Setting Up Deep Learning work station Windows

satish1v
2 min readAug 3, 2017

While ago I wrote about how to set up a deep learning workstation on AWS and I still believe that is ultimatum but it comes with the cost . So here we go for the another option .

Note : Verified the installation in Windows 10 . Need to check on Windows 8

Docker :

This is something which I started using it for a short while but turned out to be a great option to use it .

Step 1 : Install Docker community edition for windows .

Once it is installed select docker from the start menu . It will take sometime to start with so please be patient . Also to run docker you need to enable Hyper-v .This will be automatically done by docker .

Step 2 : open Powershell from start menu and type

docker pull waleedka/modern-deep-learning

This will pull all the necessary things needed for the deep learning platform . It will be totally 3 GB to download expect the delay .

Once it is installed then you should be able type

docker run -it -p 8888:8888 -p 6006:6006 -v /sharedfolder:/root/sharedfolder waleedka/modern-deep-learning bash

This will bring up a Ubuntu machine inside windows .

Step 3: Bring up Jupyter Notebook

This installation comes with all the batteries which means it has all the software ready and now it is time to use it . Type the following line in powershell

jupyter notebook — allow-root

Now go to you browser and type http://127.0.0.1:8888 this should bring the Juypter notebook .

Step 4 : Verify the Installation

Click on the New Icon and Select Python 3

Once selected it will open in the new window and Type “import keras”

Type Shift+enter to run the code and if the output is below then you are ready !!

--

--