installation.md

Installation

The project is based on the pytorch 1.9.0 with python 3.7.

We use a 3090 to train the LA and a 32G V100 to train the BRaTS19.

1. Clone the Git repo

$ git clone https://github.com/yyliu01/TraCoCo.git
$ cd TraCoCo

2. Install dependencies

1. create conda env

   $ conda env create -f tracoco.yml

2. install the torch 1.9.0

   $ conda activate tracoco
   # IF cuda version < 11.0
   $ pip install torch==1.9.0 torchvision==0.10.0 torchaudio==0.9.0
   # IF cuda version >= 11.0 (e.g., 30x or above)
   $ pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html

3. Prepare dataset

1. download the LA, Pancreas and BRaTs19 datasets from this google drive link.
2. organize the filepath as the tree struct shown below.

TraCoCo/
├── Code
│   ├── UnetBRATS
│   ├── VnetLA
│   └── VnetPancreas
├── Datasets
│   ├── BRATS19
│   ├── Left_Atrium
│   └── Pancreas
├── docs
├── k8s_launch
│   └── jobs
└── scripts