下面代碼的功能是先訓(xùn)練一個(gè)簡(jiǎn)單的模型，然后保存模型，同時(shí)保存到一個(gè)pb文件當(dāng)中，后續(xù)可以從pd文件里讀取權(quán)重值。

創(chuàng)新互聯(lián)堅(jiān)持“要么做到，要么別承諾”的工作理念，服務(wù)領(lǐng)域包括：網(wǎng)站制作、成都網(wǎng)站建設(shè)、企業(yè)官網(wǎng)、英文網(wǎng)站、手機(jī)端網(wǎng)站、網(wǎng)站推廣等服務(wù)，滿足客戶于互聯(lián)網(wǎng)時(shí)代的石鼓網(wǎng)站設(shè)計(jì)、移動(dòng)媒體設(shè)計(jì)的需求，幫助企業(yè)找到有效的互聯(lián)網(wǎng)解決方案。努力成為您成熟可靠的網(wǎng)絡(luò)建設(shè)合作伙伴！

import tensorflow as tf
import numpy as np
import os
import h6py
import pickle
from tensorflow.python.framework import graph_util
from tensorflow.python.platform import gfile
#設(shè)置使用指定GPU
os.environ['CUDA_VISIBLE_DEVICES'] = '1'
#下面這段代碼是在訓(xùn)練好之后將所有的權(quán)重名字和權(quán)重值羅列出來，訓(xùn)練的時(shí)候需要注釋掉
reader = tf.train.NewCheckpointReader('./model.ckpt-100')
variables = reader.get_variable_to_shape_map()
for ele in variables:
  print(ele)
  print(reader.get_tensor(ele))


x = tf.placeholder(tf.float32, shape=[None, 1])
y = 4 * x + 4

w = tf.Variable(tf.random_normal([1], -1, 1))
b = tf.Variable(tf.zeros([1]))
y_predict = w * x + b


loss = tf.reduce_mean(tf.square(y - y_predict))
optimizer = tf.train.GradientDescentOptimizer(0.5)
train = optimizer.minimize(loss)

isTrain = False#設(shè)成True去訓(xùn)練模型
train_steps = 100
checkpoint_steps = 50
checkpoint_dir = ''


saver = tf.train.Saver() # defaults to saving all variables - in this case w and b
x_data = np.reshape(np.random.rand(10).astype(np.float32), (10, 1))

with tf.Session() as sess:
  sess.run(tf.global_variables_initializer())
  if isTrain:
    for i in xrange(train_steps):
      sess.run(train, feed_dict={x: x_data})
      if (i + 1) % checkpoint_steps == 0:
        saver.save(sess, checkpoint_dir + 'model.ckpt', global_step=i+1)
  else:
    ckpt = tf.train.get_checkpoint_state(checkpoint_dir)
    if ckpt and ckpt.model_checkpoint_path:
      saver.restore(sess, ckpt.model_checkpoint_path)
    else:
      pass   
    print(sess.run(w))
    print(sess.run(b))
    graph_def = tf.get_default_graph().as_graph_def()
    #通過修改下面的函數(shù)，個(gè)人覺得理論上能夠?qū)崿F(xiàn)修改權(quán)重，但是很復(fù)雜，如果哪位有好辦法，歡迎指教
    output_graph_def = graph_util.convert_variables_to_constants(sess, graph_def, ['Variable'])
    with tf.gfile.FastGFile('./test.pb', 'wb') as f:
      f.write(output_graph_def.SerializeToString())


with tf.Session() as sess:
#對(duì)應(yīng)最后一部分的寫，這里能夠?qū)?duì)應(yīng)的變量取出來
  with gfile.FastGFile('./test.pb', 'rb') as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())
  res = tf.import_graph_def(graph_def, return_elements=['Variable:0'])
  print(sess.run(res))
  print(sess.run(graph_def))

文章題目：tensorflow保存模型和取出中間權(quán)重例子-創(chuàng)新互聯(lián)
轉(zhuǎn)載來于：http://weahome.cn/article/csjpsh.html