model.load_weightsでエラーがでる

うまくモデルの重みを読み込むことができません。
このようなエラーを調べてみたのですが、見つかりませんでした。
方法があったら教えていただきたいです。

Python
1  model = Sequential()
2  model.add(Dense(100, activation='Mish'))
3  model.add(Dense(100, activation='Mish'))
4  model.add(Dense(MAX_ANSWER+1,activation="softmax"))
5
6  model.compile(loss='categorical_crossentropy',optimizer='adamax',metrics=['accuracy'])
7  if flag:
8    pass
9  else:  
10    model.load_weights('transfer0.h5')          
11  model.fit(x_train, y_train, epochs=1, batch_size=1, validation_data=(x_test, y_test))
12  model.save_weights('transfer0.h5')    
13  predict_prob=model.predict(x_test)
14  predict_classes=np.argmax(predict_prob,axis=1)
15  true_classes = y_test1
16  print(confusion_matrix(true_classes, predict_classes))

<ipython-input-41-3d2d5d6f3f30> in learner(x_train, y_train, x_test, y_test, y_test1, flag)
    213     pass
    214   else:
--> 215     model.load_weights('transfer0.h5')
    216   model.fit(x_train, y_train, epochs=1, batch_size=1, validation_data=(x_test, y_test))
    217   model.save_weights('transfer0.h5')

/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py in load_weights(self, filepath, by_name, skip_mismatch, options)
   2198     if not self._is_graph_network and not self.built:
   2199       raise ValueError(
-> 2200           'Unable to load weights saved in HDF5 format into a subclassed '
   2201           'Model which has not created its variables yet. Call the Model '
   2202           'first, then load the weights.')

ValueError: Unable to load weights saved in HDF5 format into a subclassed Model which has not created its variables yet. Call the Model first, then load the weights.