This is the go-to method! Contrary to the two last gradient descent algorithms that we saw, instead of using the full dataset or a single instance, the Mini Batch, as the name indicates ,computes the gradients on small random sets of instances cal…