Skip to content

vishav/neuralstyletransfer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Artistic style transfer using convolutional neural network

The repository contains the code for artistic style transfer using convolutional neural network written using keras and Tensorflow as mentioned in the A Neural Algorithm of Artistic Style.

Important I ran this code on my laptop which had Nvidia GPU installed, which is why I used tensorflow-gpu to take advantage of my laptop GPU. If your machine doesn't have GPU, then modify line 56 in the requirements.txt from tensorflow-gpu==1.9.0 to tensorflow==1.9.0

Setup

  1. Make sure that Python (3.6), pip and conda are installed in your system.
  2. Now, create a new conda environment as: conda create --name artisticenv python=3.6 pip
  3. Now, activate this environment as follows:
    • Windows: activate artisticenv
    • Linux and macOS: source activate artisticenv
  4. Clone this repository: git clone https://github.com/vishav/neuralstyletransfer.git
  5. Change directory to neuralstyletransfer
  6. Install the project requirements from requirements.txt as follows: pip install - r requirements.txt
  7. Now, start jupyter and run the notebook: jupyter notebook
drawing drawing drawing
drawing drawing drawing
drawing drawing drawing