标签除上每幅图像,标签除上每幅图像

AlexNet(亚历克斯Krizhevsky,ILSVRC二〇一二季军)适合做图像分类。层自左向右、自上向下读取,关联层分为一组,中度、宽度减小,深度扩张。深度扩张减弱互联网总括量。

读书笔记TF016:CNN实现、数据集、TFRecord、加载图像、模型、磨练、调试,tf016tfrecord

亚历克斯Net(AlexKrizhevsky,ILSVRC2013亚军)适合做图像分类。层自左向右、自上向下读取,关联层分为一组,中度、宽度减小,深度扩展。深度扩大裁减互联网总结量。

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。包罗的120种狗图像。十分八教练,五分之三测量试验。产品模型需求预留原始数据交叉验证。每幅图像JPEG格式(奥德赛GB),尺寸不一。

图像转TFRecord文件,有助加快陶冶,简化图像标签匹配,图像分离利用检查点文件对模型进行不间断测量检验。调换图像格式把颜色空间转灰度,图像修改统一尺寸,标签除上每幅图像。练习前只实行一回预处理,时间较长。

glob.glob
枚举钦赐路径目录,显示数据集文件结构。“*”通配符能够兑现模糊查找。文件名中8个数字对应ImageNet种类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文书名,品种对应文件夹名称。依赖品种对图像分组。枚举每一种品种图像,五分一图像划入测验集。检查各样种类测量试验图疑似否至少有整整图像的18%。目录和图像协会到多少个与每种门类有关的字典,包罗各品种全部图像。分类图像组织到字典中,简化选用分类图像及分类进度。

预管理阶段,依次遍历全部分类图像,打开列表普通话件。用dataset图像填充TFRecord文件,把项目包括进去。dataset键值对应文件列表标签。record_location
存储TFRecord输出路线。枚举dataset,当前目录用于文书划分,每隔100m幅图像,磨炼样本消息写入新的TFRecord文件,加速写操作进程。不能够被TensorFlow识别为JPEG图像,用try/catch忽略。转为灰度图减弱计算量和内部存款和储蓄器占用。tf.cast把昂CoraGB值调换成[0,1)区间内。标签按字符串存款和储蓄较高速,最棒转变为整数索引或独热编码秩1张量。

展开每幅图像,调换为灰度图,调度尺寸,增加到TFRecord文件。tf.image.resize_images函数把富有图像调治为一样尺寸,不思量长度宽度比,有扭动。裁剪、边界填充能保持图像长度宽度比。

按部就班TFRecord文件读取图像,每一次加载一点点图像及标签。修改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文件加载磨炼图像。各样TFRecord文件满含多幅图像。tf.parse_single_example只从文件提取单个样本。批运算可同不经常间磨练多幅图像或单幅图像,需求足够系统内部存款和储蓄器。

图像转灰度值为[0,1)浮点类型,相称convolution2d目的在于输入。卷积输出第1维和最终一维不退换,中间两维发生变化。tf.contrib.layers.convolution2d创建立模型型第1层。weights_initializer设置正态随机值,第一组滤波器填充正态遍及随机数。滤波器设置trainable,新闻输入网络,权值调解,升高模型准确率。
max_pool把出口降采集样品。ksize、strides
([1,2,2,1]),卷积输出形状减半。输出形状减小,不更改滤波器数量(输出通道)或图像批数量尺寸。减弱重量,与图像(滤波器)中度、宽度有关。愈来愈多输出通道,滤波器数量扩充,2倍于第一层。八个卷积和池化层减弱输入高度、宽度,增添吃水。比比较多架构,卷积层和池化层抢先5层。磨炼调节和测验时间更加长,能配合越多更眼花缭乱格局。
图像每一种点与出口神经元建设构造全连接。softmax,全连接层必要二阶张量。第1维区分图像,第2维输入张量秩1张量。tf.reshape
提示和采纳任何所有维,-1把最终池化层调解为远大秩1张量。
池化层张开,网络当前意况与预测全连接层整合。weights_initializer接收可调用参数,lambda表明式重临截断正态分布,钦定分布标准差。dropout
削减模型中神经元主要性。tf.contrib.layers.fully_connected
输出后边全数层与磨练中分类的全连接。每一个像素与分类关联。网络每一步将输入图像转化为滤波减小尺码。滤波器与标签相称。收缩磨练、测量试验网络计算量,输出更具一般性。

陶冶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)总结模型损失。数次迭代,每趟提高模型精确率。超越50%分拣函数(tf.nn.softmax)须要数值类型标签。每一个标签调换代表富含全数分类列表索引整数。tf.map_fn
相称种种标签并重回种类列表索引。map依附目录列表成立包涵分类列表。tf.map_fn
可用钦命函数对数据流图张量映射,生成仅蕴含各种标签在具有类标签列表索引秩1张量。tf.nn.softmax用索引预测。

调解CNN,观望滤波器(卷积核)每轮迭代变化。设计出色CNN,第贰个卷积层工作,输入权值被随意初步化。权值通过图像激活,激活函数输出(特征图)随机。特征图可视化,输出外观与原始图相似,被施加静力(static)。静力由全体权值的即兴激发。经过多轮迭代,权值被调解拟合演练反馈,滤波器趋于一致。网络未有,滤波器与图像差异细小方式类似。tf.image_summary得报到并且接受集练习后的滤波器和特征图容易视图。数据流图图像概要输出(image
summary
output)从完整掌握所采纳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,每一个值都主要,输入超过八分之四分量非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实践》

