快捷導(dǎo)航

tensorflow建立一個(gè)簡(jiǎn)單的神經(jīng)網(wǎng)絡(luò)的方法

更新時(shí)間：2018年02月10日 14:44:31 作者：Mr丶Caleb

本篇文章主要介紹了tensorflow建立一個(gè)簡(jiǎn)單的神經(jīng)網(wǎng)絡(luò)的方法，小編覺(jué)得挺不錯(cuò)的，現(xiàn)在分享給大家，也給大家做個(gè)參考。一起跟隨小編過(guò)來(lái)看看吧

本筆記目的是通過(guò)tensorflow實(shí)現(xiàn)一個(gè)兩層的神經(jīng)網(wǎng)絡(luò)。目的是實(shí)現(xiàn)一個(gè)二次函數(shù)的擬合。

如何添加一層網(wǎng)絡(luò)

代碼如下：

def add_layer(inputs, in_size, out_size, activation_function=None):
  # add one more layer and return the output of this layer
  Weights = tf.Variable(tf.random_normal([in_size, out_size]))
  biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
  Wx_plus_b = tf.matmul(inputs, Weights) + biases
  if activation_function is None:
    outputs = Wx_plus_b
  else:
    outputs = activation_function(Wx_plus_b)
  return outputs

注意該函數(shù)中是xW+b，而不是Wx+b。所以要注意乘法的順序。x應(yīng)該定義為[類別數(shù)量，數(shù)據(jù)數(shù)量]， W定義為[數(shù)據(jù)類別，類別數(shù)量]。

創(chuàng)建一些數(shù)據(jù)

# Make up some real data
x_data = np.linspace(-1,1,300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

numpy的linspace函數(shù)能夠產(chǎn)生等差數(shù)列。start,stop決定等差數(shù)列的起止值。endpoint參數(shù)指定包不包括終點(diǎn)值。

numpy.linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None)[source] 
Return evenly spaced numbers over a specified interval. 
Returns num evenly spaced samples, calculated over the interval [start, stop].

noise函數(shù)為添加噪聲所用，這樣二次函數(shù)的點(diǎn)不會(huì)與二次函數(shù)曲線完全重合。

numpy的newaxis可以新增一個(gè)維度而不需要重新創(chuàng)建相應(yīng)的shape在賦值，非常方便，如上面的例子中就將x_data從一維變成了二維。

添加占位符，用作輸入

# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32, [None, 1])

添加隱藏層和輸出層

# add hidden layer
l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, activation_function=None)

計(jì)算誤差，并用梯度下降使得誤差最小

# the error between prediciton and real data
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices=[1]))
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

完整代碼如下：

from __future__ import print_function
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

def add_layer(inputs, in_size, out_size, activation_function=None):
  # add one more layer and return the output of this layer
  Weights = tf.Variable(tf.random_normal([in_size, out_size]))
  biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
  Wx_plus_b = tf.matmul(inputs, Weights) + biases
  if activation_function is None:
    outputs = Wx_plus_b
  else:
    outputs = activation_function(Wx_plus_b)
  return outputs

# Make up some real data
x_data = np.linspace(-1,1,300)[:, np.newaxis]
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32, [None, 1])
# add hidden layer
l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)
# add output layer
prediction = add_layer(l1, 10, 1, activation_function=None)

# the error between prediciton and real data
loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),
           reduction_indices=[1]))
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

# important step
init = tf.initialize_all_variables()
sess = tf.Session()
sess.run(init)

# plot the real data
fig = plt.figure()
ax = fig.add_subplot(1,1,1)
ax.scatter(x_data, y_data)
plt.ion()
plt.show()

for i in range(1000):
  # training
  sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
  if i % 50 == 0:
    # to visualize the result and improvement
    try:
      ax.lines.remove(lines[0])
    except Exception:
      pass
    prediction_value = sess.run(prediction, feed_dict={xs: x_data})
    # plot the prediction
    lines = ax.plot(x_data, prediction_value, 'r-', lw=5)
    plt.pause(0.1)

運(yùn)行結(jié)果：