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.
"""
images, labels = cifar10_input.distorted_inputs(
batch_size=FLAGS.batch_size)
if FLAGS.use_fp16:
images = tf.cast(images, tf.float16)
labels = tf.cast(labels, tf.float16)
return images, labels
def build_model():
print('build model')
g = tf.Graph()
with g.as_default(), tf.device(
tf.train.replica_device_setter(FLAGS.ps_tasks)):
# inputs, labels = imagenet_input(is_training=True)
inputs, labels = cifar10_input.distorted_inputs(
'../cifar10/cifar-10-batches-bin', 100)
with slim.arg_scope(
mobilenet_v1.mobilenet_v1_arg_scope(is_training=True)):
print('default layers')
logits, _ = mobilenet_v1.mobilenet_v1(
inputs,
is_training=True,
depth_multiplier=FLAGS.depth_multiplier,
num_classes=FLAGS.num_classes,
)
print('start train')
one_hot_labels = tf.one_hot(tf.cast(labels, dtype=tf.uint8),
FLAGS.num_classes,
dtype=tf.float32)
slim.losses.softmax_cross_entropy(one_hot_labels, logits)
if FLAGS.quantize:
tf.contrib.quantize.create_training_graph(
quant_delay=get_quant_delay())
total_loss = tf.losses.get_total_loss(name='total_loss')
# configure the learning rate using an exponential decay
num_epochs_per_decay = 2.5
imagenet_size = 1271167
decay_steps = int(imagenet_size / FLAGS.batch_size *
num_epochs_per_decay)
learning_rate = tf.train.exponential_decay(
get_learning_rate(),
tf.train.get_or_create_global_step(),
decay_steps,
_LEARNING_RATE_DECAY_FACTOR,
staircase=True)
opt = tf.train.GradientDescentOptimizer(learning_rate)
train_tensor = slim.learning.create_train_op(total_loss, optimizer=opt)
slim.summaries.add_scalar_summary(total_loss, 'total_loss', 'losses')
slim.summaries.add_scalar_summary(learning_rate, 'learning_rate', 'traing')
return g, train_tensor
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')
data_dir = os.path.join(FLAGS.data_dir, 'cifar-10-batches-bin')
images, labels = cifar10_input.distorted_inputs(
data_dir=data_dir, batch_size=FLAGS.batch_size)
if FLAGS.use_fp16:
images = tf.cast(images, tf.float16)
labels = tf.cast(labels, tf.float16)
return images, labels
import tensorflow as tf
from cifar10 import cifar10_input, cifar10
import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
batch_size = 100
# data_dir='D:\\学习笔记\\ai\\dataSets\\cifar10\\cifar-10-binary.tar'
data_dir = 'D:\\学习笔记\\ai\\dataSets\\cifar10\\cifar-10-batches-bin'
# data_dir='D:\\学习笔记\\ai\\dataSets\\cifar10'
cifar10.maybe_download_and_extract()
images_train, label_train = cifar10_input.distorted_inputs(
data_dir=data_dir, batch_size=batch_size)
images_test, label_test = cifar10_input.inputs(eval_data=True,
data_dir=data_dir,
batch_size=batch_size)
X = tf.placeholder(shape=images_train.get_shape(), dtype=tf.float32)
y = tf.placeholder(shape=label_train.get_shape(), dtype=tf.int32)
w_cnn = tf.Variable(tf.truncated_normal(shape=(4, 4, 3, 24), stddev=0.1))
b_cnn = tf.Variable(tf.ones(shape=24))
cnn = tf.nn.conv2d(X, w_cnn, strides=(1, 1, 1, 1), padding='SAME', name='cnn')
cnn = tf.reshape(cnn, shape=(batch_size, -1))
cnn_shape = cnn.get_shape()
relu_cnn = tf.nn.relu(cnn)
print('cnn shape', cnn_shape)
w_fc = tf.Variable(tf.truncated_normal(shape=(13824, 10), stddev=0.1))
b_fc = tf.Variable(tf.ones(shape=10))
cross_entropy_mean = tf.reduce_mean(tf.reduce_sum(cross_entropy))
tf.add_to_collection("losses", cross_entropy_mean)
return tf.add_n(tf.get_collection("losses"), name="total_loss")
if __name__ == "__main__":
#设置最大迭代次数
max_steps = 10000
#设置每次训练的数据大小
batch_size = 128
#下载解压数据
cifar10.maybe_download_and_extract()
# 设置数据的存放目录
cifar10_dir = "D:/dataset/cifar10_data/cifar-10-batches-bin"
#获取数据增强后的训练集数据
images_train, labels_train = cifar10_input.distorted_inputs(
cifar10_dir, batch_size)
#获取裁剪后的测试数据
images_test, labels_test = cifar10_input.inputs(eval_data=True,
data_dir=cifar10_dir,
batch_size=batch_size)
#定义模型的输入和输出数据
image_holder = tf.placeholder(dtype=tf.float32,
shape=[batch_size, 24, 24, 3])
label_holder = tf.placeholder(dtype=tf.int32, shape=[batch_size])
#设计第一层卷积
weight1 = variable_with_weight_loss(shape=[5, 5, 3, 64], std=5e-2, w1=0)
kernel1 = tf.nn.conv2d(image_holder, weight1, [1, 1, 1, 1], padding="SAME")
bais1 = tf.Variable(tf.constant(0.0, dtype=tf.float32, shape=[64]))
