迁移学习就是将一个问题上训练好的模型通过简单的调整使其适用于一个新的问题
使用数据集:http://download.tensorflow.org/example_images/flower_photos.tgz
使用Google提供的训练好的模型Inception-v3:http://storage.googleapis.com/download.tensorflow.org/models/inception_dec_2015.zip
解压缩
import glob
import os.path
import random
import random
import numpy as np
import tensorflow as tf
from tensorflow.python.platform import gfile
#瓶颈层的节点数
BOTTLENECK_TENSOR_SIZE =2048
#代表瓶颈层结果的张量名称
BOTTLENECK_TENSOR_NAME = 'pool_3/_reshape:0'
JPEG_DATA_TENSOR_NAME = 'DecodeJpeg/contents:0'
#google训练好的Inception-v3模型文件目录和文件名
MODEL_DIR = r'E:\data\inception_dec_2015'
MODEL_FILE = 'tensorflow_inception_graph.pb'
#训练后得到的特征向量保存地址
CACHE_DIR = r'E:\test\transfer_learning\bottleneck'
#图片数据文件夹
INPUT_DATA = r'E:\data\flower_photos\flower_photos'
#验证数据百分比
VALIDATION_PERCENTAGE = 10
#测试数据百分比
TEST_PERCENTAGE = 10
LEARNING_RATE = 0.01
STEPS = 4000
BATCH = 100
#从数据文件夹读取所有的图片列表并按训练、验证、测试数据分开
def creat_image_lists(testing_percentage, validation_percentage):
result = {} #所得的图片都存在result这个字典里
sub_dirs = [x[0] for x in os.walk(INPUT_DATA)] #获取当前目录下所有的子目录
#第一个目录是当前目录,不考虑
is_root_dir = True
for sub_dir in sub_dirs:
if is_root_dir:
is_root_dir = False
continue
#获取当前目录下所有的有效图片文件
extensions = ['jpg', 'jpeg', 'JPG', 'JPEG']
file_list = []
dir_name = os.path.basename(sub_dir)
for extension in extensions:
file_glob = os.path.join(INPUT_DATA, dir_name, '*.'+extension)
file_list.extend(glob.glob(file_glob))
if not file_list: continue
#通过目录名获取类别的名称
label_name = dir_name.lower()
#初始化
training_images = []
testing_images = []
validation_images = []
for file_name in file_list:
base_name = os.path.basename(file_name)
#随机将数据分到数据集、测试集和验证集
chance = np.random.randint(100)
if chance < validation_percentage:
validation_images.append(base_name)
elif chance < (testing_percentage + validation_percentage):
testing_images.append(base_name)
else:
training_images.append(base_name)
#将当前类别的数放如结果字典
result[label_name]={
'dir':dir_name,
'training':training_images,
'testing':testing_images,
'validation':validation_images,
}
return result
#通过类别名称、所属数据集和图片编号获取一张图片的地址
#image_lists给出所有图片信息
#image_dir根目录
#index需要获取的图片编号
#category指定需要获取的图片是在训练数据集、测试数据集还是验证数据集
def get_image_path(image_lists, image_dir, label_name, index, category):
label_lists = image_lists[label_name] #获取给定类别中所有图片
#根据所属数据集的名称获取集合中的全部图片信息
category_list = label_lists[category]
mod_index = index % len(category_list)
#获取图片文件名
base_name = category_list[mod_index]
sub_dir = label_lists['dir']
#最终的地址为数据根目录的地址加上类别的文件夹加上图片的名称
full_path = os.path.join(image_dir, sub_dir, base_name)
return full_path
#通过类别名称、所属数据集和图片编号获取经过Inception-v3模型处理之后的特征向量文件地址
def get_bottleneck_path(image_lists, label_name, index, category):
return get_image_path(image_lists, CACHE_DIR, label_name, index, category) + '.txt'
#使用加载的训练好的Inception-v3模型处理一张图片,得到这个图片的特征向量
def run_bottleneck_on_image(sess, image_data, image_data_tensor, bottleneck_tensor):
#将当前图片作为输入计算瓶颈张量的值,即为这张图片的新的特征向量
bottleneck_values = sess.run(bottleneck_tensor, {image_data_tensor:image_data})
#经过卷积神经网路处理的结果为一个四维数组,将这个结果压缩成一个特征向量
bottleneck_values = np.squeeze(bottleneck_values)
return bottleneck_values
#获取一张图片经过Inception-v3模型处理之后的特征向量,先寻找已经计算且保存下来的特征向量,若找不到则保存到文件
def get_or_creat_bottleneck(sess, image_lists, label_name, index, category, jpeg_data_tensor, bottleneck_tensor):
#获取一张图片对应的特征向量文件的路径
label_lists = image_lists[label_name]
sub_dir = label_lists['dir']
sub_dir_path = os.path.join(CACHE_DIR, sub_dir)
if not os.path.exists(sub_dir_path): os.makedirs(sub_dir_path)
bottleneck_path = get_bottleneck_path(image_lists, label_name, index, category)
#如果这个向量文件不存在,通过Inception-v3模型来计算特征向量,并将计算的结果存入文件
if not os.path.exists(bottleneck_path):
image_path = get_image_path(image_lists, INPUT_DATA, label_name, index, category)
image_data = gfile.FastGFile(image_path, 'rb').read()
#计算特征向量
bottleneck_values = run_bottleneck_on_image(sess, image_data, jpeg_data_tensor, bottleneck_tensor)
#将计算结果存入文件
bottleneck_string = ','.join(str(x) for x in bottleneck_values)
with open(bottleneck_path, 'w') as bottleneck_file:
bottleneck_file.write(bottleneck_string)
else:
#直接从文件中获取图片相应的向量
with open(bottleneck_path, 'r') as bottleneck_file:
