Deep Neural Network from Scratch in Rust 🦀- Part 4- Loss Function and Back Propagation

After Forward Propagation we need to define a loss function to calculate how wrong our model is at this moment. For a simple binary classification problem, the loss function is given below. $$Cost:J_{(w,b)} = -\frac{1}{m}[\hat{Y}log(A^{[L]} + (1-\hat...