Tensorflow之構(gòu)建自己的圖片數(shù)據(jù)集TFrecords的方法

更新時間：2018年02月07日 10:51:12 作者：tengxing007

本篇文章主要介紹了Tensorflow之構(gòu)建自己的圖片數(shù)據(jù)集TFrecords的方法，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

學(xué)習(xí)谷歌的深度學(xué)習(xí)終于有點眉目了，給大家分享我的Tensorflow學(xué)習(xí)歷程。

tensorflow的官方中文文檔比較生澀，數(shù)據(jù)集一直采用的MNIST二進(jìn)制數(shù)據(jù)集。并沒有過多講述怎么構(gòu)建自己的圖片數(shù)據(jù)集tfrecords。

流程是：制作數(shù)據(jù)集—讀取數(shù)據(jù)集—-加入隊列

先貼完整的代碼：

#encoding=utf-8
import os
import tensorflow as tf
from PIL import Image

cwd = os.getcwd()

classes = {'test','test1','test2'}
#制作二進(jìn)制數(shù)據(jù)
def create_record():
  writer = tf.python_io.TFRecordWriter("train.tfrecords")
  for index, name in enumerate(classes):
    class_path = cwd +"/"+ name+"/"
    for img_name in os.listdir(class_path):
      img_path = class_path + img_name
      img = Image.open(img_path)
      img = img.resize((64, 64))
      img_raw = img.tobytes() #將圖片轉(zhuǎn)化為原生bytes
      print index,img_raw
      example = tf.train.Example(
        features=tf.train.Features(feature={
          "label": tf.train.Feature(int64_list=tf.train.Int64List(value=[index])),
          'img_raw': tf.train.Feature(bytes_list=tf.train.BytesList(value=[img_raw]))
        }))
      writer.write(example.SerializeToString())
  writer.close()

data = create_record()

#讀取二進(jìn)制數(shù)據(jù)
def read_and_decode(filename):
  # 創(chuàng)建文件隊列,不限讀取的數(shù)量
  filename_queue = tf.train.string_input_producer([filename])
  # create a reader from file queue
  reader = tf.TFRecordReader()
  # reader從文件隊列中讀入一個序列化的樣本
  _, serialized_example = reader.read(filename_queue)
  # get feature from serialized example
  # 解析符號化的樣本
  features = tf.parse_single_example(
    serialized_example,
    features={
      'label': tf.FixedLenFeature([], tf.int64),
      'img_raw': tf.FixedLenFeature([], tf.string)
    }
  )
  label = features['label']
  img = features['img_raw']
  img = tf.decode_raw(img, tf.uint8)
  img = tf.reshape(img, [64, 64, 3])
  img = tf.cast(img, tf.float32) * (1. / 255) - 0.5
  label = tf.cast(label, tf.int32)
  return img, label

if __name__ == '__main__':
  if 0:
    data = create_record("train.tfrecords")
  else:
    img, label = read_and_decode("train.tfrecords")
    print "tengxing",img,label
    #使用shuffle_batch可以隨機(jī)打亂輸入 next_batch挨著往下取
    # shuffle_batch才能實現(xiàn)[img,label]的同步,也即特征和label的同步,不然可能輸入的特征和label不匹配
    # 比如只有這樣使用,才能使img和label一一對應(yīng),每次提取一個image和對應(yīng)的label
    # shuffle_batch返回的值就是RandomShuffleQueue.dequeue_many()的結(jié)果
    # Shuffle_batch構(gòu)建了一個RandomShuffleQueue，并不斷地把單個的[img,label],送入隊列中
    img_batch, label_batch = tf.train.shuffle_batch([img, label],
                          batch_size=4, capacity=2000,
                          min_after_dequeue=1000)

    # 初始化所有的op
    init = tf.initialize_all_variables()

    with tf.Session() as sess:
      sess.run(init)
      # 啟動隊列
      threads = tf.train.start_queue_runners(sess=sess)
      for i in range(5):
        print img_batch.shape,label_batch
        val, l = sess.run([img_batch, label_batch])
        # l = to_categorical(l, 12)
        print(val.shape, l)

制作數(shù)據(jù)集

#制作二進(jìn)制數(shù)據(jù)
def create_record():
  cwd = os.getcwd()
  classes = {'1','2','3'}
  writer = tf.python_io.TFRecordWriter("train.tfrecords")
  for index, name in enumerate(classes):
    class_path = cwd +"/"+ name+"/"
    for img_name in os.listdir(class_path):
      img_path = class_path + img_name
      img = Image.open(img_path)
      img = img.resize((28, 28))
      img_raw = img.tobytes() #將圖片轉(zhuǎn)化為原生bytes
      #print index,img_raw
      example = tf.train.Example(
        features=tf.train.Features(
          feature={
            "label": tf.train.Feature(int64_list=tf.train.Int64List(value=[index])),
            'img_raw': tf.train.Feature(bytes_list=tf.train.BytesList(value=[img_raw]))
          }
        )
      )
      writer.write(example.SerializeToString())
  writer.close()

