美文网首页深度学习
TensorFlow-slim 训练 CNN 分类模型(续)

TensorFlow-slim 训练 CNN 分类模型(续)

作者: 公输睚信 | 来源:发表于2018-04-20 21:57 被阅读733次

            在前面的文章 TensorFlow-slim 训练 CNN 分类模型 我们已经使用过 tf.contrib.slim 模块来构建和训练模型了,今天我们继续这一话题,但稍有不同的是我们不再定义数据输入输出的占位符,而是使用 tf.contrib.slim 模块来导入 tfrecord 数据进行训练。这样的训练方式主要有两个优点:1.数据是并行读取的,而且并非一次性全部导入到内存,因此可以缓解内存不足的问题;2.使用 tf.contrib.slim 模块的封装函数 slim.learning.train 来训练,可以直接使用 Tensroboard 来监督损失以及准确率等曲线,还可以在中断训练后继续从上次保存的位置继续进行训练。

            我们考虑的问题仍然是 10 分类由 Captcha 生成的数字,具体请访问:TensorFlow 训练 CNN 分类器,使用的模型与文章 TensorFlow-slim 训练 CNN 分类模型 的相同。为了方便阅读,再次将模型的代码粘贴如下(命名为:model.py):

    # -*- coding: utf-8 -*-
    """
    Created on Fri Mar 30 16:54:02 2018
    
    @author: shirhe-lyh
    """
    
    import tensorflow as tf
    
    from abc import ABCMeta
    from abc import abstractmethod
    
    slim = tf.contrib.slim
    
    
    class BaseModel(object):
        """Abstract base class for any model."""
        __metaclass__ = ABCMeta
        
        def __init__(self, num_classes):
            """Constructor.
            
            Args:
                num_classes: Number of classes.
            """
            self._num_classes = num_classes
            
        @property
        def num_classes(self):
            return self._num_classes
        
        @abstractmethod
        def preprocess(self, inputs):
            """Input preprocessing. To be override by implementations.
            
            Args:
                inputs: A float32 tensor with shape [batch_size, height, width,
                    num_channels] representing a batch of images.
                
            Returns:
                preprocessed_inputs: A float32 tensor with shape [batch_size, 
                    height, widht, num_channels] representing a batch of images.
            """
            pass
        
        @abstractmethod
        def predict(self, preprocessed_inputs):
            """Predict prediction tensors from inputs tensor.
            
            Outputs of this function can be passed to loss or postprocess functions.
            
            Args:
                preprocessed_inputs: A float32 tensor with shape [batch_size,
                    height, width, num_channels] representing a batch of images.
                
            Returns:
                prediction_dict: A dictionary holding prediction tensors to be
                    passed to the Loss or Postprocess functions.
            """
            pass
        
        @abstractmethod
        def postprocess(self, prediction_dict, **params):
            """Convert predicted output tensors to final forms.
            
            Args:
                prediction_dict: A dictionary holding prediction tensors.
                **params: Additional keyword arguments for specific implementations
                    of specified models.
                    
            Returns:
                A dictionary containing the postprocessed results.
            """
            pass
        
        @abstractmethod
        def loss(self, prediction_dict, groundtruth_lists):
            """Compute scalar loss tensors with respect to provided groundtruth.
            
            Args:
                prediction_dict: A dictionary holding prediction tensors.
                groundtruth_lists: A list of tensors holding groundtruth
                    information, with one entry for each image in the batch.
                    
            Returns:
                A dictionary mapping strings (loss names) to scalar tensors
                    representing loss values.
            """
            pass
        
            
    class Model(BaseModel):
        """A simple 10-classification CNN model definition."""
        
        def __init__(self,
                     is_training,
                     num_classes):
            """Constructor.
            
            Args:
                is_training: A boolean indicating whether the training version of
                    computation graph should be constructed.
                num_classes: Number of classes.
            """
            super(Model, self).__init__(num_classes=num_classes)
            
            self._is_training = is_training
            
        def preprocess(self, inputs):
            """Predict prediction tensors from inputs tensor.
            
            Outputs of this function can be passed to loss or postprocess functions.
            
            Args:
                preprocessed_inputs: A float32 tensor with shape [batch_size,
                    height, width, num_channels] representing a batch of images.
                
