🍀 Challenge: Building the Discriminator

Problem statement

A partial Discriminator class is provided to you. Your task is to define the:

• Network architecture

  • The input layer should have 784 nodes.
  • The hidden layer should have 200 nodes with the LeakyReLU activation applied to the output. The gradient of the left half of the function is $0.02$. The hidden layer output should be normalized.
  • The output layer should have 1 node with the Sigmoid activation function applied to the output.

• Loss function as BCE.

• Optimiser as Adam. The learning rate should be 0.0001.