3D-Vision-Transformer

Self Configuring and Adapting Vision Transformer for Segmentation of 3D Images

This repository contains the implementation of a self-configuring and adapting vision transformer designed for the segmentation of 3D images.

Getting Started

Prerequisites

• Anaconda (for managing the environment)

Installation

1. Clone the repository:

   git clone https://github.com/Anne-Andresen/3D-Vision-transformer.git
   cd Former

2. Create and activate the conda environment:

   conda env create -f environment.yml
   source activate Former

3. Create and activate the conda environment:

   pip install -e .

Training

To train the model, follow these steps:

1. Convert the dataset:

   nnFormer_convert_decathlon_task -i ../DATASET/Former/Former_data/Task01_OAR
   
   

2. Plan and preprocess the dataset:

   nnFormer_plan_and_preprocess -t 1
   

3. Train the model:

   nnFormer_train DATASET_NAME_OR_ID 3d_lowres FOLD [--npz]
   

• Example for DATASET_NAME_OR_ID:

  nnFormer_train -t 1

• Example for FOLD values: [0, 1, 2, 3, 4]

Contact

For any questions or support, you can reach me via email at:

aha.andresen@gmail.com anan@clin.au.dk

TODO

Update README

Note: This README file is a work in progress and will be updated as the project evolves.