Home

Awesome

<h1>TornadoAi</h1> <h3>A zero coding human in the loop (HITL) collaborative data annotation plateform</h3> <h2>What is TornadoAi</h2> <p>Tornado is an open source Human-in-the-loop machine learning tool. It helps you label your dataset on the fly while training your model through a simple web user interface. It supports all data types: structured, text and image.</p> <h2>Live demo</h2> <p>In this demo https://www.youtube.com/watch?v=zBe6b_vxs_I the user collaborates with the computer to label a set of 1300 pump impeller images through 4 human in the loop machine learning iterations. The goal of labelling the images is the put in place a classifier to automatize the metal casting quality process within a manufactory.</p> <div align="center"><img src="https://raw.githubusercontent.com/slrbl/human-in-the-loop-machine-learning-tool-tornado/master/docs/demo_metal_casting.png" width="100%"/></div> <h2>Data flow</h2> <div align="center"><img src="https://raw.githubusercontent.com/slrbl/human-in-the-loop-machine-learning-tool-tornado/master/docs/data_flow.png" width="90%"/></div> <h2>Technology stack</h2> <div align="center"><img src="https://raw.githubusercontent.com/slrbl/human-in-the-loop-machine-learning-tool-tornado/master/docs/technology_stack.png" width="90%"/></div> <h2>Usage</h2> <h3>Build Tornado Docker image</h3> <code>$ docker-compose build</code> <h3>Launch Tornado image</h3> <code>$ docker-compose up</code> <br> <p>Tornado will then be available at http://localhost:3000</p> <h2>Details and how to use</h2> All information are available at: <br> https://www.linkedin.com/pulse/tornado-zero-coding-active-learning-ml-tool-walid-daboubi/