【Tensorflow】Cifar-10
tensorflow官方CIFAR-10 教程學習筆記
主要包括以下四部分:
環境:
- python 3.6
- tensorflow 1.8
目錄
cifar10
- tf.app.flags.DEFINE_xxx() && tf.app.flags.FLAGS.flag_xxx
# tf.app.flags.DEFINE_string("param_name", "default_val", "description")
tf.app.flags.DEFINE_xxx() :新增命令列可選引數,主要是在命令列執行程式時傳參使用,如果不傳引數,則執行預設引數。
tf.app.flags.FLAGS.flag_xxx則取得上述引數。
- activation_summary(x)
tensorboard 的相關建立初始化。
- _variable_on_cpu(name, shape, initializer)
指定在cpu建立/獲取變數。
tf.device() 指定模型執行的具體裝置,可以指定執行在GPU還是CUP上,以及哪塊GPU上。如果安裝的是GPU版本的tensorflow,機器上有支援的GPU,也正確安裝了顯示卡驅動、CUDA和cuDNN,預設情況下,Session會在GPU上執行。tensorflow中不同的GPU使用/gpu:0和/gpu:1區分,而CPU不區分裝置號,統一使用 /cpu:0
- _variable_with_weight_decay(name, shape, stddev, wd)
建立/獲取變數,並把帶權重衰減的變數值加入loss。
- distorted_inputs()&inputs()
cifar10 data輸入,這2個函式會從CIFAR-10二進位制檔案中讀取圖片檔案。distorted做了一系列隨機變換(翻轉、亮度對比度等)的方法來人為的擴充套件資料集。
- inference(images)
預測,兩個卷積層兩個全連線層。
- loss(logits, labels)
計算loss。
sparse_labels = tf.reshape(labels, [FLAGS.batch_size, 1])
行向量reshape成列向量:
例:labels = sparse_labels =
indices = tf.reshape(tf.range(FLAGS.batch_size), [FLAGS.batch_size, 1])
生成index大小的index:
例:indices =
concated = tf.concat([indices, sparse_labels], 1)
連線以上兩個列向量,concated shape->[batch_size,2]
例:concated =
dense_labels = tf.sparse_to_dense(concated,
[FLAGS.batch_size, NUM_CLASSES],
1.0, 0.0)
得到[FLAGS.batch_size, NUM_CLASSES]大小的onehot標籤矩陣
例:
[ [0,1,0,0,0,0,0,0,0,0]
[0,0,0,1,0,0,0,0,0,0]
[0,0,0,0,0,1,0,0,0,0]
[0,0,0,0,0,0,0,1,0,0]
[0,0,0,0,0,0,0,0,0,1]]
- def _add_loss_summaries(total_loss)
moving average loss op.
apply方法會為每個變數(也可以指定特定變數)建立各自的shadow variable, 即影子變數。
之所以叫影子變數,是因為它會全程跟隨訓練中的模型變數。影子變數會被初始化為模型變數的值,
然後,每訓練一個step,就更新一次。
- train(total_loss, global_step)
cifar-10模型訓練
tf.control_dependencies()
用來控制計算流圖的,給圖中的某些計算指定順序:
with g.control_dependencies([a, b, c]):
# `d` and `e` will only run after `a`, `b`, and `c` have executed.
d = ...
e = ...
- maybe_download_and_extract()
下載Binary 版的cifar10資料。
完整程式碼:
"""Builds the CIFAR-10 network.
Summary of available functions:
# Compute input images and labels for training. If you would like to run
# evaluations, use input() instead.
inputs, labels = distorted_inputs()
# Compute inference on the model inputs to make a prediction.
predictions = inference(inputs)
# Compute the total loss of the prediction with respect to the labels.
loss = loss(predictions, labels)
# Create a graph to run one step of training with respect to the loss.
train_op = train(loss, global_step)
"""
# pylint: disable=missing-docstring
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import gzip
import os
import re
import sys
import tarfile
import tensorflow.python.platform
from six.moves import urllib
import tensorflow as tf
import cifar10_input
FLAGS = tf.app.flags.FLAGS
# Basic model parameters.
tf.app.flags.DEFINE_integer('batch_size', 128,
"""Number of images to process in a batch.""")
tf.app.flags.DEFINE_string('data_dir', 'cifar10_data',
"""Path to the CIFAR-10 data directory.""")
# Global constants describing the CIFAR-10 data set.
IMAGE_SIZE = cifar10_input.IMAGE_SIZE
NUM_CLASSES = cifar10_input.NUM_CLASSES
NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN = cifar10_input.NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN
NUM_EXAMPLES_PER_EPOCH_FOR_EVAL = cifar10_input.NUM_EXAMPLES_PER_EPOCH_FOR_EVAL
# Constants describing the training process.
MOVING_AVERAGE_DECAY = 0.9999 # The decay to use for the moving average.
NUM_EPOCHS_PER_DECAY = 350.0 # Epochs after which learning rate decays.
LEARNING_RATE_DECAY_FACTOR = 0.1 # Learning rate decay factor.
INITIAL_LEARNING_RATE = 0.1 # Initial learning rate.
# If a model is trained with multiple GPU's prefix all Op names with tower_name
# to differentiate the operations. Note that this prefix is removed from the
# names of the summaries when visualizing a model.
TOWER_NAME = 'tower'
DATA_URL = 'http://www.cs.toronto.edu/~kriz/cifar-10-binary.tar.gz'
def _activation_summary(x):
"""Helper to create summaries for activations.
Creates a summary that provides a histogram of activations.
Creates a summary that measure the sparsity of activations.
