step 0. Create a conda environment and activate it.
conda create -n seglocnet python=3.10
conda activate seglocnetstep 1. Install PyTorch following official instructions.
conda install pytorch==2.1.2 torchvision==0.16.2 pytorch-cuda=11.8 -c pytorch -c nvidiastep 2. Install mmengine, mmcv, mmdet.
pip install -U openmim
mim install mmengine==0.10.5
mim install 'mmcv>=2.0.0rc4'
mim install 'mmdet>=3.0.0'step 3. Install argoverse-api following the instructions.
git clone https://github.com/argoai/argoverse-api.gitDownload hd_maps.tar.gz from their website and extract into the root directory of the argoverse-api repo.
pip install -e ${path_to_root_directory_of_the_argoverse-api_repo}step 4. Clone and install SegLocNet.
git clone https://github.com/ZhouZijie77/SegLocNet.git
cd SegLocNet
pip install -v -e .
python projects/SegLocNet/setup.py developstep 5. Install other dependencies.
pip install utm==0.7.0 rtree==1.3.0 lxml==5.3.0