[TOC]

## Basic running example

1. Clone [the repository](https://github.com/okdimok/gpu_compose_example) to the **local** (i.e. not mounted) file system.

git clone https://github.com/okdimok/gpu_compose_example.git

2. Build

docker compose -f gpu_compose_example/main-example-compose/docker-compose.yml build

3. Run

docker compose -f gpu_compose_example/main-example-compose/docker-compose.yml up

4. Connect to port **8000** on the host, as specified in the Jupyter Command.

## Simpler Example Without Compose

docker run --rm -it --ipc=host --ulimit memlock=-1 --ulimit stack=67108864  \
                --gpus='"device=0"' \
                --volume $HOME:$HOME --volume $HOME:/root \
                --network host \
                -shm-size=4gb \
                --name inference_calculator \
                --tmpfs /tmp \
                nvcr.io/nvidia/pytorch:23.10-py3

## Additional Examples

[Train Yolo V4 in TAO](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/resources/cv_samples/version/v1.4.1/files/yolo_v4/yolo_v4.ipynb)