We provide the instructions to install the dependency packages.
We test the code in the following environments, other versions may also be compatible:
- CUDA 11.1
- Python 3.7
- Pytorch 1.8.1
First, clone the repository locally.
git clone https://github.com/LANMNG/STHQ.git
Then, install Pytorch 1.8.1 using the conda environment.
conda install pytorch==1.8.1 torchvision==0.9.1 torchaudio==0.8.1 -c pytorch
Install the necessary packages and pycocotools.
pip install -r requirements.txt
pip install 'git+https://github.com/facebookresearch/fvcore'
pip install -U 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'
Finally, compile CUDA operators.
cd models/ops
python setup.py build install
cd ../..