Args:
x: Tensor
Returns:
nothing
"""
# Remove 'tower_[0-9]/' from the name in case this is a multi-GPU training
# session. This helps the clarity of presentation on tensorboard.
# [0-9]* :這個匹配0個或0個以上的任何數字
tensor_name = re.sub('%s_[0-9]*/' % TOWER_NAME, '', x.op.name)
tf.summary.histogram(tensor_name + '/activations', x)
tf.summary.scalar(tensor_name + '/sparsity', tf.nn.zero_fraction(x))
def _variable_on_cpu(name, shape, initializer):
"""Helper to create a Variable stored on CPU memory.
Args:
name: name of the variable
shape: list of ints
initializer: initializer for Variable
Returns:
Variable Tensor
"""
# 在TensorFlow中,模型可以在本地的GPU和CPU中執行,使用者可以指定模型執行的裝置
with tf.device('/cpu:0'):
# 獲取已存在的變數(要求不僅名字,而且初始化方法等各個引數都一樣),如果不存在,就新建一個。
# 可以用各種初始化方法,不用明確指定值。
var = tf.get_variable(name, shape, initializer=initializer)
return var
def _variable_with_weight_decay(name, shape, stddev, wd):
"""Helper to create an initialized Variable with weight decay.權重衰減
Note that the Variable is initialized with a truncated normal distribution.
A weight decay is added only if one is specified.
Args:
name: name of the variable
shape: list of ints
stddev: standard deviation of a truncated Gaussian
wd: add L2Loss weight decay multiplied by this float. If None, weight
decay is not added for this Variable.
Returns:
Variable Tensor
"""
var = _variable_on_cpu(name, shape,
tf.truncated_normal_initializer(stddev=stddev))
if wd:
# 為了控制模型的複雜程度,會在loss function中加入正則項
# Computes half the L2 norm of a tensor without the `sqrt`
# output = sum(t ** 2) / 2
weight_decay = tf.multiply(tf.nn.l2_loss(var), wd, name='weight_loss')
# # 對應的正則項加入集合losses
# 把變數放入一個集合,把很多變數變成一個列表
tf.add_to_collection('losses', weight_decay)
return var
def distorted_inputs():
"""Construct distorted input for CIFAR training using the Reader ops.
Returns:
images: Images. 4D tensor of [batch_size, IMAGE_SIZE, IMAGE_SIZE, 3] size.
labels: Labels. 1D tensor of [batch_size] size.
Raises:
ValueError: If no data_dir
"""
if not FLAGS.data_dir:
raise ValueError('Please supply a data_dir')
# 用於路徑拼接檔案路徑
data_dir = os.path.join(FLAGS.data_dir, 'cifar-10-batches-bin')
return cifar10_input.distorted_inputs(data_dir=data_dir,
batch_size=FLAGS.batch_size)
def inputs(eval_data):
"""Construct input for CIFAR evaluation using the Reader ops.
Args:
eval_data: bool, indicating if one should use the train or eval data set.
Returns:
images: Images. 4D tensor of [batch_size, IMAGE_SIZE, IMAGE_SIZE, 3] size.
labels: Labels. 1D tensor of [batch_size] size.
Raises:
ValueError: If no data_dir
"""
if not FLAGS.data_dir:
raise ValueError('Please supply a data_dir')
data_dir = os.path.join(FLAGS.data_dir, 'cifar-10-batches-bin')
return cifar10_input.inputs(eval_data=eval_data, data_dir=data_dir,
batch_size=FLAGS.batch_size)
# 模型的預測流程
def inference(images):
"""Build the CIFAR-10 model.
Args:
images: Images returned from distorted_inputs() or inputs().
Returns:
Logits.
"""
# We instantiate all variables using tf.get_variable() instead of
# tf.Variable() in order to share variables across multiple GPU training runs.
# If we only ran this model on a single GPU, we could simplify this function
# by replacing all instances of tf.get_variable() with tf.Variable().
#
# conv1
with tf.variable_scope('conv1') as scope:
kernel = _variable_with_weight_decay('weights', shape=[5, 5, 3, 64],
stddev=1e-4, wd=0.0)
conv = tf.nn.conv2d(images, kernel, [1, 1, 1, 1], padding='SAME')
biases = _variable_on_cpu('biases', [64], tf.constant_initializer(0.0))
bias = tf.nn.bias_add(conv, biases)
conv1 = tf.nn.relu(bias, name=scope.name)
_activation_summary(conv1)
# pool1
pool1 = tf.nn.max_pool(conv1, ksize=[1, 3, 3, 1], strides=[1, 2, 2, 1],
padding='SAME', name='pool1')
# norm1
norm1 = tf.nn.lrn(pool1, 4, bias=1.0, alpha=0.001 / 9.0, beta=0.75,
name='norm1')
# conv2
with tf.variable_scope('conv2') as scope:
kernel = _variable_with_weight_decay('weights', shape=[5, 5, 64, 64],
stddev=1e-4, wd=0.0)
conv = tf.nn.conv2d(norm1, kernel, [1, 1, 1, 1], padding='SAME')
biases = _variable_on_cpu('biases', [64], tf.constant_initializer(0.1))
bias = tf.nn.bias_add(conv, biases)
conv2 = tf.nn.relu(bias, name=scope.name)
_activation_summary(conv2)
# norm2
norm2 = tf.nn.lrn(conv2, 4, bias=1.0, alpha=0.001 / 9.0, beta=0.75,
name='norm2')
# pool2
pool2 = tf.nn.max_pool(norm2, ksize=[1, 3, 3, 1],
strides=[1, 2, 2, 1], padding='SAME', name='pool2')
