七、tensorflow之构建网络。

在前面的学习中，我们学会了如何通过tensorflow添加新层，具体代码如下：

import tensorflow as tf
import numpy as np
def add_layer(inputs, input_size, out_size, activation_function == None):
  Weights = tf.Variable(tf.random_normal([input_size,   
  out_size]))
  biases = tf.Variable(tf.zeros([1, out_size]))
  Wx_plus_b = tf.matmul(inputs,Weights) + biases
  if activation_function == None:
    outputs = Wx_plus_b
  else:
    outputs = activation_function(Wx_plus_b)
  return outputs

上段代码定义了增加层的方法，下面产生数据模拟这一过程。

x_data = np.linspace(-1, 1, 300)[:, np.newaxis]
noise = np.random_normal(0, 0.005, x_data)
y_data = np.square(x_data)-0.5+noise
xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32,[None, 1])

构建一个两层的神经网络。

lay1 = add_layer(xs, 1, 10, activation_function = tf.nn.relu)
prediction = add_layer(layer1, 10, 1, activation_function = None)
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction), reduce_indices=[1]))

下面是训练：

optimizer = tf.train.GradientDescentOptimizer(0.1)
train_step = optimizer.minimize(loss)
init = tf.initialize_all_variables()

接下来就是激活框架：

with tf.Session() as sess:
  sess.run(init)
  for i in range(1000):
    sess.run(train_step, feed_dict={xs: x_data, ys:
    y_data})
    if i % 50 == 0:
      print(sess.run(i, sess.run(loss, feed_dict={xs:
      x_data, ys: y_data})))

完整的过程就是这样咯。