TF從文件中讀取數據
從文件中讀取數據
在TensorFlow中進行模型訓練時,在官網給出的三種讀取方式,中最好的文件讀取方式就是將利用隊列進行文件讀取,而且步驟有兩步:
- 把樣本數據寫入TFRecords二進制文件
- 從隊列中讀取
TFRecords二進制文件,能夠更好的利用內存,更方便的移動和復制,並且不需要單獨的標記文件
下面官網給出的,對mnist文件進行操作的code,具體代碼請參考:tensorflow-master\tensorflow\examples\how_tos\reading_data\convert_to_records.py
(https://www.sogou.com/link?url=DSOYnZeCC_pKZzihDKzFgzQoUkRGi7SFyAyslJcA_SlXxobSKiNyJA..)
生成TFRecords文件
定義主函數,給訓練、驗證、測試數據集做轉換:
def main(unused_argv): # Get the data. data_sets = mnist.read_data_sets(FLAGS.directory, dtype=tf.uint8, reshape=False, validation_size=FLAGS.validation_size) # Convert to Examples and write the result to TFRecords. convert_to(data_sets.train, 'train') convert_to(data_sets.validation, 'validation') convert_to(data_sets.test, 'test')
轉換函數的作用convert_to的主要功能是,將數據填入到協議緩沖區,並化為一個字符串,然後寫入到TFRecords文件。
def convert_to(data_set, name): """Converts a dataset to tfrecords.""" images = data_set.images labels = data_set.labels num_examples = data_set.num_examples if images.shape[0] != num_examples: raise ValueError('Images size %d does not match label size %d.' % (images.shape[0], num_examples)) rows = images.shape[1] # 28 cols = images.shape[2] # 28 depth = images.shape[3] # 1. 是黑白圖像,所以是單通道 filename = os.path.join(FLAGS.directory, name + '.tfrecords') print('Writing', filename) writer = tf.python_io.TFRecordWriter(filename) for index in range(num_examples): image_raw = images[index].tostring() # 寫入協議緩存區,height,width,depth,label編碼成int64類型,image_raw 編碼成二進制 example = tf.train.Example(features=tf.train.Features(feature={ 'height': _int64_feature(rows), 'width': _int64_feature(cols), 'depth': _int64_feature(depth), 'label': _int64_feature(int(labels[index])), 'image_raw': _bytes_feature(image_raw)})) writer.write(example.SerializeToString()) # 序列化為字符串 writer.close()
編碼函數如下:
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]))
完整代碼:
import tensorflow as tf
import os
import argparse
import sys
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
#1.0 生成TFRecords 文件
from tensorflow.contrib.learn.python.learn.datasets import mnist
FLAGS = None
# 編碼函數如下:
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 convert_to(data_set, name):
"""Converts a dataset to tfrecords."""
images = data_set.images
labels = data_set.labels
num_examples = data_set.num_examples
if images.shape[0] != num_examples:
raise ValueError('Images size %d does not match label size %d.' %
(images.shape[0], num_examples))
rows = images.shape[1] # 28
cols = images.shape[2] # 28
depth = images.shape[3] # 1. 是黑白圖像,所以是單通道
filename = os.path.join(FLAGS.directory, name + '.tfrecords')
print('Writing', filename)
writer = tf.python_io.TFRecordWriter(filename)
for index in range(num_examples):
image_raw = images[index].tostring()
# 寫入協議緩存區,height,width,depth,label編碼成int64類型,image_raw 編碼成二進制
example = tf.train.Example(features=tf.train.Features(feature={
'height': _int64_feature(rows),
'width': _int64_feature(cols),
'depth': _int64_feature(depth),
'label': _int64_feature(int(labels[index])),
'image_raw': _bytes_feature(image_raw)}))
writer.write(example.SerializeToString()) # 序列化為字符串
writer.close()
def main(unused_argv):
# Get the data.
data_sets = mnist.read_data_sets(FLAGS.directory,
dtype=tf.uint8,
reshape=False,
validation_size=FLAGS.validation_size)
# Convert to Examples and write the result to TFRecords.
convert_to(data_sets.train, 'train')
convert_to(data_sets.validation, 'validation')
convert_to(data_sets.test, 'test')
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument(
'--directory',
type=str,
default='MNIST_data/',
help='Directory to download data files and write the converted result'
)
parser.add_argument(
'--validation_size',
type=int,
default=5000,
help=""" Number of examples to separate from the training data for the validation
set. """
)
FLAGS, unparsed = parser.parse_known_args()
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
運行結束後,在/tmp/data下生成3個文件,即train.tfrecords,validation.tfrecords和test.tfrecords.
