GAN的魅力在於它能夠生成逼真的圖像，而在這篇文章中，我們將探索如何利用GAN來生成手寫數字。

1. 導入所需的庫

import numpy as np import matplotlib.pyplot as plt import tensorflow as tf from tensorflow.keras import layers

2. 加載和準備數據集
   使用MNIST數據集做訓練
   

(x_train, _), (_, _) = tf.keras.datasets.mnist.load_data()
x_train = x_train.astype('float32') / 255.0  # 正規化
x_train = np.expand_dims(x_train, axis=-1)  # 增加通道維度

3. 定義生成器模型
   生成器會從隨機噪聲中生成手寫數字。

def build_generator():
    model = tf.keras.Sequential()
    model.add(layers.Dense(256, input_dim=100, activation='relu'))
    model.add(layers.BatchNormalization())
    model.add(layers.Dense(512, activation='relu'))
    model.add(layers.BatchNormalization())
    model.add(layers.Dense(1024, activation='relu'))
    model.add(layers.BatchNormalization())
    model.add(layers.Dense(28 * 28 * 1, activation='tanh'))
    model.add(layers.Reshape((28, 28, 1)))
    return model

4. 定義判別器模型
   判別器會判斷圖像是真實的還是由生成器生成的。

def build_discriminator():
    model = tf.keras.Sequential()
    model.add(layers.Flatten(input_shape=(28, 28, 1)))
    model.add(layers.Dense(512, activation='relu'))
    model.add(layers.Dense(256, activation='relu'))
    model.add(layers.Dense(1, activation='sigmoid'))  # 二元分類
    return model

5. 構建和編譯GAN模型
   將生成器和判別器連接起來，並編譯GAN模型。

generator = build_generator()
discriminator = build_discriminator()

discriminator.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])

# GAN模型
discriminator.trainable = False  # 固定判別器
gan_input = layers.Input(shape=(100,))
generated_image = generator(gan_input)
gan_output = discriminator(generated_image)
gan = tf.keras.Model(gan_input, gan_output)
gan.compile(loss='binary_crossentropy', optimizer='adam')

6. 訓練GAN模型
   訓練過程中，我們將不斷生成圖像並訓練判別器。

def train_gan(epochs, batch_size):
    for epoch in range(epochs):
        # 隨機選擇真實圖像
        idx = np.random.randint(0, x_train.shape[0], batch_size)
        real_images = x_train[idx]

        # 生成假圖像
        noise = np.random.normal(0, 1, (batch_size, 100))
        generated_images = generator.predict(noise)

        # 訓練判別器
        discriminator_loss_real = discriminator.train_on_batch(real_images, np.ones((batch_size, 1)))
        discriminator_loss_fake = discriminator.train_on_batch(generated_images, np.zeros((batch_size, 1)))
        discriminator_loss = 0.5 * np.add(discriminator_loss_real, discriminator_loss_fake)

        # 訓練生成器
        noise = np.random.normal(0, 1, (batch_size, 100))
        generator_loss = gan.train_on_batch(noise, np.ones((batch_size, 1)))

        if epoch % 100 == 0:
            print(f"Epoch: {epoch}, Discriminator Loss: {discriminator_loss[0]}, Generator Loss: {generator_loss}")

train_gan(epochs=10000, batch_size=64)

7. 生成和顯示手寫數字
   在訓練結束後，生成一些手寫數字並顯示出來。

def generate_and_display_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  # 反正規化到0到1之間

    plt.figure(figsize=(10, 10))
    for i in range(num_images):
        plt.subplot(1, num_images, i + 1)
        plt.imshow(generated_images[i, :, :, 0], cmap='gray')
        plt.axis('off')
    plt.show()

generate_and_display_images(10)

這項技術不僅展示了人工智慧的強大能力，也讓我們能夠創造出全新的數位藝術作品。GAN的潛力無窮，它在影像生成、風格轉換等多個領域都有著廣泛的應用。