Skip to content

artificialneural.network: Interactive Artificial Neural Network, built with tensorflow.js and p5.js

Notifications You must be signed in to change notification settings

galinator9000/artificialneural.network

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Visit the tool: galinator9000.github.io/artificialneural.network

What even is this?

This is an interactive web page that runs a toy neural network right on your browser!

It visualizes the learning processes of neural networks; the building block of the AI revolution we are witnessing today.

This tool utilizes libraries like p5.js for being able to visualize the neural network and tensorflow.js in order to run the underlying mechanisms of them.

To get started;

  • Choose a sample dataset from the dataset section and click on 'Compile Dataset'

  • If you'd like to, you can fully configure your neural network from the 'Network Structure' part on right-bottom. However, the tool will set a default network configuration that depends on the chosen dataset's problem to get you started.

  • Click on "Compile network" for initializing the network with random parameters.

  • You can then start the learning process, either with the single sample it has chosen from the dataset or make it learn the whole dataset.

You can click on any of the neurons to see how strong their connections are and how they relate with each other.

Play as you like!

About

artificialneural.network: Interactive Artificial Neural Network, built with tensorflow.js and p5.js

Resources

Stars

Watchers

Forks

Packages

No packages published