            Returns:
                prediction_dict: A dictionary holding prediction tensors to be
                    passed to the Loss or Postprocess functions.
            """
            preprocessed_inputs = tf.to_float(inputs)
            preprocessed_inputs = tf.subtract(preprocessed_inputs, 128.0)
            preprocessed_inputs = tf.div(preprocessed_inputs, 128.0)
            return preprocessed_inputs
        
        def predict(self, preprocessed_inputs):
            """Predict prediction tensors from inputs tensor.
            
            Outputs of this function can be passed to loss or postprocess functions.
            
            Args:
                preprocessed_inputs: A float32 tensor with shape [batch_size,
                    height, width, num_channels] representing a batch of images.
                
            Returns:
                prediction_dict: A dictionary holding prediction tensors to be
                    passed to the Loss or Postprocess functions.
            """
            with slim.arg_scope([slim.conv2d, slim.fully_connected],
                                activation_fn=tf.nn.relu):
                net = preprocessed_inputs
                net = slim.repeat(net, 2, slim.conv2d, 32, [3, 3], scope='conv1')
                net = slim.max_pool2d(net, [2, 2], scope='pool1')
                net = slim.repeat(net, 2, slim.conv2d, 64, [3, 3], scope='conv2')
                net = slim.max_pool2d(net, [2, 2], scope='pool2')
                net = slim.repeat(net, 2, slim.conv2d, 128, [3, 3], scope='conv3')
                net = slim.flatten(net, scope='flatten')
                net = slim.dropout(net, keep_prob=0.5, 
                                   is_training=self._is_training)
                net = slim.fully_connected(net, 512, scope='fc1')
                net = slim.fully_connected(net, 512, scope='fc2')
                net = slim.fully_connected(net, self.num_classes, 
                                           activation_fn=None, scope='fc3')
            prediction_dict = {'logits': net}
            return prediction_dict
        
        def postprocess(self, prediction_dict):
            """Convert predicted output tensors to final forms.
            
            Args:
                prediction_dict: A dictionary holding prediction tensors.
                **params: Additional keyword arguments for specific implementations
                    of specified models.
                    
            Returns:
                A dictionary containing the postprocessed results.
            """
            logits = prediction_dict['logits']
            logits = tf.nn.softmax(logits)
            classes = tf.cast(tf.argmax(logits, axis=1), dtype=tf.int32)
            postprecessed_dict = {'classes': classes}
            return postprecessed_dict
        
        def loss(self, prediction_dict, groundtruth_lists):
            """Compute scalar loss tensors with respect to provided groundtruth.
            
            Args:
                prediction_dict: A dictionary holding prediction tensors.
                groundtruth_lists: A list of tensors holding groundtruth
                    information, with one entry for each image in the batch.
                    
            Returns:
                A dictionary mapping strings (loss names) to scalar tensors
                    representing loss values.
            """
            logits = prediction_dict['logits']
            loss = tf.reduce_mean(
                tf.nn.sparse_softmax_cross_entropy_with_logits(
                    logits=logits, labels=groundtruth_lists))
            loss_dict = {'loss': loss}
            return loss_dict
    

            下面重点说明怎么使用 tf.contrib.slim 模块来训练。本文所有代码见 github: slim_cnn_test

    一、slim.learning.train 训练 CNN 模型

            我们已经知道通过定义数据入口、出口的占位符 tf.placeholder 可以对上面定义的模型进行训练,但现在我们的目标是全部使用 tf.contrib.slim 来进行托管式的训练。要完成这个目标,需要借助两个函数的辅助,分别是上一篇文章 TensorFlow 自定义生成 .record 文件 定义的 get_record_dataset 函数和 slim 模块封装的 slim.learning.train 函数,前者用于获取训练数据,后者则执行对模型的训练。以下,先将训练用的代码列举出来(命名为 train.py):

    #!/usr/bin/env python3
    # -*- coding: utf-8 -*-
    """
    Created on Fri Mar 30 19:27:44 2018
    @author: shirhe-lyh
    """
    
    """Train a CNN model to classifying 10 digits.
    Example Usage:
    ---------------
    python3 train.py \
        --record_path: Path to training tfrecord file.
        --logdir: Path to log directory.
    """
    
    import tensorflow as tf
    
    import model
    
    slim = tf.contrib.slim
    flags = tf.app.flags
    
    flags.DEFINE_string('record_path', None, 'Path to training tfrecord file.')
    flags.DEFINE_string('logdir', None, 'Path to log directory.')
    FLAGS = flags.FLAGS
    
    
    def get_record_dataset(record_path,
                           reader=None, image_shape=[28, 28, 3], 
                           num_samples=50000, num_classes=10):
        """Get a tensorflow record file.
        
