Skip to content

Analysed and classified waste present in images and performed statistical analysis.

Notifications You must be signed in to change notification settings

Aaryan-Chokshi/Garbage-Detector-And-Classifier

Repository files navigation

Garbage image classification using YOLOv8n

Performed classification task of identifying 28 different waste categories from the WaRP - Waste Recycling Plant Dataset (available at Kaggle - link) using a YOLOv8n model.

Trained the model over forty iterations to ensure appropriate training. The training was monitored from WandB, which gives real-time info regarding the loss function, accuracy improvements and the result of latest training over a small validation set.

We were able to get 64% precision and box loss to be 0.76. The attached files here contain the final trained weights and data to help train and refine the model further.

About

Analysed and classified waste present in images and performed statistical analysis.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published