# local3 基於修正線性啟用的全連線層.
with tf.variable_scope('local3') as scope:
# Move everything into depth so we can perform a single matrix multiply.
dim = 1
for d in pool2.get_shape()[1:].as_list(): # get_shape,返回的是一個元組 ;as_list(): 以list形式,[1:]:w*h*channel
dim *= d
reshape = tf.reshape(pool2, [FLAGS.batch_size, dim])
weights = _variable_with_weight_decay('weights', shape=[dim, 384],
stddev=0.04, wd=0.004)
biases = _variable_on_cpu('biases', [384], tf.constant_initializer(0.1))
local3 = tf.nn.relu(tf.matmul(reshape, weights) + biases, name=scope.name)
_activation_summary(local3)
# local4
with tf.variable_scope('local4') as scope:
weights = _variable_with_weight_decay('weights', shape=[384, 192],
stddev=0.04, wd=0.004)
biases = _variable_on_cpu('biases', [192], tf.constant_initializer(0.1))
local4 = tf.nn.relu(tf.matmul(local3, weights) + biases, name=scope.name)
_activation_summary(local4)
# softmax, i.e. softmax(WX + b)
with tf.variable_scope('softmax_linear') as scope:
weights = _variable_with_weight_decay('weights', [192, NUM_CLASSES],
stddev=1/192.0, wd=0.0)
biases = _variable_on_cpu('biases', [NUM_CLASSES],
tf.constant_initializer(0.0))
softmax_linear = tf.add(tf.matmul(local4, weights), biases, name=scope.name)
_activation_summary(softmax_linear)
return softmax_linear
def loss(logits, labels):
"""Add L2Loss to all the trainable variables.
Add summary for for "Loss" and "Loss/avg".
Args:
logits: Logits from inference().
labels: Labels from distorted_inputs or inputs(). 1-D tensor
of shape [batch_size]
Returns:
Loss tensor of type float.
"""
# Reshape the labels into a dense Tensor of
# shape [batch_size, NUM_CLASSES].
sparse_labels = tf.reshape(labels, [FLAGS.batch_size, 1])
# 生成一個index表明一個batch裡面每個樣本對應的序號
indices = tf.reshape(tf.range(FLAGS.batch_size), [FLAGS.batch_size, 1])
# 在維度1連線 [FLAGS.batch_size, 2]
concated = tf.concat([indices, sparse_labels], 1)
# 呼叫tf.sparse_to_dense輸出一個onehot標籤的矩陣
dense_labels = tf.sparse_to_dense(concated,
[FLAGS.batch_size, NUM_CLASSES],
1.0, 0.0)
# Calculate the average cross entropy loss across the batch.
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits = logits, labels = dense_labels, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# The total loss is defined as the cross entropy loss plus all of the weight
# decay terms (L2 loss).
return tf.add_n(tf.get_collection('losses'), name='total_loss')
def _add_loss_summaries(total_loss):
"""Add summaries for losses in CIFAR-10 model.
Generates moving average for all losses and associated summaries for
visualizing the performance of the network.
Args:
total_loss: Total loss from loss().
Returns:
loss_averages_op: op for generating moving averages of losses.
"""
# Compute the moving average of all individual losses and the total loss.
loss_averages = tf.train.ExponentialMovingAverage(0.9, name='avg')
losses = tf.get_collection('losses')
loss_averages_op = loss_averages.apply(losses + [total_loss])
'''
apply方法會為每個變數(也可以指定特定變數)建立各自的shadow variable, 即影子變數。
之所以叫影子變數,是因為它會全程跟隨訓練中的模型變數。影子變數會被初始化為模型變數的值,
然後,每訓練一個step,就更新一次。
'''
# Attach a scalar summary to all individual losses and the total loss; do the
# same for the averaged version of the losses.
for l in losses + [total_loss]:
# Name each loss as '(raw)' and name the moving average version of the loss
# as the original loss name.
tf.summary.scalar(l.op.name +' (raw)', l)
tf.summary.scalar(l.op.name, loss_averages.average(l))
return loss_averages_op
def train(total_loss, global_step):
"""Train CIFAR-10 model.
Create an optimizer and apply to all trainable variables. Add moving
average for all trainable variables.
Args:
total_loss: Total loss from loss().
global_step: Integer Variable counting the number of training steps
processed.
Returns:
train_op: op for training.
"""
# Variables that affect learning rate.
num_batches_per_epoch = NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN / FLAGS.batch_size
decay_steps = int(num_batches_per_epoch * NUM_EPOCHS_PER_DECAY)
# Decay the learning rate exponentially based on the number of steps.
lr = tf.train.exponential_decay(INITIAL_LEARNING_RATE, # The initial learning rate.
global_step, # Global step to use for the decay computation. Must not be negative.
decay_steps,
LEARNING_RATE_DECAY_FACTOR, # 衰減速率,即每一次學習都衰減為原來的LEARNING_RATE_DECAY_FACTOR
staircase=True) # If True decay the learning rate at discrete intervals
# staircase=True,那就表明每decay_steps次計算學習速率變化,更新原始學習速率,如果是False,那就是每一步都更新學習速率。
tf.summary.scalar('learning_rate', lr)
# Generate moving averages of all losses and associated summaries.
loss_averages_op = _add_loss_summaries(total_loss)
# Compute gradients.
# 每步更新loss
with tf.control_dependencies([loss_averages_op]):
opt = tf.train.GradientDescentOptimizer(lr)
grads = opt.compute_gradients(total_loss)
# Apply gradients.
apply_gradient_op = opt.apply_gradients(grads, global_step=global_step)
# Add histograms for trainable variables.
# 返回需要訓練的變數列表
for var in tf.trainable_variables():
tf.summary.histogram(var.op.name, var)
# Add histograms for gradients.
for grad, var in grads:
if grad is not None:
tf.summary.histogram(var.op.name + '/gradients', grad)
# Track the moving averages of all trainable variables.
variable_averages = tf.train.ExponentialMovingAverage(
MOVING_AVERAGE_DECAY, global_step)
variables_averages_op = variable_averages.apply(tf.trainable_variables())
with tf.control_dependencies([apply_gradient_op, variables_averages_op]):
# Does nothing. Only useful as a placeholder for control edges
train_op = tf.no_op(name='train')
return train_op
def maybe_download_and_extract():
"""Download and extract the tarball from Alex's website."""
