eeva/tensorflow-haskell
1
0
Fork 0
mirror of https://github.com/tensorflow/haskell.git synced 2024-11-30 06:49:44 +01:00
Code Issues Releases Wiki Activity

Document use of nvidia docker version 2 (#208)

Browse source
This commit is contained in:
Christian Berentsen 2018-08-17 22:48:13 +02:00 committed by fkm3
parent 8a0be8ebb7
commit c978837cd3
2 changed files with 35 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
README.md
View file

@ -69,6 +69,7 @@ Check your stack version with `stack --version` in a terminal.
As an expedient we use [docker](https://www.docker.com/) for building. Once you have docker As an expedient we use [docker](https://www.docker.com/) for building. Once you have docker
working, the following commands will compile and run the tests. working, the following commands will compile and run the tests.
```
git clone --recursive https://github.com/tensorflow/haskell.git tensorflow-haskell git clone --recursive https://github.com/tensorflow/haskell.git tensorflow-haskell
cd tensorflow-haskell cd tensorflow-haskell
IMAGE_NAME=tensorflow/haskell:v0 IMAGE_NAME=tensorflow/haskell:v0
@ -76,24 +77,41 @@ working, the following commands will compile and run the tests.
# TODO: move the setup step to the docker script. # TODO: move the setup step to the docker script.
stack --docker --docker-image=$IMAGE_NAME setup stack --docker --docker-image=$IMAGE_NAME setup
stack --docker --docker-image=$IMAGE_NAME test stack --docker --docker-image=$IMAGE_NAME test
```
There is also a demo application: There is also a demo application:
```
cd tensorflow-mnist cd tensorflow-mnist
stack --docker --docker-image=$IMAGE_NAME build --exec Main stack --docker --docker-image=$IMAGE_NAME build --exec Main
```
### Docker GPU support ### Stack + Docker + GPU
If you want to use GPU you can do: If you want to use GPU you can do:
IMAGE_NAME=tensorflow/haskell:1.3.0-gpu ```
IMAGE_NAME=tensorflow/haskell:1.9.0-gpu
docker build -t $IMAGE_NAME docker/gpu docker build -t $IMAGE_NAME docker/gpu
```
We need stack to use nvidia-docker by using a 'docker' wrapper script. This will shadow the normal docker command. ### Using nvidia-docker version 2
See [Nvidia docker 2 install instructions](https://github.com/nvidia/nvidia-docker/wiki/Installation-(version-2.0))
```
stack --docker --docker-image=$IMAGE_NAME setup
stack --docker --docker-run-args "--runtime=nvidia" --docker-image=$IMAGE_NAME test
```
### Using nvidia-docker classic
Stack needs to use `nvidia-docker` instead of the normal `docker` for GPU support. We must wrap 'docker' with a script. This script will shadow the normal `docker` command.
```
ln -s `pwd`/tools/nvidia-docker-wrapper.sh <somewhere in your path>/docker ln -s `pwd`/tools/nvidia-docker-wrapper.sh <somewhere in your path>/docker
stack --docker --docker-image=$IMAGE_NAME setup stack --docker --docker-image=$IMAGE_NAME setup
stack --docker --docker-image=$IMAGE_NAME test stack --docker --docker-image=$IMAGE_NAME test
```
## Build on macOS ## Build on macOS

2
docker/gpu/Dockerfile
View file

@ -1,6 +1,6 @@
# Prepare the image with: # Prepare the image with:
# docker build -t tensorflow/haskell:1.9.0-gpu docker/gpu # docker build -t tensorflow/haskell:1.9.0-gpu docker/gpu
FROM gcr.io/tensorflow/tensorflow:1.9.0-gpu FROM tensorflow/tensorflow:1.9.0-gpu
LABEL maintainer="TensorFlow authors <tensorflow-haskell@googlegroups.com>" LABEL maintainer="TensorFlow authors <tensorflow-haskell@googlegroups.com>"
RUN apt-get update RUN apt-get update