TFRecords文件包含了tf.train.Example 協(xié)議內(nèi)存塊(protocol buffer)(協(xié)議內(nèi)存塊包含了字段 Features)。我們可以寫一段代碼獲取你的數(shù)據(jù)，將數(shù)據(jù)填入到Example協(xié)議內(nèi)存塊(protocol buffer)，將協(xié)議內(nèi)存塊序列化為一個字符串，并且通過tf.python_io.TFRecordWriter 寫入到TFRecords文件。

讀取數(shù)據(jù)集

#讀取二進(jìn)制數(shù)據(jù)
def read_and_decode(filename):
  # 創(chuàng)建文件隊列,不限讀取的數(shù)量
  filename_queue = tf.train.string_input_producer([filename])
  # create a reader from file queue
  reader = tf.TFRecordReader()
  # reader從文件隊列中讀入一個序列化的樣本
  _, serialized_example = reader.read(filename_queue)
  # get feature from serialized example
  # 解析符號化的樣本
  features = tf.parse_single_example(
    serialized_example,
    features={
      'label': tf.FixedLenFeature([], tf.int64),
      'img_raw': tf.FixedLenFeature([], tf.string)
    }
  )
  label = features['label']
  img = features['img_raw']
  img = tf.decode_raw(img, tf.uint8)
  img = tf.reshape(img, [64, 64, 3])
  img = tf.cast(img, tf.float32) * (1. / 255) - 0.5
  label = tf.cast(label, tf.int32)
  return img, label

一個Example中包含F(xiàn)eatures，F(xiàn)eatures里包含F(xiàn)eature（這里沒s）的字典。最后，F(xiàn)eature里包含有一個 FloatList，或者ByteList，或者Int64List

加入隊列

with tf.Session() as sess:
      sess.run(init)
      # 啟動隊列
      threads = tf.train.start_queue_runners(sess=sess)
      for i in range(5):
        print img_batch.shape,label_batch
        val, l = sess.run([img_batch, label_batch])
        # l = to_categorical(l, 12)
        print(val.shape, l)

這樣就可以的到和tensorflow官方的二進(jìn)制數(shù)據(jù)集了，

注意：

啟動隊列那條code不要忘記，不然卡死
使用的時候記得使用val和l，不然會報類型錯誤：TypeError: The value of a feed cannot be a tf.Tensor object. Acceptable feed values include Python scalars, strings, lists, or numpy ndarrays.
算交叉熵時候：cross_entropy=tf.nn.sparse_softmax_cross_entropy_with_logits(logits,labels)算交叉熵
最后評估的時候用tf.nn.in_top_k(logits,labels,1)選logits最大的數(shù)的索引和label比較
cross_entropy = -tf.reduce_sum(y_*tf.log(y_conv))算交叉熵，所以label必須轉(zhuǎn)成one-hot向量

實例2：將圖片文件夾下的圖片轉(zhuǎn)存tfrecords的數(shù)據(jù)集。

############################################################################################ 
#!/usr/bin/python2.7 
# -*- coding: utf-8 -*- 
#Author : zhaoqinghui 
#Date  : 2016.5.10 
#Function: image convert to tfrecords  
############################################################################################# 
 
import tensorflow as tf 
import numpy as np 
import cv2 
import os 
import os.path 
from PIL import Image 
 
#參數(shù)設(shè)置 
############################################################################################### 
train_file = 'train.txt' #訓(xùn)練圖片 
name='train'   #生成train.tfrecords 
output_directory='./tfrecords' 
resize_height=32 #存儲圖片高度 
resize_width=32 #存儲圖片寬度 
############################################################################################### 
def _int64_feature(value): 
  return tf.train.Feature(int64_list=tf.train.Int64List(value=[value])) 
 
def _bytes_feature(value): 
  return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value])) 
 
def load_file(examples_list_file): 
  lines = np.genfromtxt(examples_list_file, delimiter=" ", dtype=[('col1', 'S120'), ('col2', 'i8')]) 
  examples = [] 
  labels = [] 
  for example, label in lines: 
    examples.append(example) 
    labels.append(label) 
  return np.asarray(examples), np.asarray(labels), len(lines) 
 
