Керас функциональный API ValueError - PullRequest
Новая
       5

Керас функциональный API ValueError

0 голосов
/ 23 февраля 2019

Я построил эту модель в последовательном режиме, и она работала нормально, я использую те же самые данные, у нее тот же тип и форма, что и в последовательной модели.Теперь я пытаюсь воссоздать его в функциональном API и получаю сообщение об ошибке.

train_x форма (68, 1, 50) train_y форма (68,)

Воткод, который я использую: 

inputs = Input(shape = (1, 50))
x = LSTM(4, return_sequences = True)(inputs)
x = LSTM(4, return_sequences = True)(x)
x = LSTM(4, return_sequences = True)(x)
x = LSTM(4, return_sequences = True)(x)
predictions = Dense(1, activation = 'relu')(x)
model = Model(inputs = inputs, outputs = predictions)
model.compile(optimizer = 'rmsprop', loss = 'mean_squared_error', metrics = ['accuracy'])
model.fit(x = train_x, y = train_y, batch_size = 17, epochs = 100)

сообщение об ошибке: 

ValueError                                Traceback (most recent call last)
<ipython-input-51-246abb1e3d55> in <module>()
      1 model = Model(inputs = inputs, outputs = predictions)
      2 model.compile(optimizer = 'rmsprop', loss = 'mean_squared_error', metrics = ['accuracy'])
----> 3 model.fit(x = train_x, y = train_y, batch_size = 17, epochs = 100)

C:\Python35\lib\site-packages\keras\engine\training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, **kwargs)
   1403             class_weight=class_weight,
   1404             check_batch_axis=False,
-> 1405             batch_size=batch_size)
   1406         # prepare validation data
   1407         if validation_data:

C:\Python35\lib\site-packages\keras\engine\training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, check_batch_axis, batch_size)
   1297                                     output_shapes,
   1298                                     check_batch_axis=False,
-> 1299                                     exception_prefix='model target')
   1300         sample_weights = _standardize_sample_weights(sample_weight,
   1301                                                      self._feed_output_names)

C:\Python35\lib\site-packages\keras\engine\training.py in _standardize_input_data(data, names, shapes, check_batch_axis, exception_prefix)
    119                                  ' to have ' + str(len(shapes[i])) +
    120                                  ' dimensions, but got array with shape ' +
--> 121                                  str(array.shape))
    122             for j, (dim, ref_dim) in enumerate(zip(array.shape, shapes[i])):
    123                 if not j and not check_batch_axis:

ValueError: Error when checking model target: expected dense_4 to have 3 dimensions, but got array with shape (68, 1)

Вот последовательная модель, которая работает для меня. 

model = Sequential()
model.add(LSTM(units = 3, return_sequences = True, input_shape = (1, 50)))
model.add(LSTM(units = 3, return_sequences=True))
model.add(LSTM(units = 3, return_sequences=True))
model.add(LSTM(units = 3, return_sequences=True))
model.add(LSTM(units = 3, return_sequences=True))
model.add(LSTM(units = 3, return_sequences=True))
model.add(LSTM(units = 3))
model.add(Dense(1, activation = 'relu'))
model.compile(loss = 'mean_squared_error', optimizer = 'adagrad', metrics = ['accuracy'])
model.fit(x = train_x, y = train_y, batch_size = 3, epochs = 100, validation_split = 0.2, shuffle = True)
Predictions = model.predict(x = test_x).reshape((29,))
error = (sum((test_y - Predictions)**2))**(1.0/2.0)

print (error)

...