Neural Style Transfer (TensorFlow / Jupyter / AWS GPU)

This jupyter notebook contains a TensorFlow implementation of Neural Style Transfer. It runs on an AWS GPU instance and allows editing the notebook in your browser. This setup is ideal for playing around with neural style if you do not have a GPU machine locally.

Running the notebook on an AWS GPU instance

1. Go to the Amazon EC2 Console and launch an instance.

   • Choose AMI: Select a Deep Learning Ubuntu AMI from the AWS Marketplace.

   • Choose Instance Type: Select a GPU compute p2.xlarge Instance.

   • Configure Security Group: Create a custom TCP rule to allow port 8888. This should only be allowed for your public IP.

   • Add Storage: (Optional) Increase size to 20 GiB

2. SSH to your E2 instance

3. Clone this repository

git clone https://github.com/pvt88/neural-creative-transfer-tensorflow.git

4. Set up SSL certificates and Jupyter

Firstly, we will need to generate a password for your notebook. Open a python shell and run:

>>> from notebook.auth import passwd
>>> passwd()
Enter password: 
Verify password: 
'sha1:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
>>>

The last line is the hash of your password. Copy your hased password and update the setup.sh script with it:

c.NotebookApp.keyfile = u'$SSL_CERTIFICATE_DIR/mykey.key'
c.NotebookApp.open_browser = False
c.NotebookApp.password = u'sha1:<your hashed password>'

You can now run the setup script to install tensorflow, cuda and the other required packages. It will also configure your SSL certificate and jupyter notebook.

bash setup.sh

5. Launch the jupyter notebook

... to be continued