如何将TensorFlow数据集API与密集层结合使用

时间:2017-09-14 14:11:26

标签: python tensorflow deep-learning

我正在尝试使用TensorFlow documentation中显示的输入管道的数据集API,并使用几乎相同的代码:

tr_data = Dataset.from_tensor_slices((train_images, train_labels))
tr_data = tr_data.map(input_parser, NUM_CORES, output_buffer_size=2000)
tr_data = tr_data.batch(BATCH_SIZE)
tr_data = tr_data.repeat(EPOCHS)

iterator = dataset.make_one_shot_iterator()
next_example, next_label = iterator.get_next()

# Script throws error here
loss = model_function(next_example, next_label)

with tf.Session(...) as sess:
    sess.run(tf.global_variables_initializer())

     while True:
        try:
            train_loss = sess.run(loss)
        except tf.errors.OutOfRangeError:
            print("End of training dataset.")
            break

这应该更快,因为它避免使用慢的feed_dicts。但我无法使用我的模型,这是一个简化的LeNet架构。 问题是我tf.layers.dense中的model_function(),它需要一个已知的输入形状(我猜是因为它必须事先知道权重的数量)。但next_examplenext_label只能通过在会话中运行它们来获得它们的形状。在评估它们之前,它们的形状是未定义的?

声明model_function()会引发此错误:

  

ValueError:Dense输入的最后一个维度应该是   定义。找到None

目前,我不知道我是否以预期的方式使用此数据集API,或者是否有解决方法。

提前致谢!

编辑1: 下面是我的模型,它会在第一个密集层

处抛出错误
def conv_relu(input, kernel_shape):
    # Create variable named "weights".
    weights = tf.get_variable("weights", kernel_shape,
        initializer=tf.random_normal_initializer())
    # Create variable named "biases".
    biases = tf.get_variable("biases", kernel_shape[3],
        initializer=tf.constant_initializer(0.0))
    conv = tf.nn.conv2d(input, weights,
        strides=[1, 1, 1, 1], padding='VALID')
    return tf.nn.relu(conv + biases)

def fully(input, output_dim):
    assert len(input.get_shape())==2, 'Wrong input shape, need flattened tensor as input'
    input_dim = input.get_shape()[1]

    weight = tf.get_variable("weight", [input_dim, output_dim],
        initializer=tf.random_normal_initializer())
    bias = tf.get_variable('bias', [output_dim],
        initializer=tf.random_normal_initializer())

    fully = tf.nn.bias_add(tf.matmul(input, weight), bias)
    return fully


def simple_model(x):

    with tf.variable_scope('conv1'):
        conv1 = conv_relu(x, [3,3,1,10])
        conv1 = tf.nn.max_pool(conv1,[1,2,2,1],[1,2,2,1],'SAME')

    with tf.variable_scope('conv2'):
        conv2 = conv_relu(conv1, [3,3,10,10])
        conv2 = tf.nn.max_pool(conv2,[1,2,2,1],[1,2,2,1],'SAME')

    with tf.variable_scope('conv3'):
        conv3 = conv_relu(conv2, [3,3,10,10])
        conv3 = tf.nn.max_pool(conv3,[1,2,2,1],[1,2,2,1],'SAME')

    flat = tf.contrib.layers.flatten(conv3)
    with tf.variable_scope('fully1'):
        fully1 = tf.layers.dense(flat, 1000)
        fully1 = tf.nn.relu(fully1)

    with tf.variable_scope('fully2'):
        fully2 = tf.layers.dense(fully1, 100)
        fully2 = tf.nn.relu(fully2)

    with tf.variable_scope('output'):
        output = tf.layers.dense(fully2, 4)
        fully1 = tf.nn.relu(output)


    return output

编辑2:

在这里,您可以看到张量的打印件。请注意,next_example没有形状

  

next_example:Tensor(“IteratorGetNext:0”,dtype = float32)
  next_label:Tensor(“IteratorGetNext:1”,shape =(?,4),dtype = float32)

1 个答案:

答案 0 :(得分:3)

我自己找到了答案。

遵循此thread,如果您事先了解图像尺寸,则只需使用tf.Tensor.set_shape设置形状。

def input_parser(img_path, label):

    # read the img from file
    img_file = tf.read_file(img_path)
    img_decoded = tf.image.decode_image(img_file, channels=1)
    img_decoded = tf.image.convert_image_dtype(img_decoded, dtype=tf.float32)
    img_decoded.set_shape([90,160,1]) # This line was missing

    return img_decoded, label

如果tensorflow文档包含这一行,那就太好了。