10. Demonstration of GAN with image output

 import numpy as np

import matplotlib.pyplot as plt

from tensorflow import keras

from tensorflow.keras import layers

# Generator model

generator = keras.Sequential([

    layers.Dense(256, input_dim=100),

    layers.LeakyReLU(alpha=0.2),

    layers.Dense(512),

    layers.LeakyReLU(alpha=0.2),

    layers.Dense(784, activation='tanh'),

    layers.Reshape((28, 28, 1))

])

# Generate random images

def generate_images(num_images):

    noise = np.random.normal(0, .1, (num_images, 100))

    generated_images = generator.predict(noise)

    generated_images = 0.5 * generated_images + 0.5

    return generated_images

# Generate and display images

num_images = 25

generated_images = generate_images(num_images)

fig, axs = plt.subplots(5, 5)

cnt = 0

for i in range(5):

    for j in range(5):

        axs[i, j].imshow(generated_images[cnt, :, :, 0], cmap='magma')

        axs[i, j].axis('off')

        cnt += 1

plt.show()

------------------------------------------------------------------------

cmap values: 

In addition to 'gray', which is commonly used for grayscale images, matplotlib provides several other colormaps (cmaps) that you can use to display images. Here are some commonly used colormaps:

1. Sequential colormaps:

   • 'viridis': A perceptually uniform colormap ranging from blue to yellow.
   • 'plasma': A colormap that goes from dark purple to bright yellow.
   • 'inferno': A colormap with black, red, and yellow colors.
   • 'magma': A colormap that goes from black to white through purple and pink.
   • 'cividis': A colormap designed to be easily interpreted by individuals with colorblindness.

2. Diverging colormaps:

   • 'coolwarm': A colormap that transitions from cool colors (blues) to warm colors (reds).
   • 'bwr': A blue-white-red diverging colormap.
   • 'RdBu': A red-white-blue diverging colormap.

3. Qualitative colormaps:

   • 'tab10': A colormap with 10 distinct colors.
   • 'tab20': A colormap with 20 distinct colors.

4. Miscellaneous colormaps:

   • 'jet': A popular colormap that goes from blue to green to red to yellow.
   • 'rainbow': A colormap that spans the entire visible spectrum.
   • 'hot': A colormap ranging from black to red and white.
   • 'gray_r': The reverse grayscale colormap.
   • 'cool': A colormap that transitions from cyan to purple.

These are just a few examples, and there are many more colormaps available in matplotlib.