LeNet/BigDL

Notebooks for LetNet/BigDL deep learning workshop

The workshop environment is available for download in a Docker container. You can run it locally. The contrainer comes with the following pre-installed systems/libraries:

• Ubuntu Linux base image
• Spark
• BigDL
• Python 2
• Anaconda and related Python packages: NumPy, Scikit, etc.
• Tensorboard

Important: docker containers are not persistent. All your changes will be lost on restart. Download any copies of the notebooks you want to preserve to your local computer.

Installation on Mac/Linux

Install Docker service

Go to https://www.docker.com/community-edition#/download, download an appropriate installation, and install Docker.

Start Docker

Run Docker application. To confirm that you have docker running run the command:

sudo docker ps

This will show running containers. You should see an empty list, but you shouldn't see any errors.

Download the Docker image

In Terminal, run a command

sudo docker pull kalininalex/lenet-bigdl

Docker will download the image.

Run the Image

In Terminal, run a command

sudo docker run -p 8888:8888 -p 6006:6006 -it kalininalex/lenet-bigdl

The docker container will start, and print, among other things, a message that looks like:

  Copy/paste this URL into your browser when you connect for the first time,
  to login with a token:
      http://localhost:8888/?token=805eaa91983ae4b25c7bf325a59cefbb113202c6b0254321

Copy and paste it into your browser and you should see a Jupyter environment.

Navigate to notebooks\Workshop-1, and click on 1. MNIST with LeNet-BigDL.ipynb. The notebook will launch and you can start executing cells.

Monitor Training with Tensorboard

The Docker container will launch Tensorboard dashboard. You can access it via a URL:

http://localhost:6006/