Definition: A neural network is a computer system that is inspired by the biological neural network of animals.
Using Keras, we trained a convolutional neural network to recognize emotions in images. By adjusting parameters such as batch size and epochs, we were able to get an accuracy of around 49%.
Here are graphs of the training and test data comparison, along with the loss and accuracy measurements.
Original Neural Network
Epochs: 2
Batch Size: 128
Overall Model Accuracy: 38.87%
Neural Network Test 1
Epochs: 6
Batch Size: 500
Overall Model Accuracy: 48.76%
Neural Network Test 2
Epochs: 5
Batch Size: 400
Overall Model Accuracy: 49.76%
The accuracy of the model was not the highest because there was not that much data tested in the trials. If a larger batch size was tested, we suppose the accuracy would increase.