History

Jirka Borovec 298f9e5c2d Prune deprecated utils modules (#7503 ) * argparse_utils * model_utils * warning_utils * xla_device_utils * chlog * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>		2021-05-13 07:24:42 +00:00
..
base-conda	enable Dockers for PT 1.9 (#7363 )	2021-05-05 14:26:22 +02:00
base-cuda	enable Dockers for PT 1.9 (#7363 )	2021-05-05 14:26:22 +02:00
base-ipu	Add base IPU dockerfiles (#7252 )	2021-05-07 12:07:29 +00:00
base-xla	Fix requirements/adjust_versions.py (#7149 )	2021-05-04 01:06:28 +02:00
ipu-ci-runner	Add base IPU dockerfiles (#7252 )	2021-05-07 12:07:29 +00:00
nvidia	update ngc for 1.3 (#7414 )	2021-05-07 13:13:54 +02:00
release	remake nvidia docker (#6686 )	2021-03-29 09:39:06 +01:00
tpu-tests	Prune deprecated utils modules (#7503 )	2021-05-13 07:24:42 +00:00
README.md	Update README to 1.3 (#7489 )	2021-05-12 13:36:52 +02:00

README.md

Docker images

Builds images form attached Dockerfiles

You can build it on your own, note it takes lots of time, be prepared.

git clone <git-repository>
docker image build -t pytorch-lightning:latest -f dockers/conda/Dockerfile .

or with specific arguments

git clone <git-repository>
docker image build \
    -t pytorch-lightning:base-cuda-py3.8-pt1.8 \
    -f dockers/base-cuda/Dockerfile \
    --build-arg PYTHON_VERSION=3.8 \
    --build-arg PYTORCH_VERSION=1.8 \
    .

or nightly version from Coda

git clone <git-repository>
docker image build \
    -t pytorch-lightning:base-conda-py3.8-pt1.9 \
    -f dockers/base-conda/Dockerfile \
    --build-arg PYTHON_VERSION=3.8 \
    --build-arg PYTORCH_VERSION=1.9 \
    .

To run your docker use

docker image list
docker run --rm -it pytorch-lightning:latest bash

and if you do not need it anymore, just clean it:

docker image list
docker image rm pytorch-lightning:latest

Run docker image with GPUs

To run docker image with access to you GPUs you need to install

# Add the package repositories
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
sudo systemctl restart docker

and later run the docker image with --gpus all so for example

docker run --rm -it --gpus all pytorchlightning/pytorch_lightning:base-cuda-py3.7-torch1.6

Run Jupyter server

Inspiration comes from https://u.group/thinking/how-to-put-jupyter-notebooks-in-a-dockerfile

Build the docker image:

docker image build \
    -t pytorch-lightning:v1.3.1 \
    -f dockers/nvidia/Dockerfile \
    --build-arg LIGHTNING_VERSION=1.3.1 \
    .

start the server and map ports:

docker run --rm -it --runtime=nvidia -e NVIDIA_VISIBLE_DEVICES=all -p 8888:8888 pytorch-lightning:v1.3.1

Connect in local browser:
- copy the generated path e.g. http://hostname:8888/?token=0719fa7e1729778b0cec363541a608d5003e26d4910983c6
- replace the hostname by localhost