yjzhang3/leviosa

Leviosa

Introduction

Over the past three decades, the number of reported disasters has increased fourfold. Around 6.1 billion people have been affected by disasters with an estimated damage of almost 2.3 trillion dollars. An adequate level of mitigation measures and a coordinated post-disaster relief logistics management may help to reduce the loss of both human lives and economic damage.

Part I: Scan and Call

1. Frontend design in Replit
2. Backend training using Teachable Machine

Try out a real time emergency prediction by uploading your images to: Laviosa

Part II: Data Speaks

all training and parameter tuning done in MLRun

1. Preprocess the data
2. Training the model
3. Testing the model
4. Visualizations

Check out the two jupyer notebooks with detailed explanations!