TensorFlow ValueError求助：Shape需为1阶但实际为0阶

Let's break down the issues in your code step by step—there are several small but critical mistakes causing the shape error and other potential failures:

Your Original Code & Error #

Code #

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/", one_hot=True)
n_nodes_hl1 = 500
n_nodes_hl2 = 500
n_nodes_hl3 = 500
n_classes = 0
batch_size = 100
x = tf.placeholder('float',[None, 784])
y = tf.placeholder('float')
def neural_model(impuls):
    hidden_1_layer = {'weights':tf.Variable(tf.random_normal([784, n_nodes_hl1])), 'biases':tf.Variable(tf.random_normal(n_nodes_hl1))}
    hidden_2_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl1, n_nodes_hl2])), 'biases':tf.Variable(tf.random_normal(n_nodes_hl2))}
    hidden_3_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl2, n_nodes_hl3])), 'biases':tf.Variable(tf.random_normal(n_nodes_hl3))}
    output_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl3, n_classes])), 'biases':tf.Variable(tf.random_normal(n_classes))}
    l1 = tf.add(tf.matmul(data, hidden_1_layer['weights']) + hidden_1_layer['biases'])
    l1 = tf.nn.relu(l1)
    l2 = tf.add(tf.matmul(l1, hidden_2_layer['weights']) + hidden_2_layer['biases'])
    l2 = tf.nn.relu(l2)
    l3 = tf.add(tf.matmul(l2, hidden_3_layer['weights']) + hidden_3_layer['biases'])
    l3 = tf.nn.relu(l3)
    output = tf.matmul(l3, output_layer['weights']) + output_layer['biases']
    return output
def train_neural_network(x):
    prediction = neural_model(x)
    cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction, labels=y))
    optimizer = tf.train.AdamOptimizer().minimize(cost)
    hm_epochs = 10
    with tf.Session as sess:
        sess.run(tf.global_variables_initializer())
        for epoch in hm_epochs:
            epoch_loss = 0
            for _ in range(int(mnist.train.num_examples/batch_size)):
                x, y = mnist.train.next_batch(batch_size)
                _, c = sess.run([optimizer, cost], feed_dict={x: x, y:y})
                epoch_loss += c
            print('Epoch: ', epoch, 'completed out of', hm_epochs, 'loss: ', epoch_loss)
        correct = tf.equal(tf.argmax(prediction, 1), tf.argmax(y,1))
        ac = tf.reduce_mean(tf.cast(correct, 'float'))
        print('acc: ', ac.eval({x:mnist.test_images, y:mnist.test_labels}))
train_neural_network(x)

Error Traceback #

Traceback (most recent call last):
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/common_shapes.py", line 686, in _call_cpp_shape_fn_impl
    input_tensors_as_shapes, status)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/errors_impl.py", line 473, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: Shape must be rank 1 but is rank 0 for 'random_normal_1/RandomStandardNormal' (op: 'RandomStandardNormal') with input shapes: []
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
  File "a.py", line 60, in <module>
    train_neural_network(x)
  File "a.py", line 39, in train_neural_network
    prediction = neural_model(x)
  File "a.py", line 17, in neural_model
    'biases':tf.Variable(tf.random_normal(n_nodes_hl1))}
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/random_ops.py", line 76, in random_normal
    shape_tensor, dtype, seed=seed1, seed2=seed2)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/gen_random_ops.py", line 420, in _random_standard_normal
    seed2=seed2, name=name)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
    op_def=op_def)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 2958, in create_op
    set_shapes_for_outputs(ret)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 2209, in set_shapes_for_outputs
    shapes = shape_func(op)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 2159, in call_with_requiring
    return call_cpp_shape_fn(op, require_shape_fn=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/common_shapes.py", line 627, in call_cpp_shape_fn
    require_shape_fn)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/common_shapes.py", line 691, in _call_cpp_shape_fn_impl
    raise ValueError(err.message)
ValueError: Shape must be rank 1 but is rank 0 for 'random_normal_1/RandomStandardNormal' (op: 'RandomStandardNormal') with input shapes: []

Fixes for All Issues #

1. Direct Cause of the Shape Error #

tf.random_normal() requires a rank-1 shape tensor (a list/tuple defining the output shape), but you passed a scalar integer. Fix this for all bias initializations:

# Wrong: scalar input causes rank 0 error
tf.random_normal(n_nodes_hl1)
# Correct: pass a list to define a 1D shape
tf.random_normal([n_nodes_hl1])

