- Linux with Python ≥ 3.10.
- PyTorch ≥ 2.0 and torchvision that matches the PyTorch installation.
conda create -n segearthr1 python=3.10
conda activate segearthr1
git clone https://github.com/earth-insights/SegEarth-R1.git
cd segearthr1
pip install torch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 --index-url https://download.pytorch.org/whl/cu121
pip install -r requirements.txtFollow detectron2
After preparing the required environment, run the following command to compile CUDA kernel for MSDeformAttn:
cd segearth_r1/model/mask_decoder/Mask2Former_Simplify/modeling/pixel_decoder/ops
sh make.sh