dest_directory = FLAGS.data_dir
if not os.path.exists(dest_directory):
os.makedirs(dest_directory)
filename = DATA_URL.split('/')[-1]
filepath = os.path.join(dest_directory, filename)
if not os.path.exists(filepath):
def _progress(count, block_size, total_size):
'''
:param blocknum: 已下載資料塊
:param blocksize: 資料塊大小
:param totalsize: 遠端檔案大小
:return:
'''
sys.stdout.write('\r>> Downloading %s %.1f%%' % (filename,
float(count * block_size) / float(total_size) * 100.0))
sys.stdout.flush()
# urllib.request.urlretrieve(url, local, callback) 從遠端下載資料
filepath, _ = urllib.request.urlretrieve(DATA_URL, filepath,
reporthook=_progress)
print()
statinfo = os.stat(filepath)
print('Successfully downloaded', filename, statinfo.st_size, 'bytes.')
tarfile.open(filepath, 'r:gz').extractall(dest_directory)
cifar10_input
- read_cifar10(filename_queue)
從檔名佇列讀取資料,返回CIFAR10Record類。
class CIFAR10Record(object):
pass
空類。
# 讀取固定長度位元組
reader = tf.FixedLengthRecordReader(record_bytes=record_bytes)
# 每次在filename_queue中讀取record_bytes位元組資訊 下次呼叫時會接著上次讀取的位置繼續讀取檔案
result.key, value = reader.read(filename_queue)
每次讀取固定長的的資料,即 1(label)+3*32*32(img)大小.
record_bytes = tf.decode_raw(value, tf.uint8)
tf.decode_raw將原來編碼為字串型別的變數重新變回來,這個方法在資料集dataset中很常用,因為製作圖片源資料一般寫進tfrecord裡用to_bytes的形式,也就是字串。這裡將原始資料取出來,必須制定原始資料的格式,原始資料是什麼格式這裡解析必須是什麼格式,要不然會出現形狀的不對應問題!
depth_major = tf.reshape(tf.slice(record_bytes, [label_bytes], [image_bytes]),
[result.depth, result.height, result.width])
# Convert from [depth, height, width] to [height, width, depth].
# 轉置
result.uint8image = tf.transpose(depth_major, [1, 2, 0])
Binary 版的cifar10資料, 格式如下:
<1 x label><3072 x pixel>
...
<1 x label><3072 x pixel>
第一位元組是第一張圖片的標籤值,數值範圍是 0-9,表示 10 類,record_bytes[1:] 大小為3072(3*32*32),3072中前1024個表示Red通道資料,中間1024個表示Green通道資料,最後1024個表示Blue通道資料,資料範圍是0-255,表示畫素點灰度.後面的卷積等操作需要32 * 32 * 3格式的資料,所以這裡首先reshape成[3,32,32],然後利用tf.transpose轉置,uint8image shape[32,32,3].
- _generate_image_and_label_batch(image, label, min_queue_examples,batch_size)
得到亂序的資料batch。
- distorted_inputs(data_dir, batch_size) & inputs(eval_data, data_dir, batch_size)
輸入資料。
完整程式碼:
"""Routine for decoding the CIFAR-10 binary file format."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import tensorflow.python.platform
from six.moves import xrange # pylint: disable=redefined-builtin
import tensorflow as tf
from tensorflow.python.platform import gfile
# Process images of this size. Note that this differs from the original CIFAR
# image size of 32 x 32. If one alters this number, then the entire model
# architecture will change and any model would need to be retrained.
IMAGE_SIZE = 24
# Global constants describing the CIFAR-10 data set.
NUM_CLASSES = 10
NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN = 50000
NUM_EXAMPLES_PER_EPOCH_FOR_EVAL = 10000
def read_cifar10(filename_queue):
"""Reads and parses examples from CIFAR10 data files.
Recommendation: if you want N-way read parallelism, call this function
N times. This will give you N independent Readers reading different
files & positions within those files, which will give better mixing of
examples.
Args:
filename_queue: A queue of strings with the filenames to read from.
Returns:
An object representing a single example, with the following fields:
height: number of rows in the result (32)
width: number of columns in the result (32)
depth: number of color channels in the result (3)
key: a scalar string Tensor describing the filename & record number
for this example.
label: an int32 Tensor with the label in the range 0..9.
uint8image: a [height, width, depth] uint8 Tensor with the image data
@param filename_queue 要讀取的檔名佇列
@return 某個物件,具有以下欄位:
height: 結果中的行數 (32)
width: 結果中的列數 (32)
depth: 結果中顏色通道數(3)
key: 一個描述當前抽樣資料的檔名和記錄數的標量字串
label: 一個 int32型別的標籤,取值範圍 0..9.
