RikidWai/5-shot

A Jupyter Notebook for image classification with Pytorch for CS480 ML @UWaterloo

5-shot

Kaggle Competition Project for CS480 @UWaterloo

Introduction

To classify images of household objects, from a few training examples per class. The problem is known as few-shot learning. There are only 5 training examples per class for this project.

The model predicts with ~90% accuracy.

The dataset can be found from here: https://drive.google.com/drive/folders/1fw-44z5baECSD4YLU5XLvpeNZ7LjnDjB?usp=sharing