Tensorflow模型加載與保存、Tensorboard簡單使用

 

先上代碼：

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
# -*- coding: utf-8 -*-
"""
Created on Tue Nov 14 20:34:00 2017

@author: HJL
"""

# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================

"""A deep MNIST classifier using convolutional layers.

See extensive documentation at
https://www.tensorflow.org/get_started/mnist/pros
"""
# Disable linter warnings to maintain consistency with tutorial.
# pylint: disable=invalid-name
# pylint: disable=g-bad-import-order



import argparse
import sys
#import tempfile
import time
from tensorflow.examples.tutorials.mnist import input_data

import tensorflow as tf

FLAGS = None


def deepnn(x):
  """deepnn builds the graph for a deep net for classifying digits.

  Args:
    x: an input tensor with the dimensions (N_examples, 784), where 784 is the
    number of pixels in a standard MNIST image.

  Returns:
    A tuple (y, keep_prob). y is a tensor of shape (N_examples, 10), with values
    equal to the logits of classifying the digit into one of 10 classes (the
    digits 0-9). keep_prob is a scalar placeholder for the probability of
    dropout.
  """
  # Reshape to use within a convolutional neural net.
  # Last dimension is for "features" - there is only one here, since images are
  # grayscale -- it would be 3 for an RGB image, 4 for RGBA, etc.
  with tf.name_scope('reshape'):
    x_image = tf.reshape(x, [-1, 28, 28, 1])
    tf.summary.image('input_image', x_image)

  # First convolutional layer - maps one grayscale image to 32 feature maps.
  with tf.name_scope('conv1'):
    W_conv1 = weight_variable([5, 5, 1, 32])
    b_conv1 = bias_variable([32])
    h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)
    tf.summary.histogram('W_conv1', W_conv1)
  # Pooling layer - downsamples by 2X.
  with tf.name_scope('pool1'):
    h_pool1 = max_pool_2x2(h_conv1)

  # Second convolutional layer -- maps 32 feature maps to 64.
  with tf.name_scope('conv2'):
    W_conv2 = weight_variable([5, 5, 32, 64])
    b_conv2 = bias_variable([64])
    h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)

  # Second pooling layer.
  with tf.name_scope('pool2'):
    h_pool2 = max_pool_2x2(h_conv2)

  # Fully connected layer 1 -- after 2 round of downsampling, our 28x28 image
  # is down to 7x7x64 feature maps -- maps this to 1024 features.
  with tf.name_scope('fc1'):
    W_fc1 = weight_variable([7 * 7 * 64, 1024])
    b_fc1 = bias_variable([1024])

    h_pool2_flat = tf.reshape(h_pool2, [-1, 7*7*64])
    h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)

  # Dropout - controls the complexity of the model, prevents co-adaptation of
  # features.
  with tf.name_scope('dropout'):
    keep_prob = tf.placeholder(tf.float32)
    h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

  # Map the 1024 features to 10 classes, one for each digit
  with tf.name_scope('fc2'):
    W_fc2 = weight_variable([1024, 10])
    b_fc2 = bias_variable([10])

    y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2
  return y_conv, keep_prob


def conv2d(x, W):
  """conv2d returns a 2d convolution layer with full stride."""
  return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')


def max_pool_2x2(x):
  """max_pool_2x2 downsamples a feature map by 2X."""
  return tf.nn.max_pool(x, ksize=[1, 2, 2, 1],
                        strides=[1, 2, 2, 1], padding='SAME')


def weight_variable(shape):
  """weight_variable generates a weight variable of a given shape."""
  initial = tf.truncated_normal(shape, stddev=0.1)
  return tf.Variable(initial)


def bias_variable(shape):
  """bias_variable generates a bias variable of a given shape."""
  initial = tf.constant(0.1, shape=shape)
  return tf.Variable(initial)


def main(_):
  # Import data
  mnist = input_data.read_data_sets('./', one_hot=True)

  # Create the model
  x = tf.placeholder(tf.float32, [None, 784])

  # Define loss and optimizer
  y_ = tf.placeholder(tf.float32, [None, 10])

  # Build the graph for the deep net
  y_conv, keep_prob = deepnn(x)

  with tf.name_scope('loss'):
    cross_entropy = tf.nn.softmax_cross_entropy_with_logits(labels=y_,
                                                            logits=y_conv)
  cross_entropy = tf.reduce_mean(cross_entropy)
  

  with tf.name_scope('adam_optimizer'):
    #train_step = tf.train.AdadeltaOptimizer(1e-4).minimize(cross_entropy)
    train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)

  with tf.name_scope('accuracy'):
    correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
    correct_prediction = tf.cast(correct_prediction, tf.float32)
  accuracy = tf.reduce_mean(correct_prediction)
  
  
  graph_location = "./log/"  #tempfile.mkdtemp()
  print('Saving graph to: %s' % graph_location)
  train_writer = tf.summary.FileWriter(graph_location)
  train_writer.add_graph(tf.get_default_graph())#保存默認的圖
  
