CSCI4050U_FinalProject

Final Project for CSCI4050U Machine Learning

Overview

This repository contains three notebooks to solve the classification problem of interpreting sign language gestures associated with (most of) the letters in the Latin alphabet.

Dataset

The dataset used can be found on Kaggle. There are 27455 cases in the training data and 7172 cases in the test data, each of these is a single 28x28 image. The folder containing the dataset is located in the root directory.

Classification Approaches

There are notebooks to implement the following classification approaches:

• Decision Tree
• K-Nearest Neighbours
• Convolution Net

Each of these notebooks are contained in a separate sub-directory.

Deployment

These notebooks can be modified, used, and/or deployed in the following manner:

1. Go to Google Colab
2. Open a new notebook and choose the GitHub tab
3. Copy the following url and put it into the search bar: https://github.com/MatthewSharpOTU/CSCI4050U_FinalProject/
4. Google Colab will list each of the .ipynb files in the repository. Click the desired notebook and it will open in Google Colab.
5. If desired, make a copy of the notebook (File->Save a Copy)
6. It may be necessary to update the import location for the training and testing dataframes - if so, use the following urls, respectively:

• https://raw.githubusercontent.com/MatthewSharpOTU/CSCI4050U_FinalProject/main/dataset/sign_mnist_train.csv
• https://raw.githubusercontent.com/MatthewSharpOTU/CSCI4050U_FinalProject/main/dataset/sign_mnist_test.csv

7. You're all set!