A Deep Learning Model Based on Mamba for Automatic Segmentation in Cervical Cancer Brachytherapy

Quick Start

To get started, install the required dependencies:

pip install -r requirements.txt

For the most stable installation of Mamba-related libraries on Linux, follow these steps:

1. Create a Conda Virtual Environment:

   conda create -n mamba python=3.10

2. Activate the Environment:

   conda activate mamba

3. Install PyTorch (version 2.3.1) along with torchvision and torchaudio:

   conda install cudatoolkit=11.8 -c nvidia
   pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl

4. Offline Installation of causal-conv1d and mamba-ssm Libraries:

   • causal-conv1d (v1.4.0):

     • Download the appropriate wheel file from the causal-conv1d releases page corresponding to your system's specifications (Python version, CUDA version, etc.).

   • mamba-ssm (v2.2.2):

     • Download the suitable wheel file from the mamba releases page that matches your system's configuration.

5. Install the Downloaded Wheel Files:

   pip install causal_conv1d-1.4.0+cu118torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl
   pip install mamba_ssm-2.2.2+cu118torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl

Note: Ensure that the wheel files you download match your system's Python version, CUDA version, and PyTorch version to avoid compatibility issues.