Files are part of my coursework from the course DSCI 6671 Deep Learning
The files are part of 1st assignment of Deep Learning course.
Here, we compute the forward computation of a simple Neural Network - Multi-layer Perceptron.
We do this using mathematical concepts like, matrix, vectors, derivatives of forward computation functions.
The concepts covered would be:
Activation Functions: 1: Rectified Linear Unit ; 2: Softmax function
Loss: Cross-Entropy Loss
One - hot Encoding
Evaluation metrics: 1: Accuracy ; 2: Squared Mean Absolute Error(optional, would suggest you to give it a try)