2.3 tensorflow反向传播back propagation

大部分优化器算法需要知道每布迭代的步长,这距离由学习率控制;学习率太小可能要很小才收敛,学习率太大可能会跳过最优点导致梯度消失和梯度爆炸.

1.regression models回归模型

import numpy as np
import tensorflow as tf
from tensorflow.python.framework import ops
ops.reset_default_graph()
sess = tf.Session()

# Create data
x_vals = np.random.normal(1, 0.1, 100) #均值1标准差0.1的正态分布
y_vals = np.repeat(10., 100) #10
x_data = tf.placeholder(shape=[1], dtype=tf.float32)
y_target = tf.placeholder(shape=[1], dtype=tf.float32)

# Create variable (the only parameter: A)
A = tf.Variable(tf.random_normal(shape=[1]))
# multiply model,Theoretically, A = 10
my_output = tf.multiply(x_data, A)

# L2 loss func
loss = tf.square(my_output - y_target)

# Create Optimizer优化器
my_opt = tf.train.GradientDescentOptimizer(0.02) #learning_rate=0.02
train_step = my_opt.minimize(loss)

# Initialize variables
init = tf.global_variables_initializer()
sess.run(init)

# Run Loop,选择随机x和y传入计算图,tensorflow会自动计算损失并调整A来最小化损失
for i in range(100):
    rand_index = np.random.choice(100) 
    rand_x = [x_vals[rand_index]]
    rand_y = [y_vals[rand_index]]
    sess.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y})
    if (i+1)%25==0: #每25次打印结果
        print('Step #' + str(i+1) + ' A = ' + str(sess.run(A)))
        print('Loss = ' + str(sess.run(loss, feed_dict={x_data: rand_x, y_target: rand_y})))

2. classification models分类模型

# classification model:If sigmoid(x+A) < 0.5 -> 0 else 1

# Create graph
sess = tf.Session()

# Create data
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)

# Create variable : model parameter A
A = tf.Variable(tf.random_normal(mean=10, shape=[1]))
my_output = tf.add(x_data, A)

# Now we have to add another dimension to each (batch size of 1)
my_output_expanded = tf.expand_dims(my_output, 0)
y_target_expanded = tf.expand_dims(y_target, 0)

# Initialize variables
init = tf.global_variables_initializer()
sess.run(init)

# loss func(cross entropy)
xentropy = tf.nn.sigmoid_cross_entropy_with_logits(logits=my_output_expanded, labels=y_target_expanded)

# Create Optimizer
my_opt = tf.train.GradientDescentOptimizer(0.05)
train_step = my_opt.minimize(xentropy)  

# Run loop
for i in range(1400):
    rand_index = np.random.choice(100)
    rand_x = [x_vals[rand_index]]
    rand_y = [y_vals[rand_index]]
    
    sess.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(sess.run(A)))
        print('Loss = ' + str(sess.run(xentropy, feed_dict={x_data: rand_x, y_target: rand_y})))

# Evaluate Predictions
predictions = []
for i in range(len(x_vals)):
    x_val = [x_vals[i]]
    prediction = sess.run(tf.round(tf.sigmoid(my_output)), feed_dict={x_data: x_val})
    predictions.append(prediction[0])
    
accuracy = sum(x==y for x,y in zip(predictions, y_vals))/100.
print('Ending Accuracy = ' + str(np.round(accuracy, 2)))