fc1 = fc_layer(flatten, units=512, activation=tf.nn.relu)

This code defines a fully connected layer with 512 units and ReLU activation function that takes the output of a previous layer (likely a flattened layer) as input. The layer will perform matrix multiplication between the input and its weights, add biases, and then apply the activation function to the result. The resulting output will serve as input to the next layer in the network.