接待加我微信交换:qingxingfengzi
自身的微信公众号:qingxingfengzigz
本身老伴张幸清的微信徒人号:qingqingfeifangz

http://www.bkjia.com/Pythonjc/1213552.htmlwww.bkjia.comtruehttp://www.bkjia.com/Pythonjc/1213552.htmlTechArticle学习笔记TF016:CNN实现、数据集、TFRecord、加载图像、模型、训练、调试,tf016tfrecord
亚历克斯Net(亚历克斯 Krizhevsky,ILSVRC2011季军)适合做图像分类。层自左…

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。包含的120种狗图像。百分之九十教练,40%测量检验。产品模型需求预留原始数据交叉验证。每幅图像JPEG格式(LX570GB),尺寸不一。

图像转TFRecord文件,有助加速锻炼,简化图像标签相称,图像分离利用检查点文件对模型进行不间断测量试验。调换图像格式把颜色空间转灰度,图像修改统一尺寸,标签除上每幅图像。练习前只实行三遍预管理,时间较长。

glob.glob
枚举钦命路线目录,显示数据集文件结构。“*”通配符能够兑现模糊查找。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

文本名分解为品种和相应的文本名,品种对应文件夹名称。依赖品种对图像分组。枚举各类品种图像,百分之三十图像划入测验集。检查各类体系测验图疑似否至少有总体图像的18%。目录和图像协会到四个与各种项目有关的字典,蕴涵各档期的顺序全体图像。分类图像组织到字典中,简化选取分类图像及分类进度。

预管理阶段,依次遍历全部分类图像,展开列表粤语件。用dataset图像填充TFRecord文件,把项目包蕴进去。dataset键值对应文件列表标签。record_location
存款和储蓄TFRecord输出路线。枚举dataset,当前目录用于文书划分,每隔100m幅图像,练习样本音讯写入新的TFRecord文件,加快写操作进度。不恐怕被TensorFlow识别为JPEG图像,用try/catch忽略。转为灰度图减少总计量和内部存款和储蓄器占用。tf.cast把奥德赛GB值调换成[0,1)区间内。标签按字符串存款和储蓄较便捷,最佳调换为整数索引或独热编码秩1张量。

张开每幅图像,转变为灰度图,调节尺寸,增加到TFRecord文件。tf.image.resize_images函数把全数图像调节为同一尺寸,不思虑长度宽度比,有扭动。裁剪、边界填充能保持图像长度宽度比。

依据TFRecord文件读取图像,每一趟加载一些些图像及标签。修改图像形状有助练习和输出可视化。相配全数在练习集目录下TFRecord文件加载磨炼图像。每一种TFRecord文件包涵多幅图像。tf.parse_single_example只从文件提取单个样本。批运算可同期陶冶多幅图像或单幅图像,需求丰盛系统内部存款和储蓄器。

图像转灰度值为[0,1)浮点类型,相称convolution2d期望输入。卷积输出第1维和最后一维不变,中间两维爆发变化。tf.contrib.layers.convolution2d成立模型第1层。weights_initializer设置正态随机值,第一组滤波器填充正态分布随机数。滤波器设置trainable,新闻输入互联网,权值调节,提升模型准确率。
max_pool把出口降采样。ksize、strides
([1,2,2,1]),卷积输出形状减半。输出形状减小,不退换滤波器数量(输出通道)或图像批数量尺寸。降低重量,与图像(滤波器)高度、宽度有关。越多输出通道,滤波器数量扩充,2倍于第一层。多少个卷积和池化层减弱输入中度、宽度,扩充吃水。比很多框架结构,卷积层和池化层超过5层。训练调节和测量试验时间更加长,能相称更加多更复杂情势。
图像每一种点与输出神经元构建全连接。softmax,全连接层必要二阶张量。第1维区分图像,第2维输入张量秩1张量。tf.reshape
提醒和使用其余全部维,-1把最终池化层调治为巨大秩1张量。
池化层张开,互联网当前景况与预测全连接层整合。weights_initializer接收可调用参数,lambda表达式再次来到截断正态分布,钦赐分布规范差。dropout
削减模型中神经元首要性。tf.contrib.layers.fully_connected
输出前面全体层与磨炼中分类的全连接。每一个像素与分类关联。网络每一步将输入图像转化为滤波减小尺寸。滤波器与标签相称。减弱磨练、测验网络计算量,输出更具一般性。

教练多少真实标签和模型预测结果,输入报到并且接受集演练优化器(优化每层权值)总括模型损失。数12遍迭代,每便提高模型准确率。超越55%分类函数(tf.nn.softmax)必要数值类型标签。各种标签转变代表包括全体分类列表索引整数。tf.map_fn
匹配每一个标签并重回种类列表索引。map依附目录列表创制富含分类列表。tf.map_fn
可用钦赐函数对数码流图张量映射,生成仅包括各个标签在颇具类标签列表索引秩1张量。tf.nn.softmax用索引预测。

调理CNN,阅览滤波器(卷积核)每轮迭代变化。设计精良CNN,第一个卷积层专门的学业,输入权值被放肆早先化。权值通过图像激活,激活函数输出(特征图)随机。特征图可视化,输出外观与原始图相似,被施加静力(static)。静力由全数权值的随机激发。经过多轮迭代,权值被调动拟合演习反馈,滤波器趋于同一。网络未有,滤波器与图像分裂细小方式类似。tf.image_summary得报到并且接受集练习后的滤波器和特点图简单视图。数据流图图像概要输出(image
summary
output)从总体明白所使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。

文件输入存款和储蓄在SparseTensor,抢先二分一重量为0。CNN使用稠密输入,每种值都重要,输入大部分轻重非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------------------------------------------------------"

 

仿效资料:
《面向机器智能的TensorFlow实施》

接待加笔者微信调换:qingxingfengzi
自家的微信公众号:qingxingfengzigz
本人爱妻张幸清的微信徒人号:qingqingfeifangz

相关文章