طهور الخطأ ValueError: Error when checking model target: the list of Numpy arrays that you are passing to your model is not the size the model expected في Keras أثناء محاولة استخدام fit_generator

ظهور الخطأ التالي عندما أحاول استخدام fit_generator مع نموذجي:

print(x_train.shape) # (15000, 100, 100, 3)
print(x_test.shape) # (8708, 100, 100, 3)
print(y_train.shape) # (15000, 119)
print(y_test.shape) # (8708, 119)
im=ImageDataGenerator()
data=im.flow(x_train, y_train, 16)
# النموذج
from keras.preprocessing.image import ImageDataGenerator
from keras.layers import AveragePooling2D, MaxPooling2D, Flatten, Conv2D, ZeroPadding2D,Input, Dense, Activation
from keras import layers
from keras.models import Model
x_input = Input((100,100,3))
x = Conv2D(128, (5,5))(x_input)
x = Activation('tanh')(x)
x = MaxPooling2D((3, 3))(x)
x = Conv2D(32, (5,5))(x)
x = Activation('tanh')(x)
x = MaxPooling2D((3, 3))(x)
x = Conv2D(100, (3,3))(x)
x = Activation('tanh')(x)
x = MaxPooling2D((3, 3))(x)
x = Flatten()(x)
x = Dense(256, activation='tanh')(x)
x = Dense(100, activation='tanh')(x)
x = Dense(100, activation='tanh')(x)
output1 = Dense(117, activation='softmax')(x)
output2 = Dense(2, activation='softmax')(x)
model = Model(inputs=x_input, outputs=[output1, output2])
model.compile(optimizer='rmsprop', metrics=['acc'],loss=['categorical_crossentropy'])
model.fit_generator(data, steps_per_epoch=len(x_train) / 16,
                    epochs=5, validation_data=(x_test, y_test))
-------------------------------------------------------------------------------------------------------
ValueError: Error when checking model target: the list of Numpy arrays that you are passing to your model is not the size the model expected.