epochsのError？？？

(image_train,label_train),(image_test,label_test)=mnist.load_data()
print('Image\n',image_train[0])
print("画像データの要素数",image_train.shape)
print("ラベルデータの要素数",label_train.shape)

#ラベルと画像データを表示
for i in range(0,10):
 print("ラベル",label_train[i])
 plt.imshow(image_train[i].reshape(28,28))
 plt.show()

print('Train: num of imeges, width,height:',image_train.shape)
print('Test: num of imeges, width,height:',image_test.shape)
image_train=image_train.reshape(60000,784)
image_train=image_train/255.0
image_test=image_test.reshape(10000,784)
image_test=image_test/255.0

print('Train: num of labeled images:',label_train.shape)
print('Test: num of labeled images:',label_test.shape)

for i in range(0,9):
  print('Train Case:',i,'Label:',label_train[i])
for i in range(0,9):
  print('Test Case:',i,'Label:',label_test[i])

import numpy as np
from tensorflow.python.keras.utils.np_utils import to_categorical
from matplotlib import pyplot as plt
label_train=to_categorical(label_train,10)
label_test=to_categorical(label_test,10)
print('Onehot Label:',label_train[7])
print('Onehot Label:',label_test[1])

from tensorflow.python.keras.models import Sequential
from tensorflow.python.keras.layers import Dense

model=Sequential()

model.add(
    Dense(units=64,input_shape=(748,),activation='relu')
)
model.add(
    Dense(units=10, activation='softmax')
)

model.compile(
	optimizer='adam',
	loss='categorical_crossentropy',
	metrics=['accuracy']
)

history_adam=model.fit(
	image_train,label_train,batch_size=32,
	epochs=10,validation_split=0.2
)

MNISTの画像で画像分類のコードを作成しています．
エラーコードでどうしても原因がわからないので教えてください．

ValueError Traceback (most recent call last)
<ipython-input-11-98660863e952> in <module>()
54 history_adam=model.fit(
55 image_train,label_train,batch_size=32,
---> 56 epochs=10,validation_split=0.2
57 )

ValueError: Input 0 of layer sequential_6 is incompatible with the layer: expected axis -1 of input shape to have value 748 but received input with shape [32, 784]