This project was made with the motivation of getting down the fundamental basics of Neural Networks and Deep Learning. The model was trained on the FER2013 Dataset. VGG16 is a pre-trained facial recognition neural network architecture available on Keras. The project utilises this architecture.. Activation Function -> Softmax. Loss Function -> Categorical Crossentropy. Optimizer -> Adam. Also uses early stopping mechanism to avoid overfitting. Final accuracy achieved -> 0.52~.
-
Notifications
You must be signed in to change notification settings - Fork 0
Azufendei/emotion-recognition-with-vgg16
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
This project was made with the motivation of getting down the fundamental basics of Neural Networks and Deep Learning
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published