def extract_image(filename, resize_height, resize_width): 
  image = cv2.imread(filename) 
  image = cv2.resize(image, (resize_height, resize_width)) 
  b,g,r = cv2.split(image)     
  rgb_image = cv2.merge([r,g,b])    
  return rgb_image 
 
def transform2tfrecord(train_file, name, output_directory, resize_height, resize_width): 
  if not os.path.exists(output_directory) or os.path.isfile(output_directory): 
    os.makedirs(output_directory) 
  _examples, _labels, examples_num = load_file(train_file) 
  filename = output_directory + "/" + name + '.tfrecords' 
  writer = tf.python_io.TFRecordWriter(filename) 
  for i, [example, label] in enumerate(zip(_examples, _labels)): 
    print('No.%d' % (i)) 
    image = extract_image(example, resize_height, resize_width) 
    print('shape: %d, %d, %d, label: %d' % (image.shape[0], image.shape[1], image.shape[2], label)) 
    image_raw = image.tostring() 
    example = tf.train.Example(features=tf.train.Features(feature={ 
      'image_raw': _bytes_feature(image_raw), 
      'height': _int64_feature(image.shape[0]), 
      'width': _int64_feature(image.shape[1]), 
      'depth': _int64_feature(image.shape[2]), 
      'label': _int64_feature(label) 
    })) 
    writer.write(example.SerializeToString()) 
  writer.close() 
 
def disp_tfrecords(tfrecord_list_file): 
  filename_queue = tf.train.string_input_producer([tfrecord_list_file]) 
  reader = tf.TFRecordReader() 
  _, serialized_example = reader.read(filename_queue) 
  features = tf.parse_single_example( 
    serialized_example, 
 features={ 
     'image_raw': tf.FixedLenFeature([], tf.string), 
     'height': tf.FixedLenFeature([], tf.int64), 
     'width': tf.FixedLenFeature([], tf.int64), 
     'depth': tf.FixedLenFeature([], tf.int64), 
     'label': tf.FixedLenFeature([], tf.int64) 
   } 
  ) 
  image = tf.decode_raw(features['image_raw'], tf.uint8) 
  #print(repr(image)) 
  height = features['height'] 
  width = features['width'] 
  depth = features['depth'] 
  label = tf.cast(features['label'], tf.int32) 
  init_op = tf.initialize_all_variables() 
  resultImg=[] 
  resultLabel=[] 
  with tf.Session() as sess: 
    sess.run(init_op) 
    coord = tf.train.Coordinator() 
    threads = tf.train.start_queue_runners(sess=sess, coord=coord) 
    for i in range(21): 
      image_eval = image.eval() 
      resultLabel.append(label.eval()) 
      image_eval_reshape = image_eval.reshape([height.eval(), width.eval(), depth.eval()]) 
      resultImg.append(image_eval_reshape) 
      pilimg = Image.fromarray(np.asarray(image_eval_reshape)) 
      pilimg.show() 
    coord.request_stop() 
    coord.join(threads) 
    sess.close() 
  return resultImg,resultLabel 
 
def read_tfrecord(filename_queuetemp): 
  filename_queue = tf.train.string_input_producer([filename_queuetemp]) 
  reader = tf.TFRecordReader() 
  _, serialized_example = reader.read(filename_queue) 
  features = tf.parse_single_example( 
    serialized_example, 
    features={ 
     'image_raw': tf.FixedLenFeature([], tf.string), 
     'width': tf.FixedLenFeature([], tf.int64), 
     'depth': tf.FixedLenFeature([], tf.int64), 
     'label': tf.FixedLenFeature([], tf.int64) 
   } 
  ) 
  image = tf.decode_raw(features['image_raw'], tf.uint8) 
  # image 
  tf.reshape(image, [256, 256, 3]) 
  # normalize 
  image = tf.cast(image, tf.float32) * (1. /255) - 0.5 
  # label 
  label = tf.cast(features['label'], tf.int32) 
  return image, label 
 
def test(): 
  transform2tfrecord(train_file, name , output_directory, resize_height, resize_width) #轉(zhuǎn)化函數(shù)   
  img,label=disp_tfrecords(output_directory+'/'+name+'.tfrecords') #顯示函數(shù) 
  img,label=read_tfrecord(output_directory+'/'+name+'.tfrecords') #讀取函數(shù) 
  print label 
 
if __name__ == '__main__': 
  test()

這樣就可以得到自己專屬的數(shù)據(jù)集.tfrecords了，它可以直接用于tensorflow的數(shù)據(jù)集。

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: