Advanced Machine Learning Project

This repository hosts the code, report and presentation of the final project for the Advanced Machine Learning Project.

Dataset

The dataset was downloaded from the Fruits 360 Kaggle challenge. It is not included due to its large size (~700 MB when zipped). It can be found from the official challenge webpage and looking for the version 2019.06.18 of the dataset.

How to Use

The code is saved inside Jupyter notebooks and it was run on Google Colaboratory. The dataset was imported from Google Drive, you might need to change the folder path to match you own Drive structure.

## Notes About the Folder Structure

• the Discarded folders contains code that wasn't used for the report, or models that didn't make it to the final selection. They might turn out useful anyway, so they were included in the repo
• the Models directory hosts the notebooks used to create, train and make HPO on the CNNs

About Us

Toniolo Davide

• Current Studies: Student at the Master Degree in Data Science, University of Milano Bicocca
• Background: Bachelor Degree in Physics, University of Milano Bicocca

Omar Ghetti

• Current Studies: Student at the Master Degree in Computer Science, University of Milano Bicocca
• Background: Bachelor Degree in Computer Science, University of Milano Bicocca