從隊列中讀取
讀取TFRecords文件步驟
使用隊列讀取數TFRecords 文件 數據的步驟
- 創建張量,從二進制文件讀取一個樣本數據
- 創建張量,從二進制文件隨機讀取一個mini-batch
- 把每一批張量傳入網絡作為輸入點
TensorFlow使用TFRecords文件訓練樣本的步驟
在生成文件名的序列中,設定epoch數量
訓練時,設定為無窮循環
在讀取數據時,如果捕捉到錯誤,終止
source code:tensorflow-master\tensorflow\examples\how_tos\reading_data\fully_connected_reader.py(1.2.1)
(https://blog.csdn.net/fontthrone/article/details/76728083 )
import tensorflow as tf
import os
# from tensorflow.contrib.learn.python.learn.datasets import mnist
# 註意上面的這個mnist 與 example 中的 mnist 是不同的,本文件中請使用下面的那個 mnist
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
import argparse
import os.path
import sys
import time
from tensorflow.examples.tutorials.mnist import mnist
# Basic model parameters as external flags.
FLAGS = None
# This part of the code is added by FontTian,which comes from the source code of tensorflow.examples.tutorials.mnist
# The MNIST images are always 28x28 pixels.
# IMAGE_SIZE = 28
# IMAGE_PIXELS = IMAGE_SIZE * IMAGE_SIZE
# Constants used for dealing with the files, matches convert_to_records.
TRAIN_FILE = 'train.tfrecords'
VALIDATION_FILE = 'validation.tfrecords'
def read_and_decode(filename_queue):
reader = tf.TFRecordReader()
_, serialized_example = reader.read(filename_queue)
features = tf.parse_single_example(
serialized_example,
# Defaults are not specified since both keys are required.
# 必須寫明faetures 中的 key 的名稱
features={
'image_raw': tf.FixedLenFeature([], tf.string),
'label': tf.FixedLenFeature([], tf.int64),
})
# Convert from a scalar string tensor (whose single string has
# length mnist.IMAGE_PIXELS) to a uint8 tensor with shape
# [mnist.IMAGE_PIXELS].
# 將一個標量字符串張量(其單個字符串的長度是mnist.image像素) # 0 維的Tensor
# 轉換為一個帶有形狀mnist.圖像像素的uint8張量。 # 一維的Tensor
image = tf.decode_raw(features['image_raw'], tf.uint8)
# print(tf.shape(image)) # Tensor("input/Shape:0", shape=(1,), dtype=int32)
image.set_shape([mnist.IMAGE_PIXELS])
# print(tf.shape(image)) # Tensor("input/Shape_1:0", shape=(1,), dtype=int32)
# OPTIONAL: Could reshape into a 28x28 image and apply distortions
# here. Since we are not applying any distortions in this
# example, and the next step expects the image to be flattened
# into a vector, we don't bother.
# Convert from [0, 255] -> [-0.5, 0.5] floats.
image = tf.cast(image, tf.float32) * (1. / 255) - 0.5
# print(tf.shape(image)) # Tensor("input/Shape_2:0", shape=(1,), dtype=int32)
# Convert label from a scalar uint8 tensor to an int32 scalar.
label = tf.cast(features['label'], tf.int32)
# print(tf.shape(label)) # Tensor("input/Shape_3:0", shape=(0,), dtype=int32)
return image, label
# 使用 tf.train.shuffle_batch 將前面生成的樣本隨機化,獲得一個最小批次的張量
def inputs(train, batch_size, num_epochs):
"""Reads input data num_epochs times.
Args:
train: Selects between the training (True) and validation (False) data.
batch_size: Number of examples per returned batch.
num_epochs: Number of times to read the input data, or 0/None to
train forever.
Returns:
A tuple (images, labels), where:
* images is a float tensor with shape [batch_size, mnist.IMAGE_PIXELS]
in the range [-0.5, 0.5].
* labels is an int32 tensor with shape [batch_size] with the true label,
a number in the range [0, mnist.NUM_CLASSES).
