TensorFlow--AlexNet实现

使用了数据集Oxford-IIIT Pet中的三类猫和狗的数据，猫和狗的数据分别为570多张。

Oxford-IIIT Pet包含 37 种宠物类别的图像数据集，每个类别约有 200 张图像。这些图像在比例、姿势以及光照方面有着丰富的变化。本数据集也可以用于目标检测定位。

#! /usr/bin/python
# -*- coding:utf-8 -*-

#alexnet

import os
import time
import glob
import tensorflow as tf
import numpy as np
from skimage import io,transform

#数据集地址
image_path = '/home/muyangren/projectdir/tensorflow/PetTf/pet_photos_cd/'
#模型保存地址
model_path = '/home/muyangren/projectdir/tensorflow/PetTf/pet_models'
#TensorBorad log
log_dir = '/home/muyangren/projectdir/tensorflow/PetTf/log'
#将所有图片resize成227*227
w = 227
h = 227
c = 3

#读取图片
def read_image(image_path):
	cate = [image_path + x for x in os.listdir(image_path) if os.path.isdir(image_path + x)]
	imgs = []
	labels = []
	for idx, folder in enumerate(cate):
		for im in glob.glob(folder + '/*.jpg'):
			print('reading the image:%s' %(im))
			img = io.imread(im)
			img = transform.resize(img, (w, h))
			imgs.append(img)
			labels.append(idx)
	return np.asarray(imgs, np.float32), np.asarray(labels, np.int32)
data, label=read_image(image_path)

#打乱顺序
num_example = data.shape[0]
arr = np.arange(num_example)
np.random.shuffle(arr)
data = data[arr]
label = label[arr]

#将所有数据分为训练集和验证集
ratio = 0.8
s = np.int(num_example * ratio)
x_train = data[:s]
y_train = label[:s]
x_val = data[s:]
y_val = label[s:]

############构建网络##########
#占位符
x = tf.placeholder(tf.float32, shape = [None, w, h, c], name = 'x')
y_ = tf.placeholder(tf.int32, shape = [None,], name = 'y_')

def inference(input_tensor, train, regularizer):
	with tf.variable_scope("layer1-conv1"):
		conv1_weights = tf.get_variable("weight", [11, 11, 3, 96], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		conv1_biases = tf.get_variable("biases", [96], initializer = tf.constant_initializer(0.0))
		conv1 = tf.nn.conv2d(input_tensor, conv1_weights, strides = [1, 4, 4, 1], padding = "SAME")
		relu1 = tf.nn.relu(tf.nn.bias_add(conv1, conv1_biases))

	with tf.name_scope("layer2-pool1"):
		pool1 = tf.nn.max_pool(relu1, ksize = [1, 3, 3, 1], strides = [1, 2, 2, 1], padding = "VALID")

	with tf.variable_scope("layer3-conv2"):
		conv2_weights = tf.get_variable("weight", [5, 5, 96, 256], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		conv2_biases = tf.get_variable("biases", [256], initializer = tf.constant_initializer(0.0))
		conv2 = tf.nn.conv2d(pool1, conv2_weights, strides = [1, 1, 1, 1], padding = "SAME")
		relu2 = tf.nn.relu(tf.nn.bias_add(conv2, conv2_biases))
	
	with tf.name_scope("layer4-pool2"):
		pool2 = tf.nn.max_pool(relu2, ksize = [1, 3, 3, 1], strides = [1, 2, 2, 1], padding = "VALID")

	with tf.variable_scope("layer5-conv3"):
		conv3_weights = tf.get_variable("weight", [3, 3, 256, 384], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		conv3_biases = tf.get_variable("biases", [384], initializer = tf.constant_initializer(0.0))
		conv3 = tf.nn.conv2d(pool2, conv3_weights, strides = [1, 1, 1, 1], padding = "SAME")
		relu3 = tf.nn.relu(tf.nn.bias_add(conv3, conv3_biases))