bottleneck_string = bottleneck_file.read()
bottleneck_values = [float(x) for x in bottleneck_string.split(',')]
return bottleneck_values
#随机获取一个batch的图片作为训练数据
def get_random_cached_bottlenecks(sess, n_classes, image_lists, how_many, category, jpeg_data_tensor, bottleneck_tensor):
bottlenecks = []
ground_truths = []
for _ in range(how_many):
#随机一个类别和图片的编号加入当前的训练数据
label_index = random.randrange(n_classes)
label_name = list(image_lists.keys())[label_index]
image_index = random.randrange(65536)
bottleneck = get_or_creat_bottleneck(sess, image_lists, label_name, image_index, category, jpeg_data_tensor, bottleneck_tensor)
ground_truth = np.zeros(n_classes, dtype=np.float32)
ground_truth[label_index] = 1.0
bottlenecks.append(bottleneck)
ground_truths.append(ground_truth)
return bottlenecks, ground_truths
#获取全部的测试数据。在最终测试的时候需要在所有的测试数据上计算正确率
def get_test_bottlenecks(sess, image_lists, n_classes, jpeg_data_tensor, bottleneck_tensor):
bottlenecks = []
ground_truths = []
label_name_list = list(image_lists.keys())
#枚举所有的类别和每个类别中的测试图片
for label_index, label_name in enumerate (label_name_list):
category = 'testing'
for index, unused_base_name in enumerate(image_lists[label_name][category]):
#计算对应的特征向量,将其加入最终数据的列表
bottleneck = get_or_creat_bottleneck(sess, image_lists, label_name, index, category, jpeg_data_tensor, bottleneck_tensor)
ground_truth = np.zeros(n_classes, dtype=np.float32)
ground_truth[label_index] = 1.0
bottlenecks.append(bottleneck)
ground_truths.append(ground_truth)
return bottlenecks, ground_truths
def main(_):
#读取所有图片
image_lists = creat_image_lists(TEST_PERCENTAGE, VALIDATION_PERCENTAGE)
n_classes = len(image_lists.keys())
print(n_classes)
#读取已训练好的Inception-v3模型
with gfile.FastGFile(os.path.join(MODEL_DIR, MODEL_FILE), 'rb') as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
#加载模型,返回数据输入所对应的张量以及计算瓶颈层结果所对应的张量
bottleneck_tensor, jpeg_data_tensor = tf.import_graph_def(graph_def, return_elements=[BOTTLENECK_TENSOR_NAME, JPEG_DATA_TENSOR_NAME])
#特征提取
bottleneck_input = tf.placeholder(tf.float32, [None, BOTTLENECK_TENSOR_SIZE], name='BottleneckInputPlaceholder')
#定义新的标准答案输入
ground_truth_input = tf.placeholder(tf.float32, [None, n_classes], name='GroundTruthInput')
#定义一层全连接层来解决新的图片分类问题
with tf.name_scope('final_training_ops'):
weights = tf.Variable(tf.truncated_normal([BOTTLENECK_TENSOR_SIZE, n_classes], stddev=0.001))
biases = tf.Variable(tf.zeros([n_classes]))
logits = tf.matmul(bottleneck_input, weights) + biases
final_tensor = tf.nn.softmax(logits)
#定义交叉熵损失函数
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=ground_truth_input)
cross_entropy_mean = tf.reduce_mean(cross_entropy)
train_step = tf.train.GradientDescentOptimizer(LEARNING_RATE).minimize(cross_entropy_mean)
#计算正确率
with tf.name_scope('evaluation'):
correct_prediction = tf.equal(tf.argmax(final_tensor,1), tf.argmax(ground_truth_input,1))
evaluation_step = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
with tf.Session() as sess:
init = tf.global_variables_initializer()
sess.run(init)
#训练过程
for i in range(STEPS):
#每次获取一个batch的训练数据集
train_bottlenecks, train_ground_truth = get_random_cached_bottlenecks(
sess, n_classes, image_lists, BATCH, 'training', jpeg_data_tensor, bottleneck_tensor
)
sess.run(train_step, feed_dict={bottleneck_input:train_bottlenecks, ground_truth_input: train_ground_truth})
#在验证数据上测试正确率
if i % 100 == 0 or i+1 ==STEPS:
validation_bottlenecks, validation_ground_truth = get_random_cached_bottlenecks(
sess, n_classes, image_lists, BATCH, 'validation', jpeg_data_tensor, bottleneck_tensor
)
validation_accuracy = sess.run(
evaluation_step, feed_dict={bottleneck_input:validation_bottlenecks, ground_truth_input:validation_ground_truth}
)
print('Step %d: Validation accuracy on random sampled' '%d example = %.1f%%' % (i, BATCH, validation_accuracy*100))
#在最后的测试数据上测试正确率
test_bottlenecks, test_ground_truth = get_test_bottlenecks(
sess, image_lists, n_classes, jpeg_data_tensor, bottleneck_tensor
)
test_accuracy = sess.run(evaluation_step, feed_dict={
bottleneck_input:test_bottlenecks, ground_truth_input: test_ground_truth
})
print('Final test accuracy = %.1f%%' % (test_accuracy*100))
if __name__ == '__main__':
tf.app.run()