一、迁移学习的应用价值
迁移学习的应用
图像识别：图像增强，风格转移，对象检测，皮肤癌检测
文字识别：Zero Shot翻译、情绪分类
应用价值
复用现有知识数据，已有的大量工作不至于完全丢弃
不需要在话费巨大代价重新采集和标定庞大数据（也可能无法获取数据）
对于快速出现的新领域，能够快速迁移和应用，体现时效性优势
二、Tensorflow 迁移学习
import glob
import os.path
import random
import numpy as np
import tensorflow as tf
from
2.1参数设置
# 瓶颈点的节点个数
BOTTLENECK_TENSOR_SIZE = 200
# 瓶颈点的张量名称
BOTTLENECK_TENSOR_NAME = 'pool_3/reshape:0'
# 图像输入张量所对应的名称
JPEG_DATA_TENSOR_NAME = 'DecodeJpeg/contents:0'
# model info
MODEL_DIR = './model/'
MODEL_FILE = 'tensorflow_inception_gragh.pb'
# 缓存文件夹
CACHE_DIR = './tmp/bottleneck/'
# 图像输入文件夹
INPUT_DATA = './flower_data/'
# 验证和测试数据百分比
VALIDATION_PRECENTAGE = 10
TEST_PRECENTAGE = 10
# 定义神经网络参数
LEARNING_DATA = 0.01
STEPS = 4000
BATCH = 100
2.2 数据集拆分
def cteate_image_lists(testing_percentage,validation_percentage):
result = {} #得到的结果
sub_dir = [x[0] for x in os.walk(INPUT_DATA)]
is_root_dir = True
for sub_dir in sub_dirs:
if is_root_dir:
is_root_dir = False
continue
#获取有效图片
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)
filr_list.extend(glob.glob(file_glob))
if not file_list:
continue
#类别名
label_name = dir_name.lower()
training_images = []
testing_images = []
validation_images = []
for file_name in file_list:
base_name = os.path.basename(file_name)
#随机拆分数据集
chance = np.random.randin(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)
#把结果放在字典中
result[label_name] = {
'dir': dir_name,
'training':training_images,
'testing':testing_images,
'validation':validation_images
}
return
2.3 获取图像路径
# index图片的编号
def get_image_path(image_lists,image_dir,label_name,index,category):
label_lists = image_lists[label_name]
category_list = label_lists[category]
mod_index = index % len(category_list)
base_name = category_list[mod_index]
sub_dir = label_lists('dir')
#full_path
full_path = os.path.join(images_dir,sub_dir,base_name)
return
#get bottleneck path
def get_bottlenect_path(images_dir,label_name,index,category)
return get_image_path(image_lists,CACHE_DIR,label_name,index,category)+'.txt'
2.4 处理图片获取特征向量
def run_bottlenect_on_image(sess,image_data,image_data_tensor,bottleneck_tensor):
bottleneck_values = sess.run(bottleneck_tensor,
feed_dict={image_data_tensor:image_data})
bottleneck_values = np.squeeze(bottleneck_values)
return bottleneck_values
def get_or_create_bottleneck(sess,image_lists,label_name,index,
category,jpeg_data_tensor,bottleneck_tensor):
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_bottlenect_path(image_lists,label_name,index,category)
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()
bottleneck_values = run_bottlenect_on_image(sess,image_data,jpeg_data_tensor,
bottleneck_tensor)
bottleneck_string = ','.join(str(x) for x in bottleneck_values)
with open(bottleneck_path,'w') as bottleneck_file:
bottleneck_file.write(bottleneck_string)
else:
with open(bottleneck_path,'r') as bottleneck_file:
bottleneck_string = bottleneck_file.read()
bottleneck_values = [float(x) for x in bottleneck_string.split(',')]
return
2.5 数据集的获取
def get_random_cache_bottlenecks(sess,n_classes,image_lists,how_many,category,
jpeg_data_tensor,bottleneck_tensor):
bottlenecks = []
ground_truths = []
for i in range(how_many):
label_index = random.randrange(n_classes)
label_name = list(image_lists.keys())[label_index]
image_index = random.randrange(65536)
bottleneck = get_or_create_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
def get_test_bottlenecks(sess,n_classes,image_lists,jpeg_data_tensor,bottleneck_tensor):
bottlenecks = []
ground_truths = []
label_name_list = list(image_lists.keys())
for label_index,label_name in enumerate(label_name_list):
category = 'testing'
for index,unused_base_name in enumerate(image_lists,[label_name],[category]):
bottleneck = get_or_create_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
2.6 主函数的定义
def main(_):
#读取所有图片
image_lists = create_image_lists(TEST_PRECENTAGE,VALIDATION_PRECENTAGE)
n_classes = len(image_lists.keys())
with gfile.FastGFile(os.path.join(MODEL_DIR,MODEL_FILE),'rb') as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
bottleneck_tensor,jpeg_data_tensor = tf.import_graph_def(graph_def,
return_elements=[BOTTLENECK_TENSOR_NAME,
JPEG_DATA_TENSOR_NAME])
bottleneck_input = tf.placeholder(tf.float32,[BOTTLENECK_TENSOR_SIZE],
name='BottleneckInputPlaceholder')
ground_truth_input = tf.placeholder(tf.float32,[None,n_classes],
name='GroundTruthInput')
#定义全连接层
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]) + 0.001)
logits = tf.matmul(bottleneck_input,weights) + biases #y = wx + b
final_tensor = tf.nn.softmax(logits)
#loss交叉熵
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_DATA).\
minimize(cross_entropy_mean)
#accuracy
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:
sess.run(tf.global_variables_initializer())
for i in range(STEPS):
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})
#在训练集验证
if i%100 ==0 or i+1 ==STEPS:
validation_bottlenecks,validation_ground_truth = get_random_cached_bottlenecks(sess,n_classes,
image_lists,BATCH,"training"
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{}valiation accuracy on random sampled{}example = {}%'.format(i,BATCH,valuation_accuracy*100))
#在测试集验证
test_bottlenecks,test_ground_truth = get_test_bottlenecks(sess,n_classes,
image_lists,BATCH,
jpeg_data_tensor,
bottleneck_tensor)
test_accuracy = sess.run(evaluation_step,
feed_dict={bottleneck_input:test_bottlenecks,
ground_truth_input:test_ground_truth})
print('Fanal test accuracy ={}%'.format(test_accuracy*100))
if __name__ = "__main__"