uint8image: 一個[height, width, depth]維度的影象資料
"""
class CIFAR10Record(object):
pass
result = CIFAR10Record()
# Dimensions of the images in the CIFAR-10 dataset.
# See http://www.cs.toronto.edu/~kriz/cifar.html for a description of the
# input format.
label_bytes = 1 # 2 for CIFAR-100
result.height = 32
result.width = 32
result.depth = 3
image_bytes = result.height * result.width * result.depth
# Every record consists of a label followed by the image, with a
# fixed number of bytes for each.
# 每個記錄都包含標籤資訊和圖片資訊,每個記錄都有固定的位元組數(3073 = 1 + 3072)3*32*32 = 3072
record_bytes = label_bytes + image_bytes
# Read a record, getting filenames from the filename_queue. No
# header or footer in the CIFAR-10 format, so we leave header_bytes
# and footer_bytes at their default of 0.
# 讀取固定長度位元組
reader = tf.FixedLengthRecordReader(record_bytes=record_bytes)
# 每次在filename_queue中讀取record_bytes位元組資訊 下次呼叫時會接著上次讀取的位置繼續讀取檔案
result.key, value = reader.read(filename_queue)
# Convert from a string to a vector of uint8 that is record_bytes long.
# 將原來編碼為字串型別的變數重新變回來
record_bytes = tf.decode_raw(value, tf.uint8)
# The first bytes represent the label, which we convert from uint8->int32.
# 從輸入資料input中提取出一塊切片
# 第1維偏移0,label_bytes(1)大小的數
result.label = tf.cast(
tf.slice(record_bytes, [0], [label_bytes]), tf.int32)
# The remaining bytes after the label represent the image, which we reshape
# from [depth * height * width] to [depth, height, width].
# 第1維偏移0,image_bytes大小的數
# 3072中前1024個表示Red通道資料,中間1024個表示Green通道資料,最後1024個表示Blue通道資料,所以reshape後為3*height*width
depth_major = tf.reshape(tf.slice(record_bytes, [label_bytes], [image_bytes]),
[result.depth, result.height, result.width])
# Convert from [depth, height, width] to [height, width, depth].
# 轉置
result.uint8image = tf.transpose(depth_major, [1, 2, 0])
return result
def _generate_image_and_label_batch(image, label, min_queue_examples,
batch_size):
"""Construct a queued batch of images and labels.
Args:
image: 3-D Tensor of [height, width, 3] of type.float32.
label: 1-D Tensor of type.int32
min_queue_examples: int32, minimum number of samples to retain
in the queue that provides of batches of examples.
batch_size: Number of images per batch.
Returns:
images: Images. 4D tensor of [batch_size, height, width, 3] size.
labels: Labels. 1D tensor of [batch_size] size.
"""
# Create a queue that shuffles the examples, and then
# read 'batch_size' images + labels from the example queue.
num_preprocess_threads = 16
# # Creates batches by randomly shuffling tensors 返回值是一個batch的樣本和樣本標籤
# 將佇列中資料打亂後,再讀取出來,因此佇列中剩下的資料也是亂序的
images, label_batch = tf.train.shuffle_batch(
[image, label], # tensor_list
batch_size=batch_size, # 返回的一個batch樣本集的樣本個數
num_threads=num_preprocess_threads,
capacity=min_queue_examples + 3 * batch_size,
min_after_dequeue=min_queue_examples) #min_after_dequeue,一定要保證這引數小於capacity引數的值,否則會出錯。
# Display the training images in the visualizer.
tf.summary.image('images', images)
return images, tf.reshape(label_batch, [batch_size])
def distorted_inputs(data_dir, batch_size):
"""Construct distorted input for CIFAR training using the Reader ops.
Args:
data_dir: Path to the CIFAR-10 data directory.
batch_size: Number of images per batch.
Returns:
images: Images. 4D tensor of [batch_size, IMAGE_SIZE, IMAGE_SIZE, 3] size.
labels: Labels. 1D tensor of [batch_size] size.
"""
# 獲取當前目錄,並組合成新目錄
filenames = [os.path.join(data_dir, 'data_batch_%d.bin' % i)
for i in xrange(1, 6)]
for f in filenames:
if not gfile.Exists(f):
raise ValueError('Failed to find file: ' + f)
# Create a queue that produces the filenames to read.
# 把需要的全部檔案打包為一個tf內部的queue型別
filename_queue = tf.train.string_input_producer(filenames)
'''
tf.train.string_input_producer建立了一個這樣的執行緒,新增QueueRunner到資料流圖中
string_input_producer來生成一個先入先出的佇列, 檔案閱讀器會需要它來讀取資料。
'''
# Read examples from files in the filename queue.
read_input = read_cifar10(filename_queue)
reshaped_image = tf.cast(read_input.uint8image, tf.float32)
height = IMAGE_SIZE
width = IMAGE_SIZE
# Image processing for training the network. Note the many random
# distortions applied to the image.
# Randomly crop a [height, width] section of the image.
distorted_image = tf.random_crop(reshaped_image, [height, width,3]) # 隨機裁剪為24 * 24
# Randomly flip the image horizontally.
distorted_image = tf.image.random_flip_left_right(distorted_image) # 隨機左右翻轉
# Because these operations are not commutative, consider randomizing
# randomize the order their operation.
distorted_image = tf.image.random_brightness(distorted_image,
max_delta=63)
distorted_image = tf.image.random_contrast(distorted_image,
lower=0.2, upper=1.8)
# Subtract off the mean and divide by the variance of the pixels.
# 白化(標準化)操作。tf.image.per_image_standardization 將代表一張圖片的三維矩陣中的數字均值變為0,方差變為1。
float_image = tf.image.per_image_standardization(distorted_image)
# Ensure that the random shuffling has good mixing properties.
min_fraction_of_examples_in_queue = 0.4
min_queue_examples = int(NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN *
min_fraction_of_examples_in_queue)
print ('Filling queue with %d CIFAR images before starting to train. '
'This will take a few minutes.' % min_queue_examples)
# Generate a batch of images and labels by building up a queue of examples.
return _generate_image_and_label_batch(float_image, read_input.label,
min_queue_examples, batch_size)
def inputs(eval_data, data_dir, batch_size):
"""Construct input for CIFAR evaluation using the Reader ops.