        Args:
            
        """
        if not reader:
            reader = tf.TFRecordReader
            
        keys_to_features = {
            'image/encoded': 
                tf.FixedLenFeature((), tf.string, default_value=''),
            'image/format': 
                tf.FixedLenFeature((), tf.string, default_value='jpeg'),
            'image/class/label': 
                tf.FixedLenFeature([1], tf.int64, default_value=tf.zeros([1], 
                                   dtype=tf.int64))}
            
        items_to_handlers = {
            'image': slim.tfexample_decoder.Image(shape=image_shape, 
                                                  #image_key='image/encoded',
                                                  #format_key='image/format',
                                                  channels=3),
            'label': slim.tfexample_decoder.Tensor('image/class/label', shape=[])}
        
        decoder = slim.tfexample_decoder.TFExampleDecoder(
            keys_to_features, items_to_handlers)
        
        labels_to_names = None
        items_to_descriptions = {
            'image': 'An image with shape image_shape.',
            'label': 'A single integer between 0 and 9.'}
        return slim.dataset.Dataset(
            data_sources=record_path,
            reader=reader,
            decoder=decoder,
            num_samples=num_samples,
            num_classes=num_classes,
            items_to_descriptions=items_to_descriptions,
            labels_to_names=labels_to_names)
    
    
    def main(_):
        dataset = get_record_dataset(FLAGS.record_path)
        data_provider = slim.dataset_data_provider.DatasetDataProvider(dataset)
        image, label = data_provider.get(['image', 'label'])
        inputs, labels = tf.train.batch([image, label],
                                        batch_size=64,
                                        allow_smaller_final_batch=True)
        
        cls_model = model.Model(is_training=True, num_classes=10)
        preprocessed_inputs = cls_model.preprocess(inputs)
        prediction_dict = cls_model.predict(preprocessed_inputs)
        loss_dict = cls_model.loss(prediction_dict, labels)
        loss = loss_dict['loss']
        postprocessed_dict = cls_model.postprocess(prediction_dict)
        classes = postprocessed_dict['classes']
        acc = tf.reduce_mean(tf.cast(tf.equal(classes, labels), 'float'))
        tf.summary.scalar('loss', loss)
        tf.summary.scalar('accuracy', acc)
        
        optimizer = tf.train.MomentumOptimizer(learning_rate=0.01, momentum=0.9)
        train_op = slim.learning.create_train_op(loss, optimizer,
                                                 summarize_gradients=True)
        
        slim.learning.train(train_op=train_op, logdir=FLAGS.logdir,
                            save_summaries_secs=20, save_interval_secs=120)
        
    if __name__ == '__main__':
        tf.app.run()
    

            进行训练时,我们先根据上一篇文章的方式将所有训练图像(以及相应的标签)写成 tfrecord 文件,这一步如果训练数据足够多的话会耗费较长时间,但可以方便后续的数据读取(读取数据很快)。然后,我们接着上一篇文章的后一部分,要把这些数据读出来喂给模型。我们已经定义好了函数 get_record_dataset,它读取 .record 文件并返回一个 slim.dataset.Dataset 类对象。此时,用 slim.dataset_data_provider.DatasetDataProvider 作用一下,便可以用该类的函数 .get 来方便的将数据并行的提取出来(见上一篇文章)。因为并行的缘故,它每次返回的是单张图像,这时你可以对图像进行非批量的预处理,也可以直接使用 tf.train.batch 来生成指定大小的一个批量然后进行批量预处理。

