我正在上在线Udacity课程“ TensorFlow深度学习入门”。运行代码时出现以下错误-

Warning: unknown JFIF revision number 0.00 
Corrupt JPEG data: 296 extraneous bytes before marker 0xd9

我正在尝试使用带有MNIST数据集的MobileNet学习迁移学习。我的部分代码是-

    # getting tensorflow dataset here and splitting
    splits = dfds.Split.ALL.subsplit(weighted=(80,20))
    splits,info = dfds.load('cats_vs_dogs',with_info = True,as_supervised = True,split = splits)
    (train_data,test_data)=splits

    num_examples = info.splits['train'].num_examples
    num_classes = info.features['label'].num_classes

    # getting batches of resized train data
    def format_image(image,label):
        image = tf.image.resize(image,(IMG_SIZE,IMG_SIZE))
        image = image/255.0
        return image,label

    train_batches      = 
    train_data.shuffle(num_examples//4).map(format_image).batch(Batch_SIZE).prefetch(1)
    validation_batches = test_data.map(format_image).batch(Batch_SIZE).prefetch(1)

    image_batch,label_batch = next(iter(train_batches.take(1)))
    image_batch = image_batch.numpy()
    label_batch = label_batch.numpy()

    #code to fit model after building and compiling,and params made non-trainable
    EPOCHS = 8
    history = m.fit(train_batches,epochs = EPOCHS,verbose = 2,validation_data = validation_batches)

我在网上找到的解决方案与OpenCV或PhP有关，均不适用于我的代码。如果有人可以帮助我，我将不胜感激。谢谢！

系统信息- 操作系统-Linux 16.04