2. Incorrect Number of Classes #

MNIST has 10 handwritten digit classes (0-9), but you set n_classes = 0—this breaks the output layer and softmax calculation:

n_classes = 10

3. Mismatched Parameter Name #

Your neural_model function accepts impuls but references data (an undefined variable). Fix the parameter name to match:

def neural_model(data):  # Changed from 'impuls' to 'data'
    # ... rest of the function uses 'data' correctly

4. Redundant tf.add() Calls #

You're using both tf.add() and the + operator together, which is unnecessary. Replace lines like:

l1 = tf.add(tf.matmul(data, hidden_1_layer['weights']) + hidden_1_layer['biases'])

With either:

# Option 1: Use + operator
l1 = tf.matmul(data, hidden_1_layer['weights']) + hidden_1_layer['biases']
# Option 2: Use tf.add() properly
l1 = tf.add(tf.matmul(data, hidden_1_layer['weights']), hidden_1_layer['biases'])

5. Missing Parentheses for tf.Session() #

Your session initialization is missing parentheses, which will throw an error when starting the session:

# Wrong
with tf.Session as sess:
# Correct
with tf.Session() as sess:

6. Iterating Over an Integer #

hm_epochs = 10 is an integer, so for epoch in hm_epochs: will throw a TypeError. Use range() instead:

for epoch in range(hm_epochs):

7. Variable Name Collision #

You're overwriting the placeholder variables x and y when loading batches, breaking your feed_dict. Rename the batch variables:

# Wrong: overwrites placeholders
x, y = mnist.train.next_batch(batch_size)
# Correct: use unique names
batch_x, batch_y = mnist.train.next_batch(batch_size)
# Update feed_dict accordingly
_, c = sess.run([optimizer, cost], feed_dict={x: batch_x, y: batch_y})

8. Incorrect Test Data Access #

You used mnist.test_images instead of the correct mnist.test.images (the MNIST dataset uses dot notation for test data):

print('Accuracy: ', accuracy.eval({x:mnist.test.images, y:mnist.test.labels}))

Corrected Full Code #

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/", one_hot=True)

n_nodes_hl1 = 500
n_nodes_hl2 = 500
n_nodes_hl3 = 500
n_classes = 10
batch_size = 100

x = tf.placeholder('float',[None, 784])
y = tf.placeholder('float')

def neural_model(data):
    hidden_1_layer = {
        'weights': tf.Variable(tf.random_normal([784, n_nodes_hl1])),
        'biases': tf.Variable(tf.random_normal([n_nodes_hl1]))
    }
    hidden_2_layer = {
        'weights': tf.Variable(tf.random_normal([n_nodes_hl1, n_nodes_hl2])),
        'biases': tf.Variable(tf.random_normal([n_nodes_hl2]))
    }
    hidden_3_layer = {
        'weights': tf.Variable(tf.random_normal([n_nodes_hl2, n_nodes_hl3])),
        'biases': tf.Variable(tf.random_normal([n_nodes_hl3]))
    }
    output_layer = {
        'weights': tf.Variable(tf.random_normal([n_nodes_hl3, n_classes])),
        'biases': tf.Variable(tf.random_normal([n_classes]))
    }

    l1 = tf.matmul(data, hidden_1_layer['weights']) + hidden_1_layer['biases']
    l1 = tf.nn.relu(l1)

    l2 = tf.matmul(l1, hidden_2_layer['weights']) + hidden_2_layer['biases']
    l2 = tf.nn.relu(l2)

    l3 = tf.matmul(l2, hidden_3_layer['weights']) + hidden_3_layer['biases']
    l3 = tf.nn.relu(l3)

    output = tf.matmul(l3, output_layer['weights']) + output_layer['biases']
    return output

def train_neural_network(x):
    prediction = neural_model(x)
    cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction, labels=y))
    optimizer = tf.train.AdamOptimizer().minimize(cost)
    hm_epochs = 10

    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch in range(hm_epochs):
            epoch_loss = 0
            for _ in range(int(mnist.train.num_examples/batch_size)):
                batch_x, batch_y = mnist.train.next_batch(batch_size)
                _, c = sess.run([optimizer, cost], feed_dict={x: batch_x, y: batch_y})
                epoch_loss += c
            print('Epoch: ', epoch+1, 'completed out of', hm_epochs, 'loss: ', epoch_loss)
        
        correct = tf.equal(tf.argmax(prediction, 1), tf.argmax(y,1))
        accuracy = tf.reduce_mean(tf.cast(correct, 'float'))
        print('Accuracy: ', accuracy.eval({x:mnist.test.images, y:mnist.test.labels}))

train_neural_network(x)

内容的提问来源于stack exchange，提问作者M.Utku