            一旦将训练数据提取出来,我们就可以把它们传给模型了(如上述代码 main 函数中间片段),最后的两句 tf.summary.sccalar 表示把损失和准确率写入到训练日志文件,方便之后我们在 tensorboard 中观察损失、准确率曲线。然后,再声明使用动量的随机梯度下降优化算法,紧接着便来到了训练的最后一步:

    slim.learning.train(train_op=train_op, logdir=FLAGS.logdir,
                        save_summaries_secs=20, save_interval_secs=True)
    

    这个函数将所有训练的迭代过程都封装起来,包括日志文件书写和模型保存。函数

    slim.learning.train(train_op, logdir, train_step_fn=train_step,
                        train_step_kwargs=_USE_DEFAULT,
                        log_every_n_steps=1, graph=None, master='',
                        is_chief=True, global_step=None,
                        number_of_steps=None, init_op=_USE_DEFAULT,
                        init_feed_dict=None, local_init_op=_USE_DEFAULT,
                        init_fn=None, ready_op=_USE_DEFAULT,
                        summary_op=_USE_DEFAULT,
                        save_summaried_secs=600,
                        summary_writer=_USE_DEFAULT,
                        startup_delay_steps=0, saver=None,
                        save_interval_secs=600, sync_optimizer=None,
                        session_config=None, session_wrapper=None,
                        trace_every_n_steps=None,
                        ignore_live_threads=False)
    

    参数众多,其中重要的有:1.train_op,指定优化算法;2.logdir,指定训练数据保存文件夹;3.save_summaries_secs,指定每隔多少秒更新一次日志文件(对应 tensorboard 刷新一次的时间);4.save_interval_secs,指定每隔多少秒保存一次模型。

            注意这段代码并没有定义数据传入的占位符,因此模型训练完成之后,我们并不知道怎么用,囧。不过,没关系,我们先训练完再说,在终端执行:

    python3 train.py --record_path path/to/.record --logdir path_to_log_directory
    

    会在 logdir 指定的路径下生成多个训练文件,而且每隔 save_interval_secs 会自动更新这些文件。当觉得模型训练到可以终止的时候,可以使用 Ctrl + C 来强制终止,或者通过指定参数 number_of_steps 来终止。而当觉得有必要对所训练的模型继续进行训练时,重新执行上述命令即可。如果你想查看训练的损失和准确率变化情况,在终端使用命令:

    tensorboard --logdir path_to_log_directory
    

    即可,至于查看的时间可以在训练过程中,也可以在训练结束后。

            好了,有关使用 tf.contrib.slim 模块来进行模型训练的内容就讲完了,接下来还要解决一个重要的问题:没有定义占位符,我们怎么调用模型进行推断?

    二、自定义模型导出(.pb 格式)

            上面训练的模型会在 logdir 目录下保存为 .ckpt 格式的文件, 但一个致命的问题是没有数据入口,不知如何用它来对图像进行分类。为此,需要人为的添加数据入口、出口的结点。

            要达到这个目的,我们需要仔细的研究一下下面这个用于模型导出的文件:

    TensorFlow models/research/object_detection/export.py

    然后基于该文件做部分修改,用来导出我们的模型。请期待下一篇文章,我们将完成这一任务。

    预告:下一篇文章将详细说明自定义的将 .ckpt 文件导出为 .pb 文件,敬请期待。

    相关文章

      网友评论

      • 俺是读书人:作者您好:我在在运行train.py的时候为什么会出现这样的错误,我刚刚开始学这方面的知识,有点不明白
        File "/home/swx/pycharm/pythoncode/tensorflow_cnn/day3/train.py", line 84, in main
        acc = tf.reduce_mean(tf.cast(tf.equal(classes, labels), 'float'))
        File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/gen_math_ops.py", line 721, in equal
        result = _op_def_lib.apply_op("Equal", x=x, y=y, name=name)
        TypeError: Input 'y' of 'Equal' Op has type int64 that does not match type int32 of argument 'x'.
        公输睚信:@俺是读书人 这个错误说 labels 的数据类型与 classes 的数据类型不一致,你在这条语句 acc = tf.reduce_mean(tf.cast(tf.equal(classes, labels), 'float')) 之前加入一句 labels = tf.cast(labels, tf.int32) 应该就可以了

      本文标题:TensorFlow-slim 训练 CNN 分类模型(续)

      本文链接:https://www.haomeiwen.com/subject/lrsykftx.html