CIFAR100与VGG13实战

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

os.environ['TF_CPP_MIN_LOG_LEVEL']='2'
tf.random.set_seed(2345)

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

(x, y), (x_test, y_test) = datasets.cifar100.load_data()
y = tf.squeeze(y, axis=1)
y_test = tf.squeeze(y_test, axis=1)
print(x.shape, y.shape, x_test.shape, y_test.shape)

train_db = tf.data.Dataset.from_tensor_slices((x, y))
train_db = train_db.map(preprocess).shuffle(10000).batch(64)

test_db = tf.data.Dataset.from_tensor_slices((x_test, y_test))
test_db = test_db.map(preprocess).batch(64)

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

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

    # unit 2 (卷积核个数128)
    layers.Conv2D(128, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.Conv2D(128, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding="same"),

    # unit 3 (卷积核个数256)
    layers.Conv2D(256, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.Conv2D(256, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding="same"),

    # unit 4 (卷积核个数512)
    layers.Conv2D(512, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.Conv2D(512, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding="same"),

    # unit 5 (卷积核个数512)
    layers.Conv2D(512, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.Conv2D(512, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding="same"),
]
# [b, h, w, N]
# [b, 32, 32, 3] => [b, 1, 1, 512]
conv_net = Sequential(conv_layers)

# full connect
fc_net = Sequential([
    layers.Dense(256, activation=tf.nn.relu),
    layers.Dense(128, activation=tf.nn.relu),
    layers.Dense(100, activation=None)
])

conv_net.build(input_shape=[None, 32, 32, 3])
fc_net.build(input_shape=[None, 512])

optimizer = optimizers.Adam(lr=1e-4)
# 初始化的 w, b
variables = conv_net.trainable_variables + fc_net.trainable_variables
# print(variables)

def main():

    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)
                # flattern => [b, 512]
                out = tf.reshape(out, [-1, 512])
                # [b, 512] => [b, 100]
                logits = fc_net(out)
                y_onehot = tf.one_hot(y, depth=100)
                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))  # 对variables进行梯度更新
            if step % 100 == 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])
            out = fc_net(out)

            pred = tf.argmax(out, axis=1)
            pred = tf.cast(pred, dtype=tf.int32)

            correct = tf.cast(tf.equal(y, pred), dtype=tf.int32)
            correct = tf.reduce_sum(correct)

            total_correct += int(correct)
            total_num += x.shape[0]


        print("epoch:%s, acc:%s" % (epoch, total_correct / total_num))

if __name__=="__main__":
    main()