trash sorting the right wAI

Final project for the Ironhack Data Analytics Bootcamp.

1 Overview:

Since sustainability and the environment are areas of interest, we chose this as a base for our project, and to focus specifically on garbage disposal. Incorrect sorting of recycling is a major problem and people often don't know how to sort their recycling. We aimed to build a model that could be fed an image of trash and classify it correctly, giving the correct recycling bin as an output. We use pytorch/resnet18 (pre-trained model).

1.1 Team:

• Emma Baker Reynolds
• Raphael Weidemann

1.2 Topic:

Garbage Classification - Image Recognition with Convolutional Neural Networks

1.2.1 Objective:

• Learn about convolutional neural networks (CNN) and Pytorch.
• Train CNN to classify different types of garbage (recycable, non-recycable).

1.2.2 Data:

Collected by Gary Thung and Mindy Yang. The Dataset contains 2527 images of six classes of garbage deployed on plain background:

• 501 glass
• 594 paper
• 403 cardboard
• 482 plastic
• 410 metal
• 137 trash

1.3 Presentation:

trash sorting the right wAI - Presentation final project Ironhack-Bootcamp