conv1 = tf.nn.relu(tf.nn.bias_add(kernel1, bais1))
pool1 = tf.nn.max_pool(conv1, [1, 3, 3, 1], [1, 2, 2, 1], padding="SAME")
def variable_with_weight_loss(shape, stddev, wl):
var = tf.Variable(tf.truncated_normal(shape, stddev=stddev))
if wl is not None:
weight_loss = tf.multiply(tf.nn.l2_loss(var), wl, name='weight_loss')
tf.add_to_collection('losses', weight_loss)
return var
# 下载数据集到指定位置
cifar10.maybe_download_and_extract()
# 提取训练集和测试集
# distort_inputs函数包含了一些数据增强操作
images_train, labels_train = cifar10_input.distorted_inputs(eval_data=True, data_dir=data_dir, batch_size=batch_size)
# 测试集数据正常读取，不做增强处理
images_test, labels_test = cifar10_input.inputs(eval_data=True, data_dir=data_dir, batch_size=batch_size)
# 设置输入数据承接
image_holder = tf.placeholder(tf.float32, [batch_size, 24, 24, 3])
label_holder = tf.placeholder(tf.float32, [batch_size])
# 第一层卷积操作
# 初始化卷积核，第一层卷积核不进行L2正则化，大小为5*5，3通道，64个卷积核，标准差为0.05
weight1 = variable_with_weight_loss(shape=[5, 5, 3, 64], stddev=5e-2, wl=0.0)
# 第一次卷积两个步长都为1，padding值为SAME
kernel1 = tf.nn.conv2d(image_holder, weight1, [1, 1, 1, 1], padding='SAME')
# 偏值为64个，初始化为0
bias1 = tf.Variable(tf.constant(0.0, shape=[64]))
# 激活函数进行非线性化
batch_size = 128
data_dir = 'F:\project\MachineLearning\DeepLearning\Data\cifar10'
def variable_with_weight_loss(shape, stddev, w1):
var = tf.Variable(tf.truncated_normal(shape, stddev=stddev))
if w1 is not None:
weight_loss = tf.multiply(tf.nn.l2_loss(var), w1, name='weight_loss')
tf.add_to_collection('losses', weight_loss)
return var
maybe_download_and_extract() # 下载解压数据
# 使用distorted_inputs 产生训练需要的数据，包括特征及其对应的label, 该方法做了数据增强
images_train, labels_train = distorted_inputs(data_dir=data_dir,
batch_size=batch_size)
# cifar10_input.inputs() 生成测试数据
image_test, labels_test = inputs(eval_data=True,
data_dir=data_dir,
batch_size=batch_size)
image_holder = tf.placeholder(tf.float32,
[batch_size, 24, 24, 3]) # 3表示有RGB三种通道
label_holder = tf.placeholder(tf.int32, [batch_size])
# 创建卷积层，注意LRN的理解
# 使用 5x5的卷积核大小，3个颜色通道，64个卷积核
# ==============
# 卷积层对数据进行特征提取， 全连接层进行组合匹配
def debug():
g = tf.Graph()
inputs = tf.placeholder(shape=[None, 28, 28, 3], dtype=tf.float32)
labels = tf.placeholder(shape=[
None,
], dtype=tf.float32)
with g.as_default(), tf.device(
tf.train.replica_device_setter(FLAGS.ps_tasks)):
# inputs, labels = imagenet_input(is_training=True)
tf_inputs, tf_labels = cifar10_input.distorted_inputs(
'../cifar10/cifar-10-batches-bin', 100)
with slim.arg_scope(
mobilenet_v1.mobilenet_v1_arg_scope(is_training=True)):
logits, _ = mobilenet_v1.mobilenet_v1(
inputs,
is_training=True,
depth_multiplier=FLAGS.depth_multiplier,
num_classes=FLAGS.num_classes,
)
one_hot_labels = tf.one_hot(tf.cast(labels, dtype=tf.uint8),
FLAGS.num_classes,
dtype=tf.float32)
slim.losses.softmax_cross_entropy(one_hot_labels, logits)
if FLAGS.quantize:
tf.contrib.quantize.create_training_graph(
quant_delay=get_quant_delay())
total_loss = tf.losses.get_total_loss(name='total_loss')
# configure the learning rate using an exponential decay
num_epochs_per_decay = 2.5
imagenet_size = 1271167
decay_steps = int(imagenet_size / FLAGS.batch_size *
num_epochs_per_decay)
learning_rate = tf.train.exponential_decay(
get_learning_rate(),
tf.train.get_or_create_global_step(),
decay_steps,
_LEARNING_RATE_DECAY_FACTOR,
staircase=True)
opt = tf.train.GradientDescentOptimizer(learning_rate)
train_tensor = slim.learning.create_train_op(total_loss, optimizer=opt)
slim.summaries.add_scalar_summary(total_loss, 'total_loss', 'losses')
slim.summaries.add_scalar_summary(learning_rate, 'learning_rate', 'traing')
init_op = [
tf.global_variables_initializer(),
tf.local_variables_initializer()
]
# accuracy
test_inputs, test_labels = cifar10_input.distorted_inputs(
'../cifar10/cifar-10-batches-bin', 1000, is_validdata=True)
accuracy = tf.metrics.accuracy(tf.argmax(logits, axis=1), test_labels)
with tf.Session(graph=g) as session:
session.run(init_op)
for step in range(10000):
_loss, _ = session.run([total_loss, train_tensor],
feed_dict={
inputs: tf_inputs,
labels: tf_labels,
})
if step % 100 == 0:
_accuracy = session.run(accuracy,
feed_dict={
inputs: test_inputs,
labels: test_labels
})
print('step: %d, loss: %s, accuracy: %s' %
(step, _loss, _accuracy))
