10 行 Python 代码构建你的第一个深度学习模型

pip install tensorflow==2.0.0-rc0

$ pip install jupyter notebook

$ jupyter notebook

import tensorflow as tf
print(tf.__version__)

mnist = tf.keras.datasets.mnist # 导入数据 mnist数据集
(x_train, y_train), (x_test, y_test) = mnist.load_data() # 将数据划分为训练数据和测试数据，其中x_train,x_test 为图像灰度值，y_train,y_test 为数据标签（指定数字具体是哪个）
x_train, x_test = x_train / 255.0, x_test / 255.0 # 输入的数据是图像的灰度值，对其进行归一化处理（灰度的最大值是255除以255）

x_train.shape # 查看训练数据大小，训练数据总数为60000的28*28图像
(60000, 28, 28)
import matplotlib.pyplot as plt
plt.imshow(x_train[0]) # 查看第一条数据的图像

y_train[0] # 查看第一条数据的标签
5

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(128, activation='relu'),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
Train on 60000 samples
Epoch 1/5
60000/60000 [==============================] - 5s 76us/sample - loss: 0.2923 - acc: 0.9144
Epoch 2/5
60000/60000 [==============================] - 4s 69us/sample - loss: 0.1389 - acc: 0.9592
Epoch 3/5
60000/60000 [==============================] - 4s 69us/sample - loss: 0.1059 - acc: 0.9681
Epoch 4/5
60000/60000 [==============================] - 4s 69us/sample - loss: 0.0875 - acc: 0.9729
Epoch 5/5
60000/60000 [==============================] - 4s 71us/sample - loss: 0.0747 - acc: 0.9765
10000/10000 - 0s - loss: 0.0780 - acc: 0.9765

model.evaluate(x_test,  y_test, verbose=2)
[0.07800268666106276, 0.9765]