	with tf.variable_scope("layer6-conv4"):
		conv4_weights = tf.get_variable("weight", [3, 3, 384, 384], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		conv4_biases = tf.get_variable("biases", [384], initializer = tf.constant_initializer(0.0))
		conv4 = tf.nn.conv2d(conv3, conv4_weights, strides = [1, 1, 1, 1], padding = "SAME")
		relu4 = tf.nn.relu(tf.nn.bias_add(conv4, conv4_biases))

	with tf.variable_scope("layer7-conv5"):
		conv5_weights = tf.get_variable("weight", [3, 3, 384, 256], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		conv5_biases = tf.get_variable("biases", [256], initializer = tf.constant_initializer(0.0))
		conv5 = tf.nn.conv2d(conv4, conv5_weights, strides = [1, 1, 1, 1], padding = "SAME")
		relu5 = tf.nn.relu(tf.nn.bias_add(conv5, conv5_biases))
	
	with tf.name_scope("layer8-pool3"):
		pool3 = tf.nn.max_pool(relu5, ksize = [1, 3, 3, 1], strides = [1, 2, 2, 1], padding = "VALID")
		nodes = 6 * 6 * 256
		reshaped = tf.reshape(pool3, [-1, nodes])

	with tf.variable_scope("layer9-fc1"):
		fc1_weights = tf.get_variable("weight", [nodes, 4096], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		if regularizer != None:
			tf.add_to_collection("losses", regularizer(fc1_weights))
		fc1_biases = tf.get_variable("bias", [4096], initializer = tf.constant_initializer(0.1))
		fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
		if train:
			fc1 = tf.nn.dropout(fc1, 0.5)

	with tf.variable_scope("layer10-fc2"):
		fc2_weights = tf.get_variable("weight", [4096, 4096], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		if regularizer != None:
			tf.add_to_collection("losses", regularizer(fc2_weights))
		fc2_biases = tf.get_variable("bias", [4096], initializer = tf.constant_initializer(0.1))
		fc2 = tf.nn.relu(tf.matmul(fc1, fc2_weights) + fc2_biases)
		if train:
			fc2 = tf.nn.dropout(fc2, 0.5)

	with tf.variable_scope("layer11-fc3"):
		fc3_weights = tf.get_variable("weight", [4096, 2], initializer = tf.truncated_normal_initializer(stddev = 0.1))
		if regularizer != None:
			tf.add_to_collection("losses", regularizer(fc3_weights))
		fc3_biases = tf.get_variable("bias", [2], initializer = tf.constant_initializer(0.1))
		logit = tf.matmul(fc2, fc3_weights) + fc3_biases
	return logit

regularizer = tf.contrib.layers.l2_regularizer(0.0001)
logits = inference(x, False, regularizer)

#将logits乘以1赋值给logits_eval, 定义name
b = tf.constant(value = 1, dtype = tf.float32)
logits_eval = tf.multiply(logits, b, name = "logits_eval")

loss = tf.nn.sparse_softmax_cross_entropy_with_logits(logits = logits, labels = y_)
train_op = tf.train.AdamOptimizer(learning_rate = 0.001).minimize(loss)
correct_prediction = tf.equal(tf.cast(tf.argmax(logits, 1), tf.int32), y_)
acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

train_acc_graph = tf.summary.scalar('train_acc_graph', acc)
#train_loss_graph = tf.summary.scalar('train_loss_graph', loss)

#按批次取数据
def minibatches(inputs = None, targets = None, batch_size = None, shuffle = False):
	assert len(inputs) == len(targets)
	if shuffle:
		indices = np.arange(len(inputs))
		np.random.shuffle(indices)
	for start_idx in range(0, len(inputs) - batch_size + 1, batch_size):
		if shuffle:
			excerpt = indices[start_idx:start_idx + batch_size]
		else:
			excerpt = slice(start_idx, start_idx + batch_size)
		yield inputs[excerpt], targets[excerpt]

def variable_summaries(var):
	with tf.name_scope('summaries'):
		mean = tf.reduce_mean(var)
		tf.summary.scalar('mean', mean)
		with tf.name_scope('stddev'):
			stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))
		tf.summary.scalar('stddev', stddev)
		tf.summary.scalar('max', tf.reduce_max(var))
		tf.summary.scalar('min', tf.reduce_min(var))
		tf.summary.histogram('histogram', var)

