Question

我一直在尝试制作一个简单的2层神经网络。我研究了tensorflow api和官方教程，我制作了一个分层模型，但在神经网络中遇到了麻烦。以下是导致错误的代码部分：

package LectureLoops;

import java.text.NumberFormat;
import java.util.Scanner;

public class ValidatingInput {

    static Scanner sc = new Scanner(System.in);
    static NumberFormat cf = NumberFormat.getCurrencyInstance();

    public static void main(String[] args) {

        int bank = 10000;
        int bet;
        int max = 9000;
        int deposit;
        cf = NumberFormat.getCurrencyInstance();


        do {
            System.out.print("Please make a deposit: ");
            deposit = sc.nextInt();
            if ((deposit > bank))
                System.out.println("Insufficient funds, please enter a valid amount");
        }// end of do
        while (deposit > bank);
        System.out.println("You have successfully made a deposit of " + cf.format(deposit));
        bank = bank - deposit;
        System.out.println("Your new bank balance is " + cf.format(bank));

        do {
            System.out.print("Enter your bet: ");
            bet = sc.nextInt();
            if ((bet > deposit))
                System.out.println("Insufficient funds please enter a valid amount or add more funds" + "\nYour current allowance is: " + cf.format(deposit));
            if ((bet > max) || (bet <= 0))
                System.out.println("Maximum bet is " + cf.format(max) + " you entered " + cf.format(bet) + "\nPlease enter a valid amount");
        }// end of do
        while ((bet <= 0) || (bet > max || bet > deposit));
        System.out.print("Thank you for your bet of " + cf.format(bet) + " Your money is good here!\n");
        deposit = deposit - bet;
        System.out.println("You have " + cf.format(deposit) + " left to spend");
    }// psvm end

}// validatinginput end

错误是：

with graph.as_default():
    tf_train_dataset = tf.placeholder(tf.float32, shape=(batch_size, image_size * image_size))
    tf_train_labels = tf.placeholder(tf.int32, shape=(batch_size, num_labels))
    tf_valid_dataset = tf.constant(valid_dataset)
    tf_test_dataset = tf.constant(test_dataset)

    weights0 = tf.Variable(tf.truncated_normal([image_size**2, num_labels]))
    biases0 = tf.Variable(tf.zeros([num_labels]))

    hidden1 = tf.nn.relu(tf.matmul(tf_test_dataset, weights0) + biases0)

    weights1 = tf.Variable(tf.truncated_normal([num_labels, image_size * image_size]))
    biases1 = tf.Variable(tf.zeros([image_size**2]))

    hidden2 = tf.nn.relu(tf.matmul(hidden1, weights1) + biases1)


    logits = tf.matmul(hidden2, weights0) + biases0

    labels = tf.expand_dims(tf_train_labels, 1)

    indices = tf.expand_dims(tf.range(0, batch_size), 1)

    concated = tf.concat(1, [indices, tf.cast(labels,tf.int32)])

    onehot_labels = tf.sparse_to_dense(concated, tf.pack([batch_size, num_labels]), 1.0, 0.0)


    loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits, onehot_labels))

    optimizer = tf.train.GradientDescentOptimizer(0.5).minimize(loss)

    train_prediction = tf.nn.softmax(logits)
    valid_prediction = tf.nn.softmax(tf.matmul(tf.nn.relu(tf.matmul(tf.nn.relu(tf.matmul(tf_valid_dataset,weights0) + biases0),weights1)+biases1),weights0)+biases0)
    test_prediction = tf.nn.softmax(tf.matmul(tf.nn.relu(tf.matmul(tf.nn.relu(tf.matmul(tf_test_dataset,weights0) + biases0),weights1)+biases1),weights0)+biases0)

以下是完整代码：http://pastebin.com/sX7RqbAf

我使用过TensorFlow和Python 2.7。我对神经网络和机器学习都很陌生，所以请原谅我任何错误，提前谢谢。

Answer 1

在你的例子中：

tf_train_labels的形状为[batch_size, num_labels]
因此labels的形状为[batch_size, 1, num_labels]
indices的形状为[batch_size, 1]

因此，当你写：

concated = tf.concat(1, [indices, tf.cast(labels,tf.int32)])

它会引发错误，因为labels和indices的第三个维度不同。 labels的第三维大小为num_labels（大概为10），indices没有第三维。

TensorFlow concat中的秩不匹配错误

1 个答案: