17.Tensorflow2.0 CIFAR与VGG实战

1. 池化和采样

• Pooling

  • Max/Avg pooling
    

• upsample

  • nearest
  • bilinear
    
    

• Relu
  

2. cifar100

• 13 layers
  

• Code

import tensorflow as tf
from tensorflow.keras import layers, optimizers, metrics, datasets, Sequential
import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
tf.random.set_seed(2345)

conv_layers = [  # 5 units of conv + max pooling
    # unit 1
    layers.Conv2D(filters=64, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.Conv2D(filters=64, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding='same'),

    # unit 2
    layers.Conv2D(filters=128, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.Conv2D(filters=128, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding='same'),

    # unit 3
    layers.Conv2D(filters=256, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.Conv2D(filters=256, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding='same'),

    # unit 4
    layers.Conv2D(filters=512, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.Conv2D(filters=512, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding='same'),

    # unit 5
    layers.Conv2D(filters=512, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.Conv2D(filters=512, kernel_size=[3, 3], padding='same', activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding='same'),
]


def pre_process(x, y):
    # [0~1]
    x = tf.cast(x, dtype=tf.float32) / 255.
    y = tf.cast(y, dtype=tf.int32)
    return x, y


# (50000, 32, 32, 3) (50000, 1) (10000, 32, 32, 3) (10000, 1)
(x, y), (x_test, y_test) = datasets.cifar100.load_data()
print(y_test)
y = tf.squeeze(y, axis=1)
train_db = tf.data.Dataset.from_tensor_slices((x, y))
train_db = train_db.shuffle(1000).map(pre_process).batch(64)

y_test = tf.squeeze(y_test, axis=1)
test_db = tf.data.Dataset.from_tensor_slices((x_test, y_test))
test_db = test_db.map(pre_process).batch(64)

sample = next(iter(train_db))
print(sample[0].shape, sample[1].shape, tf.reduce_min(sample[0]), tf.reduce_max(sample[0]))


def main():
    # 卷积层 [b, 32, 32, 3] => [b, 1, 1, 512]
    conv_net = Sequential(conv_layers)
    # 全连接层 [b, 512] => [b, 100]
    fc_net = Sequential([
        # 此处代表输出参数
        layers.Dense(256, activation=tf.nn.relu),
        layers.Dense(128, activation=tf.nn.relu),
        layers.Dense(100, activation=tf.nn.relu),
    ])

    conv_net.build(input_shape=[None, 32, 32, 3])
    fc_net.build(input_shape=[None, 512])
    optimizer = optimizers.Adam(lr=1e-4)
    # [1, 2] + [3, 4] => [1, 2, 3, 4]
    variables = conv_net.trainable_variables + fc_net.trainable_variables

    for epoch in range(50):
        for step, (x, y) in enumerate(train_db):
            with tf.GradientTape() as tape:
                # [b, 32, 32, 3] => [b, 1, 1, 512]
                out = conv_net(x)
                # flatten => [b, 512]
                out = tf.reshape(out, [-1, 512])
                # [b, 512] => [b, 100]
                logits = fc_net(out)
                # [b] => [b, 100]
                y_onehot = tf.one_hot(y, depth=100)
                # compute loss
                loss = tf.losses.categorical_crossentropy(y_onehot, logits, from_logits=True)
                loss = tf.reduce_mean(loss)

            grads = tape.gradient(loss, variables)
            optimizer.apply_gradients(zip(grads, variables))

            if step % 10 == 0:
                print(epoch, step, "loss: ", float(loss))

        total_num = 0
        total_correct = 0
        for x, y in test_db:
            out = conv_net(x)
            out = tf.reshape(out, [-1, 512])
            logits = fc_net(out)
            prob = tf.nn.softmax(logits, axis=1)
            pred = tf.argmax(prob, axis=1)
            pred = tf.cast(pred, dtype=tf.int32)

            correct = tf.cast(tf.equal(pred, y), tf.float32)
            correct = tf.reduce_sum(correct)

            total_num += x.shape[0]
            total_correct += correct

        acc = total_correct / total_num
        print("acc: ", acc)


if __name__ == '__main__':
    main()

3. 经典卷积神经网络

• ImageNet
  
• LeNet-5
  
• AlexNet
  
• VGG
  
• GoogleNet
  

4. 待续…