CSS490_Final_Project

Overview

We will have a goal of analyzing two well-known face recognition datasets (VGG Face and UTKFace) and explore if there are biases in two datasets by training two identical models (ResNet) and compare the accuracy of each model. We added Tiny ImageNet for an artificial dataset.

Execution

We recommend use coLab because of high usage of GPU

1. Clone the repository

2. Open main_colab.ipynb

3. Uncommnet lines from Import statements

from google.colab import drive  
drive.mount('/content/drive')
!pwd
!ls drive/MyDrive/CSS490/
!mkdir ./datasets
!cp drive/MyDrive/CSS490/modified_datasets.tar.gz ./datasets
!mkdir ./util
!cp -r drive/MyDrive/CSS490/util/ ./
!tar -xf ./datasets/modified_datasets.tar.gz --directory=./datasets/
!python --version  

3. Click Runtime at the top and select Run all