GarbageClassification

This project is built based on garbage classification which will sort the garbage into 6 different categories that are- cardboard, glass, metal, paper, plastic and trash. We have used Garbage Classification dataset on Kaggle and classified images using pre-train (ResNet50) convolutional neural networks as pre-processing required in a ConvNet is much lower as compared to other classification algorithms. We used ImageNets weights in the ResNet50 model with test accuracy of 89.024%, about 10% less than training accuracy.

Dataset Link