Args:
eval_data: bool, indicating if one should use the train or eval data set.
data_dir: Path to the CIFAR-10 data directory.
batch_size: Number of images per batch.
Returns:
images: Images. 4D tensor of [batch_size, IMAGE_SIZE, IMAGE_SIZE, 3] size.
labels: Labels. 1D tensor of [batch_size] size.
"""
# 根據eval_data決定讀入train or eval資料
if not eval_data:
filenames = [os.path.join(data_dir, 'data_batch_%d.bin' % i)
for i in xrange(1, 6)]
num_examples_per_epoch = NUM_EXAMPLES_PER_EPOCH_FOR_TRAIN
else:
filenames = [os.path.join(data_dir, 'test_batch.bin')]
num_examples_per_epoch = NUM_EXAMPLES_PER_EPOCH_FOR_EVAL
for f in filenames:
if not gfile.Exists(f):
raise ValueError('Failed to find file: ' + f)
# 同上
# Create a queue that produces the filenames to read.
filename_queue = tf.train.string_input_producer(filenames)
# Read examples from files in the filename queue.
read_input = read_cifar10(filename_queue)
reshaped_image = tf.cast(read_input.uint8image, tf.float32)
height = IMAGE_SIZE
width = IMAGE_SIZE
# Image processing for evaluation.
# Crop the central [height, width] of the image.
resized_image = tf.image.resize_image_with_crop_or_pad(reshaped_image,
width, height)
# Subtract off the mean and divide by the variance of the pixels.
float_image = tf.image.per_image_standardization(resized_image)
# Ensure that the random shuffling has good mixing properties.
min_fraction_of_examples_in_queue = 0.4
min_queue_examples = int(num_examples_per_epoch *
min_fraction_of_examples_in_queue)
# Generate a batch of images and labels by building up a queue of examples.
return _generate_image_and_label_batch(float_image, read_input.label,
min_queue_examples, batch_size)
tf.Graph().as_default():
tf.Graph() 表示例項化了一個類,一個用於 tensorflow 計算和表示用的資料流圖。tf.Graph().as_default() 表示將這個類例項,也就是新生成的圖作為整個 tensorflow 執行環境的預設圖。
tf.train.start_queue_runners(sess=sess)
啟動所有的QueueRunners。
cifar10_train
"""A binary to train CIFAR-10 using a single GPU.
Accuracy:
cifar10_train.py achieves ~86% accuracy after 100K steps (256 epochs of
data) as judged by cifar10_eval.py.
Speed: With batch_size 128.
System | Step Time (sec/batch) | Accuracy
------------------------------------------------------------------
1 Tesla K20m | 0.35-0.60 | ~86% at 60K steps (5 hours)
1 Tesla K40m | 0.25-0.35 | ~86% at 100K steps (4 hours)
Usage:
Please see the tutorial and website for how to download the CIFAR-10
data set, compile the program and train the model.
http://tensorflow.org/tutorials/deep_cnn/
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from datetime import datetime
import os.path
import time
import tensorflow.python.platform
from tensorflow.python.platform import gfile
import numpy as np
from six.moves import xrange # pylint: disable=redefined-builtin
import tensorflow as tf
import cifar10
FLAGS = tf.app.flags.FLAGS
# 命令函引數
tf.app.flags.DEFINE_string('train_dir', 'cifar10_train',
"""Directory where to write event logs """
"""and checkpoint.""")
tf.app.flags.DEFINE_integer('max_steps', 100000,
"""Number of batches to run.""")
tf.app.flags.DEFINE_boolean('log_device_placement', False,
"""Whether to log device placement.""")
def train():
"""Train CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Get images and labels for CIFAR-10.
images, labels = cifar10.distorted_inputs()
# Build a Graph that computes the logits predictions from the
# inference model.
logits = cifar10.inference(images)
# Calculate loss.
loss = cifar10.loss(logits, labels)
# Build a Graph that trains the model with one batch of examples and
# updates the model parameters.
train_op = cifar10.train(loss, global_step)
# Create a saver.
saver = tf.train.Saver(tf.global_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(
log_device_placement=FLAGS.log_device_placement))
sess.run(init)
# Start the queue runners.
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.summary.FileWriter(FLAGS.train_dir,
graph_def=sess.graph_def)
for step in xrange(FLAGS.max_steps):
start_time = time.time()
_, loss_value = sess.run([train_op, loss])
duration = time.time() - start_time
assert not np.isnan(loss_value), 'Model diverged with loss = NaN'
if step % 10 == 0:
num_examples_per_step = FLAGS.batch_size
examples_per_sec = num_examples_per_step / duration
sec_per_batch = float(duration)
format_str = ('%s: step %d, loss = %.2f (%.1f examples/sec; %.3f '
'sec/batch)')
print (format_str % (datetime.now(), step, loss_value,
examples_per_sec, sec_per_batch))
if step % 100 == 0:
summary_str = sess.run(summary_op)
summary_writer.add_summary(summary_str, step)
# Save the model checkpoint periodically.
if step % 1000 == 0 or (step + 1) == FLAGS.max_steps:
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
def main(argv=None): # pylint: disable=unused-argument
cifar10.maybe_download_and_extract()
if gfile.Exists(FLAGS.train_dir):
gfile.DeleteRecursively(FLAGS.train_dir) # Deletes everything under dirname recursively 遞迴的刪除
gfile.MakeDirs(FLAGS.train_dir)
train()
if __name__ == '__main__':
# tf.app.run()會呼叫main
tf.app.run()
cifar10_eval
"""Evaluation for CIFAR-10.
