https://github.com/anushadatta/butterflymeadow

🦋 Visualizing butterflies using Google's Sketch-RNN (Magenta-Tensorflow).
https://github.com/anushadatta/butterflymeadow

ml5js sketch-rnn

Last synced: about 1 year ago
JSON representation

🦋 Visualizing butterflies using Google's Sketch-RNN (Magenta-Tensorflow).

• Host: GitHub
• URL: https://github.com/anushadatta/butterflymeadow
• Owner: anushadatta
• License: mit
• Created: 2020-05-16T14:30:44.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2020-05-18T15:25:00.000Z (about 6 years ago)
• Last Synced: 2025-02-05T16:49:04.098Z (over 1 year ago)
• Topics: ml5js, sketch-rnn
• Language: JavaScript
• Homepage: https://anushadatta.github.io/ButterflyMeadow/
• Size: 1.4 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Butterfly Meadow

This project generates butterflies using the recurrent neural network model Sketch-RNN, which has been ported to TensorFlow.js under the Magenta.js project.  Magenta is an open source research project exploring the role of machine learning as a tool in creative processes.



 See the sketching action HERE!


Sketch-RNN was taught to draw by training it on millions of doodles collected from the  Quick, Draw!  game. With over 100 different models to choose from, the butterfly object is just one possible choice. All those doodles contributing to that particular dataset can be found & viewed  here. 

Visualizing neural network generated drawings in the web browser has been made possible by using  ml5js, which is a library built on top of TensorFlow.js with no other external dependencies. 

Go ahead & try doodling other objects for yourself with SketchRNN's online demo.