【Inception-v3模型】迁移学习 实战训练 花朵种类识别

参考博客：【TensorFlow】迁移学习（使用Inception-v3），非常感谢这个博主的这篇博客，我这篇博客的框架来自于这位博主，然后我针对评论区的问题以及自己的实践增加了一些内容以及解答。

github：代码

知识储备

• 迁移学习是将一个数据集上训练好的网络模型快速转移到另外一个数据集上，可以保留训练好的模型中倒数第一层之前的所有参数，替换最后一层即可，在最后层之前的网络层称之为瓶颈层。
• 迁移学习，首先尝试了Inception-V3，直接使用pool_3层的输出，接上一个全连接的分类层，使用softmax进行分类，使用Inception-V3的默认输入。

一、准备工作

1、数据集下载
2、Inception-v3模型下载

• 官方下载地址
  【科学上网】

• 百度网盘
  提取码：zmrl

• 数据集解压后的目录：

flower_photos/
    daisy/
    dandelion/
    roses/
    sunflowers/
    tulips/

数据集文件夹包含5个子文件，每一个子文件夹的名称为一种花的名称，代表了不同的类别。平均每一种花有734张图片，每一张图片都是RGB色彩模式，大小也不相同，程序将直接处理没有整理过的图像数据。

• 模型解压后的目录：

imagenet_comp_graph_label_strings.txt
tensorflow_inception_graph.pb

3、目录结构

• 需要自行创建transfer-learning/data/tmp/bottlenec/model/train.py/eval.py文件。

transfer-learning/
    data/
        flower_photos/
            ......
        tmp/
            bottleneck/
                ......
    model/
        imagenet_comp_graph_label_strings.txt
        tensorflow_inception_graph.pb
    train.py
    eval.py

二、代码实现

需要写两个文件：1、train.py 2、eval.py

1、train.py

python3 train.py

import glob
import os.path
import random
import numpy as np
import tensorflow as tf
from tensorflow.python.platform import gfile

# 数据参数
MODEL_DIR = 'model/'  # inception-v3模型的文件夹
MODEL_FILE = 'tensorflow_inception_graph.pb'  # inception-v3模型文件名
CACHE_DIR = 'data/tmp/bottleneck'  # 图像的特征向量保存地址
INPUT_DATA = 'data/flower_photos'  # 图片数据文件夹
VALIDATION_PERCENTAGE = 10  # 验证数据的百分比
TEST_PERCENTAGE = 10  # 测试数据的百分比

# inception-v3模型参数
BOTTLENECK_TENSOR_SIZE = 2048  # inception-v3模型瓶颈层的节点个数
BOTTLENECK_TENSOR_NAME = 'pool_3/_reshape:0'  # inception-v3模型中代表瓶颈层结果的张量名称
JPEG_DATA_TENSOR_NAME = 'DecodeJpeg/contents:0'  # 图像输入张量对应的名称

# 神经网络的训练参数
LEARNING_RATE = 0.01
STEPS = 1000
BATCH = 100
CHECKPOINT_EVERY = 100
NUM_CHECKPOINTS = 5


# 从数据文件夹中读取所有的图片列表并按训练、验证、测试分开
def create_image_lists(validation_percentage, test_percentage):
    result = {}  # 保存所有图像。key为类别名称。value也是字典，存储了所有的图片名称
    sub_dirs = [x[0] for x in os.walk(INPUT_DATA)]  # 获取所有子目录
    is_root_dir = True  # 第一个目录为当前目录，需要忽略

    # 分别对每个子目录进行操作
    for sub_dir in sub_dirs:
        if is_root_dir:
            is_root_dir = False
            continue

        # 获取当前目录下的所有有效图片
        extensions = {'jpg', 'jpeg', 'JPG', 'JPEG'}
        file_list = []  # 存储所有图像
        dir_name = os.path.basename(sub_dir)  # 获取路径的最后一个目录名字
        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

        # 将当前类别的图片随机分为训练数据集、测试数据集、验证数据集
        label_name = dir_name.lower()  # 通过目录名获取类别的名称
        training_images = []
        testing_images = []
        validation_images = []
        for file_name in file_list:
            base_name = os.path.basename(file_name)  # 获取该图片的名称
            chance = np.random.randint(100)  # 随机产生100个数代表百分比
            if chance < validation_percentage:
                validation_images.append(base_name)
            elif chance < (validation_percentage + test_percentage):
                testing_images.append(base_name)
            else:
                training_images.append(base_name)

        # 将当前类别的数据集放入结果字典
        result[label_name] = {
            'dir': dir_name,
            'training': training_images,
            'testing': testing_images,
            'validation': validation_images
        }

    # 返回整理好的所有数据
    return result


# 通过类别名称、所属数据集、图片编号获取一张图片的地址
def get_image_path(image_lists, image_dir, label_name, index, category):
    label_lists = image_lists[label_name]  # 获取给定类别中的所有图片
    category_list = label_lists[category]  # 根据所属数据集的名称获取该集合中的全部图片
    mod_index = index % len(category_list)  # 规范图片的索引
    base_name = category_list[mod_index]  # 获取图片的文件名
    sub_dir = label_lists['dir']  # 获取当前类别的目录名
    full_path = os.path.join(image_dir, sub_dir, base_name)  # 图片的绝对路径
    return full_path


# 通过类别名称、所属数据集、图片编号获取特征向量值的地址
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inception-v3处理图片获取特征向量
def run_bottleneck_on_image(sess, image_data, image_data_tensor,
                            bottleneck_tensor):
    bottleneck_values = sess.run(bottleneck_tensor,
                                 {image_data_tensor: image_data})
    bottleneck_values = np.squeeze(bottleneck_values)  # 将四维数组压缩成一维数组
    return bottleneck_values


# 获取一张图片经过inception-v3模型处理后的特征向量
def get_or_create_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)

    # 如果该特征向量文件不存在，则通过inception-v3模型计算并保存
    if not os.path.exists(bottleneck_path):
        image_path = get_image_path(image_lists, INPUT_DATA, label_name, index,
                                    category)  # 获取图片原始路径
        image_data = gfile.FastGFile(image_path, 'rb').read()  # 获取图片内容
        bottleneck_values = run_bottleneck_on_image(
            sess, image_data, jpeg_data_tensor,
            bottleneck_tensor)  # 通过inception-v3计算特征向量

        # 将特征向量存入文件
        bottleneck_string = ','.join(str(x) for x in bottleneck_values)
        with open(bottleneck_path, 'w') as bottleneck_file:
            bottleneck_file.write(bottleneck_string)
    else:
        # 否则直接从文件中获取图片的特征向量
        with open(bottleneck_path, 'r') as bottleneck_file:
            bottleneck_string = bottleneck_file.read()
        bottleneck_values = [float(x) for x in bottleneck_string.split(',')]

    # 返回得到的特征向量
    return bottleneck_values


# 随机获取一个batch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(65535)
        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


# 获取全部的测试数据
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]):
            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 bottlenecks, ground_truths


def main(_):
    # 读取所有的图片
    image_lists = create_image_lists(VALIDATION_PERCENTAGE, TEST_PERCENTAGE)
    n_classes = len(image_lists.keys())

    with tf.Graph().as_default() as graph:
        # 读取训练好的inception-v3模型
        with tf.gfile.GFile(os.path.join(MODEL_DIR, MODEL_FILE), 'rb') as f:
            graph_def = tf.GraphDef()
            graph_def.ParseFromString(f.read())
            # 加载inception-v3模型，并返回数据输入张量和瓶颈层输出张量
            bottleneck_tensor, jpeg_data_tensor = tf.import_graph_def(
                graph_def,
                return_elements=[
                    BOTTLENECK_TENSOR_NAME, JPEG_DATA_TENSOR_NAME
                ])

        # 定义新的神经网络输入
        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=0.1))
            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_v2(
            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))

    # 训练过程
    with tf.Session(graph=graph) as sess:
        init = tf.global_variables_initializer().run()

        # 模型和摘要的保存目录
        import time
        timestamp = str(int(time.time()))
        out_dir = os.path.abspath(
            os.path.join(os.path.curdir, 'runs', timestamp))
        print('\nWriting to {}\n'.format(out_dir))
        # 损失值和正确率的摘要
        loss_summary = tf.summary.scalar('loss', cross_entropy_mean)
        acc_summary = tf.summary.scalar('accuracy', evaluation_step)
        # 训练摘要
        train_summary_op = tf.summary.merge([loss_summary, acc_summary])
        train_summary_dir = os.path.join(out_dir, 'summaries', 'train')
        train_summary_writer = tf.summary.FileWriter(train_summary_dir,
                                                     sess.graph)
        # 开发摘要
        dev_summary_op = tf.summary.merge([loss_summary, acc_summary])
        dev_summary_dir = os.path.join(out_dir, 'summaries', 'dev')
        dev_summary_writer = tf.summary.FileWriter(dev_summary_dir, sess.graph)
        # 保存检查点
        checkpoint_dir = os.path.abspath(os.path.join(out_dir, 'checkpoints'))
        checkpoint_prefix = os.path.join(checkpoint_dir, 'model')
        if not os.path.exists(checkpoint_dir):
            os.makedirs(checkpoint_dir)
            saver = tf.train.Saver(
                tf.global_variables(), max_to_keep=NUM_CHECKPOINTS)

        for i in range(STEPS):
            # 每次获取一个batch的训练数据
            train_bottlenecks, train_ground_truth = get_random_cached_bottlenecks(
                sess, n_classes, image_lists, BATCH, 'training',
                jpeg_data_tensor, bottleneck_tensor)
            _, train_summaries = sess.run(
                [train_step, train_summary_op],
                feed_dict={
                    bottleneck_input: train_bottlenecks,
                    ground_truth_input: train_ground_truth
                })

            # 保存每步的摘要
            train_summary_writer.add_summary(train_summaries, i)

            # 在验证集上测试正确率
            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, dev_summaries = sess.run(
                    [evaluation_step, dev_summary_op],
                    feed_dict={
                        bottleneck_input: validation_bottlenecks,
                        ground_truth_input: validation_ground_truth
                    })
                print(
                    'Step %d : Validation accuracy on random sampled %d examples = %.1f%%'
                    % (i, BATCH, validation_accuracy * 100))

            # 每隔checkpoint_every保存一次模型和测试摘要
            if i % CHECKPOINT_EVERY == 0:
                dev_summary_writer.add_summary(dev_summaries, i)
                path = saver.save(sess, checkpoint_prefix, global_step=i)
                print('Saved model checkpoint to {}\n'.format(path))

        # 最后在测试集上测试正确率
        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))

        # 保存标签
        output_labels = os.path.join(out_dir, 'labels.txt')
        with tf.gfile.FastGFile(output_labels, 'w') as f:
            keys = list(image_lists.keys())
            for i in range(len(keys)):
                keys[i] = '%2d -> %s' % (i, keys[i])
            f.write('\n'.join(keys) + '\n')


if __name__ == '__main__':
    tf.app.run()

训练结果：

Writing to D:\New Scenery-AI\detection_flowers\runs\1548057429

Step 0 : Validation accuracy on random sampled 100 examples = 19.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-0

Step 100 : Validation accuracy on random sampled 100 examples = 75.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-100

Step 200 : Validation accuracy on random sampled 100 examples = 81.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-200

Step 300 : Validation accuracy on random sampled 100 examples = 78.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-300

Step 400 : Validation accuracy on random sampled 100 examples = 87.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-400

Step 500 : Validation accuracy on random sampled 100 examples = 90.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-500

Step 600 : Validation accuracy on random sampled 100 examples = 90.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-600

Step 700 : Validation accuracy on random sampled 100 examples = 89.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-700

Step 800 : Validation accuracy on random sampled 100 examples = 93.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-800

Step 900 : Validation accuracy on random sampled 100 examples = 83.0%
Saved model checkpoint to D:\New Scenery-AI\detection_flowers\runs\1548057429\checkpoints\model-900

Step 999 : Validation accuracy on random sampled 100 examples = 82.0%
Final test accuracy = 85.6%

2、eval.py

import tensorflow as tf
import numpy as np

# 模型目录
CHECKPOINT_DIR = './runs/1548061861/checkpoints'
INCEPTION_MODEL_FILE = 'model/tensorflow_inception_graph.pb'

# inception-v3模型参数
BOTTLENECK_TENSOR_NAME = 'pool_3/_reshape:0'  # inception-v3模型中代表瓶颈层结果的张量名称
JPEG_DATA_TENSOR_NAME = 'DecodeJpeg/contents:0'  # 图像输入张量对应的名称

# 测试数据
file_path = './data/flower_photos/roses/295257304_de893fc94d.jpg'
# file_path = './data/flower_photos/roses/12240303_80d87f77a3_n.jpg'
# file_path = './data/flower_photos/dandelion/7355522_b66e5d3078_m.jpg'
# file_path = './data/flower_photos/dandelion/16159487_3a6615a565_n.jpg'
# file_path = './data/flower_photos/sunflowers/6953297_8576bf4ea3.jpg'
# file_path = './data/flower_photos/sunflowers/40410814_fba3837226_n.jpg'
# file_path = './data/flower_photos/tulips/11746367_d23a35b085_n.jpg'
y_test = [1]

# 读取数据
image_data = tf.gfile.GFile(file_path, 'rb').read()

# 评估
checkpoint_file = tf.train.latest_checkpoint(CHECKPOINT_DIR)
with tf.Graph().as_default() as graph:
    with tf.Session().as_default() as sess:
        # 读取训练好的inception-v3模型
        with tf.gfile.GFile(INCEPTION_MODEL_FILE, 'rb') as f:
            graph_def = tf.GraphDef()
            graph_def.ParseFromString(f.read())

        # 加载inception-v3模型，并返回数据输入张量和瓶颈层输出张量
        bottleneck_tensor, jpeg_data_tensor = tf.import_graph_def(
            graph_def,
            return_elements=[BOTTLENECK_TENSOR_NAME, JPEG_DATA_TENSOR_NAME])

        # 使用inception-v3处理图片获取特征向量
        bottleneck_values = sess.run(bottleneck_tensor,
                                     {jpeg_data_tensor: image_data})
        # 将四维数组压缩成一维数组，由于全连接层输入时有batch的维度，所以用列表作为输入
        bottleneck_values = [np.squeeze(bottleneck_values)]

        # 加载元图和变量
        saver = tf.train.import_meta_graph('{}.meta'.format(checkpoint_file))
        saver.restore(sess, checkpoint_file)

        # 通过名字从图中获取输入占位符
        input_x = graph.get_operation_by_name(
            'BottleneckInputPlaceholder').outputs[0]

        # 我们想要评估的tensors
        predictions = graph.get_operation_by_name('evaluation/ArgMax').outputs[
            0]

        # 收集预测值
        all_predictions = []
        all_predictions = sess.run(predictions, {input_x: bottleneck_values})

# 如果提供了标签则打印正确率
if y_test is not None:
    correct_predictions = float(sum(all_predictions == y_test))
    print('\nTotal number of test examples: {}'.format(len(y_test)))
    print('Accuracy: {:g}'.format(correct_predictions / float(len(y_test))))

结果：

python3 eval.py


Total number of test examples: 1
Accuracy: 1

注意：

问：更换了图片，为什么accuracy总是为0的问题？怎么解决？
答：\detection_flowers\runs\1548061861，在你自己的目录下去找一个label.txt的文件
0 -> tulips 1 -> roses 2 -> daisy 3 -> dandelion 4 -> sunflowers
这个标签对应的就是你要测试的图片的标签，如果你选用的是roses，那么你的标签就为1。y_test = [1]，1是标签，意思就是测试的图片是这一类，也就是roses 。如果你的标签与你的图片不是对应的，那么就会出现一直是0的情况，意思就是测试错误。

结束。希望对你有用~