food-groups-classification-model

Introduction

The goal of this project is to create a model that can predict what food group a specific food item belongs to based on its nutritional information. A $k$-nearest neighbour model is used for prediction and the model is trained on a dataset of food items from Food Standards Australia New Zealand.

File structure

There are four main folders in this repository:

Folder	Description
`archive`	Contains old files that are no longer used.
`data`	Contains the data files used in our project.
`report`	Contains the LaTeX source code for our `project_group40.pdf` report.
`src`	Contains the main source code for our project.

Instructions to run `src` code

This project uses Python 3.11.3. See Additional requirements for instructions on setting up the correct Python environment and installing dependencies.

Running the code

Run the preprocessing.ipynb notebook to preprocess the data.

This will create a data/generated folder containing preprocessed-data-classification-first.csv, preprocessed-data-classification-second.csv and preprocessed-data-classification-third.csv, which are the preprocessed data files used in our project.
Run the knn.ipynb notebook to train and evaluate the $k$-nearest neighbour model.

This will generate a few plots and saved to report/figs.

Additional requirements

Set up Python environment and install dependencies

We will be using the current latest version of Python (v3.11.3 as of 24/04/2023). To install and use the correct version of Python in a virtual environment, follow the steps below on macOS.

Install pyenv, a Python version manager. See pyenv's installation guide.
```
brew install pyenv
```
Install virtualenv, a tool to create isolated Python environments:
```
pip install virtualenv
```
Install Python 3.11.3 using pyenv:
```
pyenv install 3.11.3
```
Create a virtual environment using Python 3.11.3:
```
virtualenv -p $(pyenv which python3.11.3) venv
```
Activate the virtual environment:
```
source venv/bin/activate
```
Check that the Python version is correct:
```
python --version
```
You should see Python 3.11.3.
Now that you have the correct Python version, install the dependencies into the virtual environment:
```
pip install -r requirements.txt
```
(Optional) To deactivate the virtual environment, run:
```
deactivate
```

Credits

Victoria Lyngaae
Mason Sebek
David Sha
William Spongberg

d4vidsha/food-groups-classification-model