#训练和测试数据
n_epoch = 50
batch_size = 16
saver = tf.train.Saver()
sess = tf.Session()
sess.run(tf.global_variables_initializer())

train_writer = tf.summary.FileWriter(log_dir, sess.graph)

for epoch in range(n_epoch):
	start_time = time.time()
	
	#training
	train_loss, train_acc, n_batch = 0, 0, 0
	for x_train_a, y_train_a in minibatches(x_train, y_train, batch_size, shuffle = True):
		_, err, ac, acc_graph = sess.run([train_op, loss, acc, train_acc_graph], feed_dict = {x: x_train_a, y_: y_train_a})
		train_loss += err; train_acc += ac; n_batch +=1
	print("the epoch value is {0}, train loss: {1}".format(epoch, (np.sum(train_loss) / n_batch)))
	print("the epoch value is {0}, train acc: {1}".format(epoch, (np.sum(train_acc) / n_batch)))
	train_writer.add_summary(acc_graph, epoch)

	#validation
	val_loss, val_acc, n_batch = 0, 0, 0
	for x_val_a, y_val_a in minibatches(x_val, y_val, batch_size, shuffle = False):
		err, ac, acc_graph = sess.run([loss, acc, train_acc_graph], feed_dict = {x: x_val_a, y_: y_val_a})
		val_loss += err; val_acc += ac; n_batch += 1
	print("the epoch value is {0}, validation loss: {1}".format(epoch, (np.sum(val_loss) / n_batch)))
	print("the epoch value is {0}, validation acc: {1}".format(epoch, (np.sum(val_acc) / n_batch)))
	train_writer.add_summary(acc_graph, epoch)
saver.save(sess, model_path, )
sess.close()




#! /usr/bin/python
# -*- coding:utf-8 -*-

from skimage import io,transform
import tensorflow as tf
import numpy as np


path1 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/Abyssinian_220.jpg"
path2 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/american_bulldog_217.jpg"
path3 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/basset_hound_194.jpg"
path4 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/beagle_195.jpg"
path5 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/Bengal_194.jpg"
path6 = "/home/muyangren/projectdir/tensorflow/PetTf/test_model_photos/Birman_197.jpg"

#pet_dict = {0:'Abyssinian',1:'american_bulldog',2:'basset_hound',3:'beagle',4:'Bengal',5:'Birman'}
pet_dict = {0:'dog',1:'cat'}

w=227
h=227
c=3

def read_one_image(path):
    img = io.imread(path)
    img = transform.resize(img,(w,h))
    return np.asarray(img)

with tf.Session() as sess:
    data = []
    data1 = read_one_image(path1)
    print(data1)
    data2 = read_one_image(path2)
    data3 = read_one_image(path3)
    data4 = read_one_image(path4)
    data5 = read_one_image(path5)
    data6 = read_one_image(path6)
    data.append(data1)
    data.append(data2)
    data.append(data3)
    data.append(data4)
    data.append(data5)
    data.append(data6)

    saver = tf.train.import_meta_graph('/home/muyangren/projectdir/tensorflow/PetTf/pet_models.meta')
    saver.restore(sess,tf.train.latest_checkpoint('/home/muyangren/projectdir/tensorflow/PetTf/'))

    graph = tf.get_default_graph()
    x = graph.get_tensor_by_name("x:0")
    feed_dict = {x:data}

    logits = graph.get_tensor_by_name("logits_eval:0")

    classification_result = sess.run(logits,feed_dict)

    #打印出预测矩阵
    print(classification_result)
    #打印出预测矩阵每一行最大值的索引
    print(tf.argmax(classification_result,1).eval())
    #根据索引通过字典对应分类
    output = []
    output = tf.argmax(classification_result,1).eval()
    for i in range(len(output)):
        print("the ",i+1," pet is "+pet_dict[output[i]])