Image-Processing-using-Transfer-Learning-with-Neo4j

A image recognition (basically classification) system build uisng p5.js & ml5.js libraries. Making use of node.js and a graph based NoSQL database, Neo4j Implementing transfer learning.

How to get started

First you would require Neo4j and the below code for its driver. (Refer file public/index.html)

<!--Below is the precious driver code -->
		<!-- unpkg CDN non-minified -->
		<script src="https://unpkg.com/neo4j-driver"></script>
		

Refer to file public/test1.js to see how to connect neo4j with javascript or refer to neo4j docs

Next thing you would need is Node.js in your system and install the following modules

• Express - use code ->npm install --save express
• Socket.io- use code -> npm install --save socket.io

Now you are ready to go , run Server.js form node.js and your program will run at localhost:3000

How the code works

To know how to use Neo4j with javascript visit file Public/neo.js

You will run the server.js file through node.js which will run the file Public/index.html at localhost:3000 using socket.io.

The file Public/index.html has the Neo4j driver code , socket.io , p5.js and ml5.js. It also runs the file Public/test.js which has the code for transfer learning and neo4j