank187, thanks for your interest in our work.

Could you check your CUDA Build version for PyTorch? Your cudatoolkit has version 10.2, but you are using the 11.3 cuda build for PyTorch. You need to use the 10.2 build of PyTorch.

Mayank187 commented 1 year ago

Hi @praeclarumjj3 , we have following configuration

Name: torch Version: 1.9.0 Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration Home-page: https://pytorch.org/ Author: PyTorch Team Author-email: packages@pytorch.org License: BSD-3 Location: /home/techvamp/anaconda3/envs/oneformer/lib/python3.8/site-packages Requires: typing-extensions Required-by: pytorch-lightning, timm, torchmetrics, torchvision, ultralytics

and NVIDIA-SMI gives this output

+-----------------------------------------------------------------------------+ | NVIDIA-SMI 515.105.01 Driver Version: 515.105.01 CUDA Version: 11.7 | |-------------------------------+----------------------+----------------------+

praeclarumjj3 commented 1 year ago

Hi @Mayank187, could you check the Cuda version for your Pytorch installation?

torch.__version__.split("+")[-1]

Also, I noticed that you are using PyTorch 1.9.0. I suggest you follow the installation instructions.

praeclarumjj3 commented 1 year ago

Closing due to inactivity, feel free to re-open if you have any more questions.