快捷導(dǎo)航

tensorflow訓(xùn)練中出現(xiàn)nan問題的解決

更新時間：2018年02月10日 15:25:39 作者：你不來我不老

本篇文章主要介紹了tensorflow訓(xùn)練中出現(xiàn)nan問題的解決，小編覺得挺不錯的，現(xiàn)在分享給大家，也給大家做個參考。一起跟隨小編過來看看吧

深度學(xué)習(xí)中對于網(wǎng)絡(luò)的訓(xùn)練是參數(shù)更新的過程，需要注意一種情況就是輸入數(shù)據(jù)未做歸一化時，如果前向傳播結(jié)果已經(jīng)是[0,0,0,1,0,0,0,0]這種形式，而真實結(jié)果是[1,0,0,0,0,0,0,0,0]，此時由于得出的結(jié)論不懼有概率性，而是錯誤的估計值，此時反向傳播會使得權(quán)重和偏置值變的無窮大，導(dǎo)致數(shù)據(jù)溢出，也就出現(xiàn)了nan的問題。

解決辦法：

1、對輸入數(shù)據(jù)進行歸一化處理，如將輸入的圖片數(shù)據(jù)除以255將其轉(zhuǎn)化成0-1之間的數(shù)據(jù);

2、對于層數(shù)較多的情況，各層都做batch_nomorlization;

3、對設(shè)置Weights權(quán)重使用tf.truncated_normal(0, 0.01, [3,3,1,64])生成，同時值的均值為0，方差要小一些;

4、激活函數(shù)可以使用tanh;

5、減小學(xué)習(xí)率lr。

實例：

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

mnist = input_data.read_data_sets('data',one_hot = True)

def add_layer(input_data,in_size, out_size,activation_function=None):
  Weights = tf.Variable(tf.random_normal([in_size,out_size]))
  Biases = tf.Variable(tf.zeros([1, out_size])+0.1)
  Wx_plus_b = tf.add(tf.matmul(input_data, Weights), Biases)
  if activation_function==None:
    outputs = Wx_plus_b
  else:
    outputs = activation_function(Wx_plus_b)
  #return outputs#, Weights
  return {'outdata':outputs, 'w':Weights}

def get_accuracy(t_y):
#  global l1
#  accu = tf.reduce_mean(tf.cast(tf.equal(tf.argmax(l1['outdata'],1),tf.argmax(t_y,1)), dtype = tf.float32))
  global prediction
  accu = tf.reduce_mean(tf.cast(tf.equal(tf.argmax(prediction['outdata'],1),tf.argmax(t_y,1)), dtype = tf.float32))
  return accu

X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])

#l1 = add_layer(X, 784, 10, tf.nn.softmax)
#cross_entropy = tf.reduce_mean(-tf.reduce_sum(Y*tf.log(l1['outdata']), reduction_indices= [1]))
#l1 = add_layer(X, 784, 1024, tf.nn.relu)

l1 = add_layer(X, 784, 1024, None)
prediction = add_layer(l1['outdata'], 1024, 10, tf.nn.softmax)
cross_entropy = tf.reduce_mean(-tf.reduce_sum(Y*tf.log(prediction['outdata']), reduction_indices= [1]))

optimizer = tf.train.GradientDescentOptimizer(0.000001)
train = optimizer.minimize(cross_entropy)


newW = tf.Variable(tf.random_normal([1024,10]))
newOut = tf.matmul(l1['outdata'],newW)
newSoftMax = tf.nn.softmax(newOut)

init = tf.global_variables_initializer()
with tf.Session() as sess:
  sess.run(init)
  #print(sess.run(l1_Weights))
  for i in range(2):
    X_train, y_train = mnist.train.next_batch(1)
    X_train = X_train/255  #需要進行歸一化處理
    #print(sess.run(l1['w'],feed_dict={X:X_train}))
    #print(sess.run(prediction['w'],feed_dict={X:X_train, Y:y_train}))
    #print(sess.run(l1['outdata'],feed_dict={X:X_train, Y:y_train}).shape)
    print(sess.run(prediction['outdata'],feed_dict={X:X_train, Y:y_train}))
    print(sess.run(newOut, feed_dict={X:X_train}))
    print(sess.run(newSoftMax, feed_dict={X:X_train}))
    print(y_train)
    #print(sess.run(l1['outdata'], feed_dict={X:X_train}))
    sess.run(train, feed_dict={X:X_train, Y:y_train})
    if i%100 == 0:
      #print(sess.run(cross_entropy, feed_dict={X:X_train, Y:y_train}))
      accuracy = get_accuracy(mnist.test.labels)
      print(sess.run(accuracy,feed_dict={X:mnist.test.images}))
    
    #if i%100==0:
    #print(sess.run(prediction, feed_dict={X:X_train}))
    #print(sess.run(cross_entropy, feed_dict={X:X_train,Y:y_train}))

以上就是本文的全部內(nèi)容，希望對大家的學(xué)習(xí)有所幫助，也希望大家多多支持腳本之家。

您可能感興趣的文章: