TensorFlow 机器学实战指南示例代码之 TensorFlow 实现反向传播(二)

"""
二值分类,TensorFlow 示例
"""
import os
import tensorflow as tf
import numpy as np

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

s = tf.Session()

# 从正态分布(N(-1, 1), N(3, 1)) 生成数据,同时生成目标标签,占位符和偏差变量 A
x_vals = np.concatenate((np.random.normal(-1, 1, 50), np.random.normal(3, 1, 50)))
y_vals = np.concatenate((np.repeat(0., 50), np.repeat(1., 50)))
x_data = tf.placeholder(shape=[1], dtype=tf.float32)
y_target = tf.placeholder(shape=[1], dtype=tf.float32)
A = tf.Variable(tf.random_normal(mean=10, shape=[1]))

# 增加转换操作
my_output = tf.add(x_data, A)

# 增加一个批量维度
my_output_expanded = tf.expand_dims(my_output, 0)
y_target_expanded = tf.expand_dims(y_target, 0)

# 初始化变量A
init = tf.global_variables_initializer()
s.run(init)

# 声明损失函数,交叉熵使用预测结果来表征样本标签
xentropy = tf.nn.sigmoid_cross_entropy_with_logits(labels=y_target_expanded, logits=my_output_expanded)

# 定义一个优化器
my_opt = tf.train.GradientDescentOptimizer(0.05)
train_step = my_opt.minimize(xentropy)

# 迭代训练, 每 200 次打印出损失和变量 A 的返回值
for i in range(1400):
    rand_index = np.random.choice(100)
    rand_x = [x_vals[rand_index]]
    rand_y = [y_vals[rand_index]]
    s.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y})
    if(i+1) % 200 == 0:
        print('Step #' + str(i+1) + ' A = ' + str(s.run(A)))
        print('Loss = ' + str(s.run(xentropy, feed_dict={x_data: rand_x, y_target: rand_y})))