Accuracy:
cifar10_train.py achieves 83.0% accuracy after 100K steps (256 epochs
of data) as judged by cifar10_eval.py.
Speed:
On a single Tesla K40, cifar10_train.py processes a single batch of 128 images
in 0.25-0.35 sec (i.e. 350 - 600 images /sec). The model reaches ~86%
accuracy after 100K steps in 8 hours of training time.
Usage:
Please see the tutorial and website for how to download the CIFAR-10
data set, compile the program and train the model.
http://tensorflow.org/tutorials/deep_cnn/
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from datetime import datetime
import math
import time
import tensorflow.python.platform
from tensorflow.python.platform import gfile
import numpy as np
import tensorflow as tf
import cifar10
FLAGS = tf.app.flags.FLAGS
tf.app.flags.DEFINE_string('eval_dir', 'cifar10_eval',
"""Directory where to write event logs.""")
tf.app.flags.DEFINE_string('eval_data', 'test',
"""Either 'test' or 'train_eval'.""")
tf.app.flags.DEFINE_string('checkpoint_dir', 'cifar10_train',
"""Directory where to read model checkpoints.""")
tf.app.flags.DEFINE_integer('eval_interval_secs', 60 * 5,
"""How often to run the eval.""")
tf.app.flags.DEFINE_integer('num_examples', 10000,
"""Number of examples to run.""")
tf.app.flags.DEFINE_boolean('run_once', False,
"""Whether to run eval only once.""")
def eval_once(saver, summary_writer, top_k_op, summary_op):
"""Run Eval once.
Args:
saver: Saver.
summary_writer: Summary writer.
top_k_op: Top K op.
summary_op: Summary op.
"""
with tf.Session() as sess:
ckpt = tf.train.get_checkpoint_state(FLAGS.checkpoint_dir)
if ckpt and ckpt.model_checkpoint_path:
# Restores from checkpoint
saver.restore(sess, ckpt.model_checkpoint_path)
# Assuming model_checkpoint_path looks something like:
# /my-favorite-path/cifar10_train/model.ckpt-0,
# extract global_step from it.
global_step = ckpt.model_checkpoint_path.split('/')[-1].split('-')[-1]
else:
print('No checkpoint file found')
return
# Start the queue runners.
# tf.train.Coordinator()來建立一個執行緒管理器(協調器)物件
coord = tf.train.Coordinator()
try:
threads = []
# 所有佇列管理器被預設加入圖的tf.GraphKeys.QUEUE_RUNNERS集合中
for qr in tf.get_collection(tf.GraphKeys.QUEUE_RUNNERS):
threads.extend(qr.create_threads(sess, coord=coord, daemon=True,
start=True))
num_iter = int(math.ceil(FLAGS.num_examples / FLAGS.batch_size))
true_count = 0 # Counts the number of correct predictions.
total_sample_count = num_iter * FLAGS.batch_size
step = 0
while step < num_iter and not coord.should_stop():
predictions = sess.run([top_k_op])
true_count += np.sum(predictions)
step += 1
# Compute precision @ 1.
precision = true_count / total_sample_count
print('%s: precision @ 1 = %.3f' % (datetime.now(), precision))
summary = tf.Summary()
summary.ParseFromString(sess.run(summary_op))
summary.value.add(tag='Precision @ 1', simple_value=precision)
summary_writer.add_summary(summary, global_step)
except Exception as e: # pylint: disable=broad-except
coord.request_stop(e)
coord.request_stop()
coord.join(threads, stop_grace_period_secs=10)
def evaluate():
"""Eval CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
# Get images and labels for CIFAR-10.
eval_data = FLAGS.eval_data == 'test'
images, labels = cifar10.inputs(eval_data=eval_data)
# Build a Graph that computes the logits predictions from the
# inference model.
logits = cifar10.inference(images)
# Calculate predictions.
top_k_op = tf.nn.in_top_k(logits, labels, 1)
