Получение проблемы при итерации данных mnist с использованием tenorflow

Question

Я получаю ошибку ValueError: Невозможно передать значение shape (100,) для Tensor 'Placeholder_31: 0', который имеет форму '(?, 10)' при выполнении данных mnist.

import tensorflow as tf
import numpy as np
from tensorflow import keras




mnist = tf.keras.datasets.mnist(train_images, train_labels), 
 (test_images, test_labels) = mnist.load_data()


train_images = np.array([image.flatten() for image in train_images])

test_images = np.array([image.flatten() for image in test_images])

train_images = train_images / 255

test_images = test_images / 255


print(train_images.shape[0])
print(train_images[0].shape[0])
print(len(train_images[0]))
print(train_labels.shape[0])
print(test_images.shape[0])
print(test_labels.shape[0])


# 10 Classes , 0 - 9


n_nodes_hdl_1 = 500
n_nodes_hdl_2 = 500
n_nodes_hdl_3 = 500


n_input = len(train_images[0])
n_classes = 10
batch_size = 100
learning_rate = 0.001


# height x, Shape: 28 * 28 matrix = 784
x = tf.placeholder(tf.float32,[None, n_input])
y = tf.placeholder(tf.float32,[None, n_classes])



 def neural_network_model(data):
  hidden_1_layer = 
{'weights':tf.Variable(tf.random_normal([n_input,n_nodes_hdl_1])),

'biases':tf.Variable(tf.random_normal([n_nodes_hdl_1]))}

  hidden_2_layer = 
  {'weights':tf.Variable(tf.random_normal
   ([n_nodes_hdl_1,n_nodes_hdl_2])),

   'biases':tf.Variable(tf.random_normal([n_nodes_hdl_2]))}

hidden_3_layer = {'weights':tf.Variable
(tf.random_normal([n_nodes_hdl_2,n_nodes_hdl_3])),
                  'biases':tf.Variable(tf.random_normal([n_nodes_hdl_3]))}

output_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hdl_3,n_classes])),
                  'biases':tf.Variable(tf.random_normal([n_classes]))}

# input_data x, weights w, biases b : w * x + b;

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_2_layer['biases'])
l3 = tf.nn.relu(l3)

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

return output

def train_neural_network(x):
prediction = neural_network_model(x)
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(logits=prediction,labels=y))    
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)

# cycles: feed forward + back prop
hm_epochs = 5

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

    for epoch in range(hm_epochs):
        epoch_loss = 0

        i = 0
        while i < len(train_images):                
            start = i
            end = i+batch_size

            epoch_x = np.array(train_images[start:end])
            epoch_y = np.array(train_labels[start:end])

            i,c = sess.run([optimizer,cost], feed_dict = {x:epoch_x, y:epoch_y})
            epoch_loss += c
            i +=batch_size

        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,tf.float32))
    print('Accuracy:',accuracy.eval({x:test_images, y:test_labels}))

 train_neural_network(x) 





     60000
        784
        784
        60000           
        10000
        10000
    ---------------------------------------------------------------------------


    ~/anaconda3/envs/LSTM-Human-Activity-Recognition/lib/python3.6/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata)
        927     try:
        928       result = self._run(None, fetches, feed_dict, options_ptr,
    --> 929                          run_metadata_ptr)
        930       if run_metadata:
        931         proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

    ~/anaconda3/envs/LSTM-Human-Activity-Recognition/lib/python3.6/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)
       1126                              'which has shape %r' %
       1127                              (np_val.shape, subfeed_t.name,
    -> 1128                               str(subfeed_t.get_shape())))
       1129           if not self.graph.is_feedable(subfeed_t):
       1130             raise ValueError('Tensor %s may not be fed.' % subfeed_t)

Я отправил по ошибке также вместе с кодом. Ошибка о типе Tensorflow, который он не принимает. ValueError Traceback (последний вызов был последним) в 134 отпечатка («Точность:», precision.eval ({x: test_images, y: test_labels})) 135 -> 136 train_neural_network (x) 137 138

    <ipython-input-21-f73f84d7c08f> in train_neural_network(x)
        124                 epoch_y = np.array(train_labels[start:end])
        125 
    --> 126                 i,c = sess.run([optimizer,cost], feed_dict = {x:epoch_x, y:epoch_y})
        127                 epoch_loss += c
        128                 i +=batch_size