速寄了，cpu使用率拉满跑了30min还不到一半

我不理解代码但我大受震撼

import tensorflow as tf
from tensorflow.keras import layers, optimizers, datasets, Sequential
import os
import ssl
ssl._create_default_https_context = ssl._create_unverified_context
# 忽略ssl认证否则下载数据集报错

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
# 用于忽略级别 2 及以下的消息（级别 1 是提示，级别 2 是警告，级别 3 是错误）。
tf.random.set_seed(2345)

conv_layers = [  # 5 units: conv+max pooling
    # unit1
    # padding="same" 输入和输出的图像大小相同
    layers.Conv2D(64, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),  # 创建一个卷积层
    layers.Conv2D(64, kernel_size=[3, 3], padding="same", activation=tf.nn.relu),
    layers.MaxPool2D(pool_size=[2, 2], strides=2, padding="same"), # max pooling

    # unit2
    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"),

    # unit3
    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"),

    # unit4
    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"),

    # unit5
    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"),
]


def preprocess(x, y):
    # 预处理，x转换到0-1的范围(?)
    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.shuffle(1000).map(preprocess).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]))


def main():
    # Convolution
    # [b, 32, 32, 3] -> [b, 1, 1, 512]
    conv_net = Sequential(conv_layers)
    conv_net.build(input_shape=[None, 32, 32, 3])
    # x = tf.random.normal([4, 32, 32, 3])
    # out = conv_net(x)
    # print(out.shape)

    # Full connection
    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])

    # 优化器  learning rate=0.0001
    optimizer = optimizers.Adam(learning_rate=1e-4)

    # list合并 [a,b]+[c,d]=[a,b,c,d]
    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:
                out = conv_net(x)
                # squeeze/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.keras.losses.categorical_crossentropy(y_onehot, logits, from_logits=True)
                # avg loss
                loss = tf.reduce_mean(loss)

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

            if step % 100 == 0:
                print(epoch, step, 'loss:', float(loss))

        # test
        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), dtype=tf.int32)
            correct = tf.reduce_sum(correct)

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

        acc = total_correct / total_num
        print(epoch, 'acc:', acc)


if __name__ == '__main__':
    main()