To train your model using mixed precision with Tensor Cores or using FP32, perform the following steps using the default parameters of the V-Net model on the Hippocampus head and body dataset present on the medical segmentation decathlon website.

1. Clone the repository

git clone https://github.com/NVIDIA/DeepLearningExamples
cd DeepLearningExamples/TensorFlow/Segmentation/V-Net_tf

2. Download and preprocess the dataset

The V-Net script main.py operates on Hippocampus head and body data from the medical segmentation decathlon. Upon registration, the challenge's data is made available through the following link:

• Medical segmentation decathlon Google Drive

The script download_dataset.py is provided for data download. It is possible to select the destination folder when downloading the files by using the --data_dir flag. For example:

python download_dataset.py --data_dir ./data

Once downloaded the data using the download_dataset.py script, it can be used to run the training and benchmark scripts described below, by pointing main.py to its location using the --data_dir flag.

Note: Masks are only provided for training data.

3. Build the V-Net TensorFlow container

After Docker is correctly set up, the V-Net TensorFlow container can be built with:

docker build -t vnet_tf .

4. Start an interactive session in the NGC container to run training/inference.

Run the previously built Docker container:

$ docker run --runtime=nvidia --rm -it --shm-size=1g --ulimit memlock=-1 --ulimit stack=67108864 -v /path/to/dataset:/data vnet_tf:latest bash

Note: Ensure to mount your dataset using the -v flag to make it available for training inside the NVIDIA Docker container. Data can be downloaded as well from inside the container.

5. Start training

To run training on all training data for a default configuration (for example 1/4/8 GPUs FP32/TF-AMP), run the vnet_train.py script in the ./examples directory:

usage: vnet_train.py [-h] 
                          --data_dir DATA_DIR 
                          --model_dir MODEL_DIR 
                          --gpus {1, 8} 
                          --batch_size BATCH_SIZE 
                          --epochs EPOCHS
                          OPTIONAL [--amp]

For example:

python examples/vnet_train.py --data_dir ./data/Task04_Hippocampus --model_dir ./tmp --gpus 8 --batch_size 260 --epochs 50 --amp

To run training on 9/10 of the training data and perform evaluation on the remaining 1/10, run the vnet_train_and_evaluate.py script in the ./examples directory:

usage: vnet_train_and_evaluate.py [-h] 
                          --data_dir DATA_DIR 
                          --model_dir MODEL_DIR 
                          --gpus {1, 8} 
                          --batch_size BATCH_SIZE 
                          --epochs EPOCHS
                          OPTIONAL [--amp]

This is useful to estimate the convergence point of the training. For example:

python examples/vnet_train_and_evaluate.py --data_dir ./data/Task04_Hippocampus --model_dir ./tmp --gpus 1 --batch_size 8 --epochs 260 --amp

6. Start inference/predictions To run inference on a checkpointed model, run the vnet_predict.py script in the ./examples directory:

usage: vnet_predict.py [-h] 
                         --data_dir DATA_DIR 
                         --model_dir MODEL_DIR
                         --batch_size BATCH_SIZE 
                         OPTIONAL [--amp]

For example:

python examples/vnet_predict.py --data_dir ./data/Task04_Hippocampus --model_dir ./tmp --batch_size 4 --amp