Note that an tf.train.QueueRunner is added to the graph, which
must be run using e.g. tf.train.start_queue_runners().
輸入參數:
train: Selects between the training (True) and validation (False) data.
batch_size: 訓練的每一批有多少個樣本
num_epochs: 讀取輸入數據的次數, or 0/None 表示永遠訓練下去
返回結果:
A tuple (images, labels), where:
* images is a float tensor with shape [batch_size, mnist.IMAGE_PIXELS]
範圍: [-0.5, 0.5].
* labels is an int32 tensor with shape [batch_size] with the true label,
範圍: [0, mnist.NUM_CLASSES).
註意 : tf.train.QueueRunner 被添加進 graph, 它必須用 tf.train.start_queue_runners() 來啟動線程.
"""
if not num_epochs: num_epochs = None
filename = os.path.join(FLAGS.train_dir,
TRAIN_FILE if train else VALIDATION_FILE)
with tf.name_scope('input'):
# tf.train.string_input_producer 返回一個 QueueRunner,裏面有一個 FIFQueue
filename_queue = tf.train.string_input_producer(
[filename], num_epochs=num_epochs)
# 如果樣本數據很大,可以分成若幹文件,把文件名列表傳入
# Even when reading in multiple threads, share the filename queue.
image, label = read_and_decode(filename_queue)
# Shuffle the examples and collect them into batch_size batches.
# (Internally uses a RandomShuffleQueue.)
# We run this in two threads to avoid being a bottleneck.
images, sparse_labels = tf.train.shuffle_batch(
[image, label], batch_size=batch_size, num_threads=2,
capacity=1000 + 3 * batch_size,
# Ensures a minimum amount of shuffling of examples.
# 留下一部分隊列,來保證每次有足夠的數據做隨機打亂
min_after_dequeue=1000)
return images, sparse_labels
def run_training():
"""Train MNIST for a number of steps."""
# Tell TensorFlow that the model will be built into the default Graph.
with tf.Graph().as_default():
# Input images and labels.
images, labels = inputs(train=True, batch_size=FLAGS.batch_size,
num_epochs=FLAGS.num_epochs)
# 構建一個從推理模型來預測數據的圖
logits = mnist.inference(images,
FLAGS.hidden1,
FLAGS.hidden2)
# Add to the Graph the loss calculation.
# 定義損失函數
loss = mnist.loss(logits, labels)
# 將模型添加到圖操作中
train_op = mnist.training(loss, FLAGS.learning_rate)
# 初始化變量的操作
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
# Create a session for running operations in the Graph.
# 在圖中創建一個用於運行操作的會話
sess = tf.Session()
# 初始化變量,註意:string_input_product 內部創建了一個epoch計數器
sess.run(init_op)
# Start input enqueue threads.
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=sess, coord=coord)
try:
step = 0
while not coord.should_stop():
start_time = time.time()
# Run one step of the model. The return values are
# the activations from the `train_op` (which is
# discarded) and the `loss` op. To inspect the values
# of your ops or variables, you may include them in
# the list passed to sess.run() and the value tensors
# will be returned in the tuple from the call.
_, loss_value = sess.run([train_op, loss])
duration = time.time() - start_time
# Print an overview fairly often.
if step % 100 == 0:
print('Step %d: loss = %.2f (%.3f sec)' % (step, loss_value,
duration))
step += 1
except tf.errors.OutOfRangeError:
print('Done training for %d epochs, %d steps.' % (FLAGS.num_epochs, step))
finally:
# 通知其他線程關閉
coord.request_stop()
# Wait for threads to finish.
coord.join(threads)
sess.close()
def main(_):
run_training()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument(
'--learning_rate',
type=float,
default=0.01,
help='Initial learning rate.'
)
parser.add_argument(
'--num_epochs',
type=int,
default=2,
help='Number of epochs to run trainer.'
)
parser.add_argument(
'--hidden1',
type=int,
default=128,
help='Number of units in hidden layer 1.'
)
parser.add_argument(
'--hidden2',
type=int,
default=32,
help='Number of units in hidden layer 2.'
)
parser.add_argument(
'--batch_size',
type=int,
default=100,
help='Batch size.'
)
parser.add_argument(
'--train_dir',
type=str,
default='/tmp/data',
help='Directory with the training data.'
)
FLAGS, unparsed = parser.parse_known_args()
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
TF從文件中讀取數據