# Restore the moving average version of the learned variables for eval.
variable_averages = tf.train.ExponentialMovingAverage(
cifar10.MOVING_AVERAGE_DECAY)
variables_to_restore = variable_averages.variables_to_restore()
saver = tf.train.Saver(variables_to_restore)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
graph_def = tf.get_default_graph().as_graph_def()
summary_writer = tf.summary.FileWriter(FLAGS.eval_dir,
graph_def=graph_def)
while True:
eval_once(saver, summary_writer, top_k_op, summary_op)
if FLAGS.run_once:
break
time.sleep(FLAGS.eval_interval_secs)
def main(argv=None): # pylint: disable=unused-argument
cifar10.maybe_download_and_extract()
if gfile.Exists(FLAGS.eval_dir):
gfile.DeleteRecursively(FLAGS.eval_dir)
gfile.MakeDirs(FLAGS.eval_dir)
evaluate()
if __name__ == '__main__':
tf.app.run()
參考連結:
相關推薦
【Tensorflow】Cifar-10
tensorflow官方CIFAR-10 教程學習筆記 主要包括以下四部分: 環境: python 3.6 tensorflow 1.8 目錄 完整程式碼: 完整程式碼: cifar10 tf.app.flags.DEFINE_xx
【Pytorch】CIFAR-10分類任務
CIFAR-10資料集共有60000張32*32彩色圖片,分為10類,每類有6000張圖片。其中50000張用於訓練,構成5個訓練batch,每一批次10000張圖片,其餘10000張圖片用於測試。 CIFAR-10資料集下載地址:點選下載 資料讀取,這裡選擇下載py
【計算幾何】【DP】【tarjan】Day 10.6
long long pri cout logs 前綴 ret ble freopen style T1 計算幾何+遞推 1 #include <cstdio> 2 #include <cmath> 3 double w,x,r; 4 int
【tensorFlow】tf.reshape()報錯信息 - TypeError: Expected binary or unicode string
bject port cas inpu dimen div nts sof expec 今天在使用tensoflow跑cifar10的CNN分類時候,download一個源碼,但是報錯 TypeError: Failed to convert object of type
【TensorFlow】tf.nn.softmax_cross_entropy_with_logits的用法
white 交叉 none padding tomat ros true const cross 在計算loss的時候,最常見的一句話就是 tf.nn.softmax_cross_entropy_with_logits ,那麽它到底是怎麽做的呢? 首先明確一點,loss是代
【TensorFlow】01 TensorFlow簡介與Python基礎
編譯器 n) The 腳本語言 ble rem 時間 完整 快的 TensorFlow簡介與Python基礎 2018.9.10 一、概述 TF使用數據數據流圖進行數值計算,亮點是支持異構設備分布式計算機 常用的ML庫有MXNet Torch/Pytorch Theano
【TensorFlow】(01)線性回歸
lob 超參數 教育版 ini src ont numpy mat font 特別說明 代碼地址:Github 環境說明 平臺:WIN10(教育版) 環境:Anaconda5.2(Python3.6.6) IDE:Pacharm2018.2.3(專業版) Tensor
機器學習-Logistic迴歸python實踐【3】(10.26更新)
寫在最前面:Logistic迴歸通過Sigmoid函式接受輸入然後進行預測 首先,介紹一下什麼是Sigmoid函式。大家一定聽過海維賽德階躍函式(Heaviside step function),什麼?沒聽過,好吧,換個名字,單位階躍函式,這個認識吧! 這個函式的問題在於該函式
【TensorFlow】Win10+TensorFlow-gpu1.9.0+CUDA9.0+cudnn7.1.4(2018/11/02)
折騰了一天多,終於配置成功了orz 本篇文章是2018年11月2日寫的,Win10,顯示卡為960M 下載版本為:(請注意相容性) Anaconda3 5.3.0 TensorFlow-gpu 1.9.0 CUDA9.0 cu
【TensorFlow】池化層max_pool中兩種paddding操作
max_pool()中padding引數有兩種模式valid和same模式。 Tensorflow的padding和卷積層一樣也有padding操作,兩種不同的操作輸出的結果有區別: 函式原型max_pool(value, ksize, strides, padding
【UE4】 第10講 Matinee相機過場動畫
(版權宣告,禁止轉載) &
【tensorflow】】模型優化(一)指數衰減學習率
指數衰減學習率是先使用較大的學習率來快速得到一個較優的解,然後隨著迭代的繼續,逐步減小學習率,使得模型在訓練後期更加穩定。在訓練神經網路時,需要設定學習率(learning rate)控制引數的更新速度,學習速率設定過小,會極大降低收斂速度,增加訓練時間;學習率太大,可能導致引數在最優解兩側來回振盪
【tensorflow】tf.identity()
常與tf.control_dependencies(self, control_inputs)配合使用,只有當這個上下文管理器中的是一個操作時,control_inputs才會執行。 x = tf.Variable(0.0) x_plus_1 = tf.assign_add(x, 1)
【tensorflow】tf.get_variable()和tf.Variable()的區別
1.tf.Variable() tf.Variable(initial_value, trainable=True, collections=None, validate_shape=True, name=None) ##initial_value為變數的初始值 tf.get
【tensorflow】命令列引數解析
1. tf.app.flags,用於支援接受命令列傳遞引數 import tensorflow as tf #第一個是引數名稱,第二個引數是預設值,第三個是引數描述 tf.app.flags.DEFINE_string('str_name', 'def_v_1',"descrip1")
【Tensorflow】Tensorflow的圖、會話、裝置、變數、核心
前言 基礎知識,前面我們介紹到,Tensorflow的資料流圖是由節點和邊組成的有向無環圖,此外,還涉及一些其他概念,如圖、會話、裝置、變數、核心等。 圖(Graph) import tensorflow as tf # 建立圖 # 建立一個常量運算操作,產生一個1 x 2
【Tensorflow】Tensorflow基礎知識
Tensorflow簡介 \quad\quad 在我們使用以統計方法為核心的機器學習方法的時候,重要的是做特徵工程,
【TensorFlow】多GPU訓練:示例程式碼解析
使用多GPU有助於提升訓練速度和調參效率。 本文主要對tensorflow的示例程式碼進行註釋解析:cifar10_multi_gpu_train.py 1080Ti下加速效果如下(batch=128) 單卡: 兩個GPU比單個GPU加速了近一倍 :
【tensorflow】Object DetectionAPI訓練識別自己的資料集
#一、資料準備 ###1.一個友好的標註工具 各種系統安裝已經再此介紹的很詳細了,linux下可以三行命令解決。 注意:圖片要求是png或者jpg格式 1> . 標註資訊存為xml檔案,使用該指令碼可以將所有的xml檔案轉換為1個csv檔案(自行修改xml路徑
【tensorflow】tensorflow中的全域性變數GLOBAL_VARIABLES及區域性變數LOCAL_VARIABLES
在初學tensorflow的時候,我們會發現在函式體內定義tf.variable()或者tf.get_variable()變數的時候,跟其他語言不同,在tensorflow的函式體內定義的變數並不會隨著函式的執行結束而消失。這是因為tensorflow設定的全域性變數及區域性變數與其他