  tf.summary.scalar('cross_entropy', cross_entropy)
  tf.summary.scalar('accuracy', accuracy)
  merged = tf.summary.merge_all()
  
  with tf.Session() as sess:
    #模型保存  step1
    saver = tf.train.Saver()
    checkpoint_dir="./"
    #返回checkpoint文件中checkpoint的狀態
    ckpt = tf.train.get_checkpoint_state(checkpoint_dir)
    #print(ckpt)
    if ckpt and ckpt.model_checkpoint_path:#如果存在以前保存的模型
      print('Restore the model from checkpoint %s' % ckpt.model_checkpoint_path)
        # Restores from checkpoint
      saver.restore(sess, ckpt.model_checkpoint_path)#加載模型
      start_step = int(ckpt.model_checkpoint_path.split('/')[-1].split('-')[-1])
    else:#如果不存在之前保存的模型
      sess.run(tf.global_variables_initializer())#變量初始化
      start_step = 0
      print('start training from new state')      
      
    
    
    for i in range(start_step,start_step+20000):
      batch = mnist.train.next_batch(50)
      if i % 100 == 0:
        train_accuracy = accuracy.eval(feed_dict={
            x: batch[0], y_: batch[1], keep_prob: 1.0})
        print('step %d, training accuracy %g' % (i, train_accuracy))
        #step2   每隔一段時間 保存模型
        saver.save(sess, './log/my_test_model',global_step=i)
        
        
      summary,_=sess.run([merged, train_step],feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
      train_writer.add_summary(summary, i)
      #time.sleep(0.5)

    print('test accuracy %g' % accuracy.eval(feed_dict={
        x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))

if __name__ == '__main__':
  #main()
  
  parser = argparse.ArgumentParser()
  parser.add_argument('--data_dir', type=str,
                      default='./data/MNIST/',
                      help='Directory for storing input data')
  FLAGS, unparsed = parser.parse_known_args()
  tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
  

 上述代碼輸出如下：

 

模型的加載與保存

模型的保存涉及到兩個函數：

saver = tf.train.Saver()

和

saver.save(sess, './log/my_test_model',global_step=i)

即，先創建tf.train.Saver 對象，用於后續模型保存與加載，默認保存所有參數。saver.save用於將模型及參數保存到文件中，通過傳遞一個值給可選參數 global_step ，你可以編號checkpoint 名字。上述代碼中每隔100步，將模型保存一次。保存結果如下（默認保存最新的5個模型）：

 

 對於模型的加載，涉及如下函數：

saver = tf.train.Saver()

和

saver.restore(sess, ckpt.model_checkpoint_path)

tf.train.Saver.restore(sess, save_path)

恢復之前保存的變量

這個方法運行構造器為恢復變量所添加的操作。它需要啟動圖的Session。恢復的變量不需要經過初始化，恢復作為初始化的一種方法。

save_path 參數是之前調用save() 的返回值，或調用 latest_checkpoint() 的返回值。

參數：

• sess:  用於恢復參數的Session
• save_path:  參數之前保存的路徑

 

TensorBoard簡單使用

涉及如下幾個函數：

train_writer = tf.summary.FileWriter(graph_location)
train_writer.add_graph(tf.get_default_graph())

...

tf.summary.scalar('cross_entropy', cross_entropy)#
tf.summary.scalar('accuracy', accuracy)
tf.summary.image('input_image', x_image)
tf.summary.histogram('W_conv1', W_conv1)
merged = tf.summary.merge_all()

...
summary,_=sess.run([merged, train_step],feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
train_writer.add_summary(summary, i)

 

 

• Summary：所有需要在TensorBoard上展示的統計結果。

• tf.name_scope()：為Graph中的Tensor添加層級，TensorBoard會按照代碼指定的層級進行展示，初始狀態下只繪制最高層級的效果，點擊后可展開層級看到下一層的細節。

• tf.summary.scalar()：添加標量統計結果。

• tf.summary.histogram()：添加任意shape的Tensor，統計這個Tensor的取值分布。

• tf.summary.merge_all()：添加一個操作，代表執行所有summary操作，這樣可以避免人工執行每一個summary op。

• tf.summary.FileWrite：用於將Summary寫入磁盤，需要制定存儲路徑logdir，如果傳遞了Graph對象，則在Graph Visualization會顯示Tensor Shape Information。執行summary op后，將返回結果傳遞給add_summary()方法即可。

 

最后結果：

Scalar

（對應：

tf.summary.scalar('cross_entropy', cross_entropy)
tf.summary.scalar('accuracy', accuracy)

 

）

 

對應：

tf.summary.image('input_image', x_image)

 

 

 對應：

train_writer.add_graph(tf.get_default_graph())

 

 

 

對應：

tf.summary.histogram('W_conv1', W_conv1)