TensorFlow-2.x-04-多分类逻辑回归的简洁实现

        本文不做多余解释，主要是衔接上一章TensorFlow-2.x-03-从0开始的多分类逻辑回归内容，使用TF-2.x来快速实现多分类逻辑回归。

1、获取/读取数据集，归一化数据

(x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()
print(x_train.shape,y_train.shape)

2、定义及初始化模型

def net():
    net=tf.keras.Sequential()
    net.add(tf.keras.layers.Flatten(input_shape=(28,28)))
    net.add(tf.keras.layers.Dense(10,activation=tf.nn.softmax)) #输出10个神经元，softmax用于分类
    return net

3、定义优化器

optimizer = tf.keras.optimizers.SGD(0.1)

4、模型训练

model.compile(optimizer=optimizer,
              loss = 'sparse_categorical_crossentropy', #交叉熵损失
              metrics=['accuracy']) # 准确率

model.fit(x_train,y_train,epochs=5,batch_size=256)

训练结果：

#5、模型验证

test_loss, test_acc = model.evaluate(x_test, y_test)
print('Test Acc:',test_acc)

验证结果：

附上所有源码：

import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np
from tensorflow import data as tfdata
from tensorflow.keras.datasets import fashion_mnist

# 1、获取和读取数据
(x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()
x_train = x_train / 255.0
x_test = x_test / 255.0
print(x_train.shape,y_train.shape)


#2、定义及初始化模型
def net():
    net=tf.keras.Sequential()
    net.add(tf.keras.layers.Flatten(input_shape=(28,28)))
    net.add(tf.keras.layers.Dense(10,activation=tf.nn.softmax)) #输出10个神经元，softmax用于分类
    return net

model=net()
model.summary()

#3、定义优化器
optimizer = tf.keras.optimizers.SGD(0.1)

# 4、模型训练
model.compile(optimizer=optimizer,
              loss = 'sparse_categorical_crossentropy', #交叉熵损失
              metrics=['accuracy']) # 准确率

model.fit(x_train,y_train,epochs=5,batch_size=256)

# 5、模型验证
test_loss, test_acc = model.evaluate(x_test, y_test)
print('Test Acc:',test_acc)