I build a classification model on keras (Vanilla MLP),It went quite well by using a tutorial.But what are the default weights of layers because I didn't mention any in the code.It has a feature of intializing weights, but what if we don't initialize.
It uses Xavier initialization. If you don't initialize it, by default it uses this method for initialization. If you want to know how to perform that you can use here. In the first link at the end, link of the paper is also available.
This Xavier initialization (after Glorot’s first name) is a neat trick that works well in practice. However, along came rectified linear units (ReLU), a non-linearity that is scale-invariant around 0 and does not saturate at large input values. This seemingly solved both of the problems the sigmoid function had.