Skip to content
Case study for training a neural network in a Docker container
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.pytest_cache/v/cache
src
.gitignore
Dockerfile.cpu
Dockerfile.gpu
README.md
docker-compose-cpu.yml
docker-compose-gpu.yml

README.md

Training a neural network with Docker

This the code I used in my article A Practical example of Training a Neural Network in the AWS cloud with Docker.

I prepared also ready-to-use Docker Images on Docker Hub.

Setup of a AWS instance

To train the neural network with GPU power on AWS you need to set up instance with docker-machine. The ami-0891f5dcc59fc5285 is a AMI I am sharing to the public. It has CUDO 10.1 and nvidia docker, which is needed to "activate" the GPU.

docker-machine create --driver amazonec2\
                      --amazonec2-instance-type p2.xlarge\
                      --amazonec2-ami ami-0891f5dcc59fc5285\
                      --amazonec2-vpc-id <YOUR VPC-ID>\
                      cifar10-deep-learning

Training with GPU

docker-compose -f docker-compose-gpu.yml up -d

Training with CPU

docker-compose -f docker-compose-cpu.yml up -d

Please check the article for more A Practical example of Training a Neural Network in the AWS cloud with Dockerfor more details.

You can’t perform that action at this time.