Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Setting Up Environment to Run NVDS Successfully #10

Closed
TouqeerAhmad opened this issue Aug 24, 2023 · 1 comment
Closed

Setting Up Environment to Run NVDS Successfully #10

TouqeerAhmad opened this issue Aug 24, 2023 · 1 comment

Comments

@TouqeerAhmad
Copy link

TouqeerAhmad commented Aug 24, 2023

I have looked at other issues and running NVDS successfully seems to be a challenge at the moment. I am stating my steps here in hope that someone can advise me to make corrections if I am making any mistakes.

I am using a conda environment on a Ubuntu 18.04.6 LTS

Here are my steps:

conda create --name NVDS_attempt4 python=3.8
conda activate NVDS_attempt4
conda install pytorch torchvision -c pytorch

Following the author's advised strict instructions for mmcv/mmseg:

pip install -U openmim
mim install mmengine
mim install "mmcv>=2.0.0"
pip install "mmsegmentation>=1.0.0"

At this point, I tried running the code using following instruction:

CUDA_VISIBLE_DEVICES=0 python infer_NVDS_midas_bi.py --base_dir ./demo_outputs/midas_init/market_6/ --vnum market_6 --infer_w 896 --infer_h 384

It complained about timm being missing, which I installed next:

pip install timm

And now, I see the following issue.

"ImportError: /home/touqeera/mambaforge/envs/NVDS_attempt4/lib/python3.8/site-packages/mmcv/_ext.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZN3c105ErrorC2ENS_14SourceLocationESs"

I have tried several times and have not been successful so far in setting up the environment to NVDS' liking. Please advise.

@RaymondWang987
Copy link
Owner

I have looked at other issues and running NVDS successfully seems to be a challenge at the moment. I am stating my steps here in hope that someone can advise me to make corrections if I am making any mistakes.

I am using a conda environment on a Ubuntu 18.04.6 LTS

Here are my steps:

conda create --name NVDS_attempt4 python=3.8 conda activate NVDS_attempt4 conda install pytorch torchvision -c pytorch

Following the author's advised strict instructions for mmcv/mmseg:

pip install -U openmim mim install mmengine mim install "mmcv>=2.0.0" pip install "mmsegmentation>=1.0.0"

At this point, I tried running the code using following instruction:

CUDA_VISIBLE_DEVICES=0 python infer_NVDS_midas_bi.py --base_dir ./demo_outputs/midas_init/market_6/ --vnum market_6 --infer_w 896 --infer_h 384

It complained about timm being missing, which I installed next:

pip install timm

And now, I see the following issue.

"ImportError: /home/touqeera/mambaforge/envs/NVDS_attempt4/lib/python3.8/site-packages/mmcv/_ext.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZN3c105ErrorC2ENS_14SourceLocationESs"

I have tried several times and have not been successful so far in setting up the environment to NVDS' liking. Please advise.

Please refer to the answers in Issue #1.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants