加载图像

亚历克斯Net(亚历克斯Krizhevsky,ILSVRC2013季军卡塔尔国相符做图像分类。层自左向右、自上向下读取,关联层分为风流浪漫组,中度、宽度减小,深度扩充。深度扩展收缩网络计算量。

教练模型数据集 StanfordComputer视觉站点Stanford Dogs
http://vision.stanford.edu/aditya86/ImageNetDogs/
。数据下载解压到模型代码同一路线imagenet-dogs目录下。满含的120种狗图像。九成练习,五分之二测量检验。产物模型须求预先流出原始数据交叉验证。每幅图像JPEG格式(GL450GB卡塔尔(英语:State of Qatar),尺寸不后生可畏。

图像转TFRecord文件,有助加快练习,简化图像标签般配,图像分离利用检查点文件对模型实行不间断测量检验。转变图像格式把颜色空间转灰度,图像改善统风华正茂尺寸,标签除上每幅图像。训练前只实行二次预管理,时间较长。

glob.glob
枚举钦定路径目录,显示数据集文件布局。“*”通配符可以兑现模糊查找。文件名中8个数字对应ImageNet连串WordNetID。ImageNet网址可用WordNetID查图像细节:
http://www.image-net.org/synset?wnid=n02085620

文件名分解为品种和对应的公文名,品种对应文件夹名称。依靠品种对图像分组。枚举每种门类图像,33.33%图像划入测验集。检查每一种项目测验图疑似否至少有整整图像的18%。目录和图像协会到八个与种种品种有关的字典,包含种种型全体图像。分类图像组织到字典中,简化选拔分类图像及分类进程。

葡萄娱乐官方,预管理阶段,依次遍历全数分类图像,张开列表中文件。用dataset图像填充TFRecord文件,把项目满含进去。dataset键值对应文件列表标签。record_location
存款和储蓄TFRecord输出路径。枚举dataset,当前目录用于文书划分,每隔100m幅图像,练习样板新闻写入新的TFRecord文件,加快写操作进度。不大概被TensorFlow识别为JPEG图像,用try/catch忽视。转为灰度图收缩计算量和内部存款和储蓄器占用。tf.cast把RubiconGB值转变来[0,1卡塔尔区间内。标签按字符串存款和储蓄较便捷,最棒调换为整数索引或独热编码秩1张量。

开拓每幅图像,转变为灰度图,调解尺寸,增多到TFRecord文件。tf.image.resize_images函数把富有图像调节为同一尺寸,不思忖长度宽度比,有扭动。裁剪、边界填充能保持图像长度宽度比。

根据TFRecord文件读取图像,每回加载少许图像及标签。改进图像形状有助练习和出口可视化。相称全体在教练集目录下TFRecord文件加载训练图像。每种TFRecord文件富含多幅图像。tf.parse_single_example只从文件提取单个样板。批运算可同期演练多幅图像或单幅图像,要求丰硕系统内部存款和储蓄器。

图像转灰度值为[0,1卡塔尔浮点类型,相配convolution2d盼望输入。卷积输出第1维和最后风华正茂维不退换,中间两维产生变化。tf.contrib.layers.convolution2d创立模型第1层。weights_initializer设置正态随机值,第豆蔻梢头组滤波器填充正态布满随机数。滤波器设置trainable,消息输入互联网,权值调解,进步模型精确率。
max_pool把出口降采样。ksize、strides
([1,2,2,1]卡塔尔(قطر‎,卷积输出形状减半。输出形状减小,不改造滤波器数量(输出通道卡塔尔(英语:State of Qatar)或图像批数量尺寸。减弱重量,与图像(滤波器卡塔尔(قطر‎高度、宽度有关。更加多输出通道,滤波器数量增加,2倍于第生机勃勃层。四个卷积和池化层收缩输入中度、宽度,扩大吃水。超多布局,卷积层和池化层当先5层。练习调节和测量检验时间更加长,能同盟更加多更复杂方式。
图像每种点与输出神经元营造全连接。softmax,全连接层须要二阶张量。第1维区分图像,第2维输入张量秩1张量。tf.reshape
提醒和平运动用其余全数维,-1把最后池化层调度为宏伟秩1张量。
池化层展开,互连网当前处境与预测全连接层整合。weights_initializer选用可调用参数,lambda表明式重返截断正态布满,内定布满规范差。dropout
裁减模型中神经元主要性。tf.contrib.layers.fully_connected
输出前边全部层与教练中分类的全连接。各个像素与分类关联。网络每一步将输入图像转变为滤波减小尺寸。滤波器与标签相称。裁减演习、测量检验网络总括量,输出更具经常性。

教练多少真实标签和模型预测结果,输入到教练优化器(优化每层权值卡塔尔总结模型损失。多次迭代,每一遍提高模型正确率。大多数分类函数(tf.nn.softmax卡塔尔(قطر‎必要数值类型标签。各个标签转变代表包括全部分类列表索引整数。tf.map_fn
相配每一个标签并赶回体体系表索引。map依靠目录列表创造包罗分类列表。tf.map_fn
可用钦点函数对数码流图张量映射,生成仅包罗各个标签在全体类标签列表索引秩1张量。tf.nn.softmax用索引预测。

调剂CNN,观察滤波器(卷积核卡塔尔国每轮迭代变化。设计能够CNN,第二个卷积层工作,输入权值被随机初阶化。权值通过图像激活,激活函数输出(特征图卡塔尔(قطر‎随机。特征图可视化,输出外观与原始图类似,被施加静力(static卡塔尔国。静力由全数权值的随便激发。经过多轮迭代,权值被调动拟合演练反馈,滤波器趋于生机勃勃致。互连网未有,滤波器与图像分歧渺小方式相近。tf.image_summary得到演习后的滤波器和特征图简单视图。数据流图图像概要输出(image
summary
output卡塔尔从总体掌握所选用的滤波器和输入图像特点图。TensorDebugger,迭代中以GIF动画查看滤波器变化。

文本输入存款和储蓄在SparseTensor,大多数份量为0。CNN使用稠密输入,各类值都重要,输入大部分分量非0。

 

