Skip to content

Latest commit

 

History

History
22 lines (11 loc) · 718 Bytes

README.md

File metadata and controls

22 lines (11 loc) · 718 Bytes

Facial-emotion-classifier

The Facial Emotions Classifier, is a model that accurately predicts facial expressions.

The datasets were collected from one of Kaggle's FER competitions, (https://www.kaggle.com/aspiring1/fer2013-images) as images and then manually cleaned to remove bad images that could affect the model's performance.

Three frameworks were used: PyTorch, fastai library and Tensorflow. After which PyTorch attained the highest accuracy, using se_resnext50_32x4d with a validation accuracy of 98% and Testing accuracy of 76%.

On the other hand, fastai made a Validation accuracy of 76%, using vgg19_bn.

Group Members:

Ojeifo Oziegbe

Ayodele Adebayo

'Kayode Akanni

Praise Taiwo

Sharon Ibejih