7. Generative Adversarial Networks

Activity 7.01: Implementing a DCGAN for the MNIST Fashion Dataset

Solution

1. Open a new Jupyter Notebook and name it Activity 7.01. Import the following library packages:

   # Import the required library functions
   import numpy as np
   import matplotlib.pyplot as plt
   from matplotlib import pyplot
   import tensorflow as tf
   from tensorflow.keras.layers import Input
   from tensorflow.keras.initializers import RandomNormal
   from tensorflow.keras.models import Model, Sequential
   from tensorflow.keras.layers \
   import Reshape, Dense, Dropout, Flatten,Activation
   from tensorflow.keras.layers import LeakyReLU,BatchNormalization
   from tensorflow.keras.layers import Conv2D, UpSampling2D,Conv2DTranspose
   from tensorflow.keras.datasets import fashion_mnist
   from tensorflow.keras.optimizers import Adam

2. Create a function that will generate real data samples from the fashion MNIST data:

   # Function to generate real data samples
   def realData(batch):
      ...