    import tensorflow as tf
    import glob
    from itertools import groupby
    from collections import defaultdict
    sess = tf.InteractiveSession()
    image_filenames = glob.glob("./imagenet-dogs/n02*/*.jpg")
    image_filenames[0:2]
    training_dataset = defaultdict(list)
    testing_dataset = defaultdict(list)
    image_filename_with_breed = map(lambda filename: (filename.split("/")[2], filename), image_filenames)
    for dog_breed, breed_images in groupby(image_filename_with_breed, lambda x: x[0]):
        for i, breed_image in enumerate(breed_images):
            if i % 5 == 0:
                testing_dataset[dog_breed].append(breed_image[1])
            else:
                training_dataset[dog_breed].append(breed_image[1])
        breed_training_count = len(training_dataset[dog_breed])
        breed_testing_count = len(testing_dataset[dog_breed])
        breed_training_count_float = float(breed_training_count)
        breed_testing_count_float = float(breed_testing_count)
        assert round(breed_testing_count_float / (breed_training_count_float + breed_testing_count_float), 2) > 0.18, "Not enough testing images."
    print "training_dataset testing_dataset END ------------------------------------------------------"
    def write_records_file(dataset, record_location):
        writer = None
        current_index = 0
        for breed, images_filenames in dataset.items():
            for image_filename in images_filenames:
                if current_index % 100 == 0:
                    if writer:
                        writer.close()
                    record_filename = "{record_location}-{current_index}.tfrecords".format(
                        record_location=record_location,
                        current_index=current_index)
                    writer = tf.python_io.TFRecordWriter(record_filename)
                    print record_filename + "------------------------------------------------------" 
                current_index += 1
                image_file = tf.read_file(image_filename)
                try:
                    image = tf.image.decode_jpeg(image_file)
                except:
                    print(image_filename)
                    continue
                grayscale_image = tf.image.rgb_to_grayscale(image)
                resized_image = tf.image.resize_images(grayscale_image, [250, 151])
                image_bytes = sess.run(tf.cast(resized_image, tf.uint8)).tobytes()
                image_label = breed.encode("utf-8")
                example = tf.train.Example(features=tf.train.Features(feature={
                    'label': tf.train.Feature(bytes_list=tf.train.BytesList(value=[image_label])),
                    'image': tf.train.Feature(bytes_list=tf.train.BytesList(value=[image_bytes]))
                }))
                writer.write(example.SerializeToString())
        writer.close()
    write_records_file(testing_dataset, "./output/testing-images/testing-image")
    write_records_file(training_dataset, "./output/training-images/training-image")
    print "write_records_file testing_dataset training_dataset END------------------------------------------------------"
    filename_queue = tf.train.string_input_producer(
    tf.train.match_filenames_once("./output/training-images/*.tfrecords"))
    reader = tf.TFRecordReader()
    _, serialized = reader.read(filename_queue)
    features = tf.parse_single_example(
    serialized,
        features={
            'label': tf.FixedLenFeature([], tf.string),
            'image': tf.FixedLenFeature([], tf.string),
        })
    record_image = tf.decode_raw(features['image'], tf.uint8)
    image = tf.reshape(record_image, [250, 151, 1])
    label = tf.cast(features['label'], tf.string)
    min_after_dequeue = 10
    batch_size = 3
    capacity = min_after_dequeue + 3 * batch_size
    image_batch, label_batch = tf.train.shuffle_batch(
        [image, label], batch_size=batch_size, capacity=capacity, min_after_dequeue=min_after_dequeue)
    print "load image from TFRecord END------------------------------------------------------"
    float_image_batch = tf.image.convert_image_dtype(image_batch, tf.float32)
    conv2d_layer_one = tf.contrib.layers.convolution2d(
        float_image_batch,
        num_outputs=32,
        kernel_size=(5,5),
        activation_fn=tf.nn.relu,
        weights_initializer=tf.random_normal,
        stride=(2, 2),
        trainable=True)
    pool_layer_one = tf.nn.max_pool(conv2d_layer_one,
        ksize=[1, 2, 2, 1],
        strides=[1, 2, 2, 1],
        padding='SAME')
    conv2d_layer_one.get_shape(), pool_layer_one.get_shape()
    print "conv2d_layer_one pool_layer_one END------------------------------------------------------"
    conv2d_layer_two = tf.contrib.layers.convolution2d(
        pool_layer_one,
        num_outputs=64,
        kernel_size=(5,5),
        activation_fn=tf.nn.relu,
        weights_initializer=tf.random_normal,
        stride=(1, 1),
        trainable=True)
    pool_layer_two = tf.nn.max_pool(conv2d_layer_two,
        ksize=[1, 2, 2, 1],
        strides=[1, 2, 2, 1],
        padding='SAME')
    conv2d_layer_two.get_shape(), pool_layer_two.get_shape()
    print "conv2d_layer_two pool_layer_two END------------------------------------------------------"
    flattened_layer_two = tf.reshape(
        pool_layer_two,
        [
            batch_size,
            -1
        ])
    flattened_layer_two.get_shape()
    print "flattened_layer_two END------------------------------------------------------"
    hidden_layer_three = tf.contrib.layers.fully_connected(
        flattened_layer_two,
        512,
        weights_initializer=lambda i, dtype: tf.truncated_normal([38912, 512], stddev=0.1),
        activation_fn=tf.nn.relu
    )
    hidden_layer_three = tf.nn.dropout(hidden_layer_three, 0.1)
    final_fully_connected = tf.contrib.layers.fully_connected(
        hidden_layer_three,
        120,
        weights_initializer=lambda i, dtype: tf.truncated_normal([512, 120], stddev=0.1)
    )
    print "final_fully_connected END------------------------------------------------------"
    labels = list(map(lambda c: c.split("/")[-1], glob.glob("./imagenet-dogs/*")))
    train_labels = tf.map_fn(lambda l: tf.where(tf.equal(labels, l))[0,0:1][0], label_batch, dtype=tf.int64)
    loss = tf.reduce_mean(
        tf.nn.sparse_softmax_cross_entropy_with_logits(
            final_fully_connected, train_labels))
    batch = tf.Variable(0)
    learning_rate = tf.train.exponential_decay(
        0.01,
        batch * 3,
        120,
        0.95,
        staircase=True)
    optimizer = tf.train.AdamOptimizer(
        learning_rate, 0.9).minimize(
        loss, global_step=batch)
    train_prediction = tf.nn.softmax(final_fully_connected)
    print "train_prediction END------------------------------------------------------"
    filename_queue.close(cancel_pending_enqueues=True)
    coord.request_stop()
    coord.join(threads)
    print "END------------------------------------------------------"

 

参照他事他说加以考察资料:
《面向机器智能的TensorFlow试行》

应接加笔者微信调换:qingxingfengzi
自家的Wechat公众号:qingxingfengzigz
本身老婆张幸清的Wechat民众号:qingqingfeifangz