I have a code that uses machine learning and neural network. I used TensorFlow 2.0 and Keras. it is a classification program that gives output as 0 or 1. I used ReLU as activation function. Sparse SoftMax Cross Entropy is used as loss function. there are 4 features and I used 1 hidden layer.
Find the mathematical interpretation (Equation) for loss function and backward propagation.
The links or names of the sources used to get the solution.
I can not find a mathematical equation for loss function and backward propagation for using ReLU and sparse Softmax Cross Entropy. I have looked in some research papers and Googled and couldn't find it. the solutions I find doesnt clearly indicate for which loss function and activation function that equation is for.
I am expecting a 1-3 line equation for loss function and back propagation for my program scenario and want the links or names for sources used to get to this conclusion by my helper.