最新版本：http://www.mashangxue123.com/tensorflow/tf2-tutorials-quickstart-advanced.html
英文版本：https://tensorflow.google.cn/alpha/tutorials/quickstart/advanced
初学者入门教程中，使用tf.keras.Sequential模型，只是简单的堆叠模型。 本文是专家级入门，使用 Keras 模型子类 API 构建模型，会使用更底层一点的的函数接口，自定义模型、损失、评估指标和梯度下降控制等，流程清晰。
开始，请将TensorFlow库导入您的程序：
from __future__ import absolute_import, division, print_function, unicode_literals import tensorflow as tf # 安装命令 `pip install tensorflow-gpu==2.0.0-alpha0` from tensorflow.keras.layers import Dense, Flatten, Conv2D from tensorflow.keras import Model
加载并准备MNIST数据集.。
mnist = tf.keras.datasets.mnist (x_train, y_train), (x_test, y_test) = mnist.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0 # 添加一个通道维度 x_train = x_train[..., tf.newaxis] x_test = x_test[..., tf.newaxis]
使用tf.data批处理和随机打乱数据集：
train_ds = tf.data.Dataset.from_tensor_slices( (x_train, y_train)).shuffle(10000).batch(32) test_ds = tf.data.Dataset.from_tensor_slices((x_test, y_test)).batch(32)
通过使用Keras模型子类 API构建tf.keras模型：
class MyModel(Model): def __init__(self): super(MyModel, self).__init__() self.conv1 = Conv2D(32, 3, activation='relu') self.flatten = Flatten() self.d1 = Dense(128, activation='relu') self.d2 = Dense(10, activation='softmax') def call(self, x): x = self.conv1(x) x = self.flatten(x) x = self.d1(x) return self.d2(x) model = MyModel()
选择优化器和损失函数进行训练：
loss_object = tf.keras.losses.SparseCategoricalCrossentropy() optimizer = tf.keras.optimizers.Adam()
选择指标（metrics）以衡量模型的损失和准确性。这些指标累积超过周期的值，然后打印整体结果。
train_loss = tf.keras.metrics.Mean(name='train_loss') train_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='train_accuracy') test_loss = tf.keras.metrics.Mean(name='test_loss') test_accuracy = tf.keras.metrics.SparseCategoricalAccuracy(name='test_accuracy')
使用tf.GradientTape训练模型：
@tf.function def train_step(images, labels): with tf.GradientTape() as tape: predictions = model(images) loss = loss_object(labels, predictions) gradients = tape.gradient(loss, model.trainable_variables) optimizer.apply_gradients(zip(gradients, model.trainable_variables)) train_loss(loss) train_accuracy(labels, predictions)
现在测试模型：
@tf.function
def test_step(images, labels):
predictions = model(images)
t_loss = loss_object(labels, predictions)
test_loss(t_loss)
test_accuracy(labels, predictions)
EPOCHS = 5
for epoch in range(EPOCHS):
for images, labels in train_ds:
train_step(images, labels)
for test_images, test_labels in test_ds:
test_step(test_images, test_labels)
template = 'Epoch {}, Loss: {}, Accuracy: {}, Test Loss: {}, Test Accuracy: {}'
print (template.format(epoch+1,
train_loss.result(),
train_accuracy.result()*100,
test_loss.result(),
test_accuracy.result()*100))
Epoch 1, Loss: 0.13177014887332916, Accuracy: 96.06000518798828, Test Loss: 0.05814294517040253, Test Accuracy: 98.04999542236328
...
Epoch 5, Loss: 0.042211469262838364, Accuracy: 98.72000122070312, Test Loss: 0.05708516761660576, Test Accuracy: 98.3239974975586
现在，图像分类器在该数据集上的准确度达到约98％。要了解更多信息，请阅读 TensorFlow教程.。
