CODEWITHSUNDEEP
pythonneural-networkkeras
Abrar Hasin
Abrar Hasin

Reputation: 41

AssertionError while trying to fit model python

I keep on running into an AssertionError when trying to fit a model. I did some reading on when Python raises an AssertionError. The backlog is as follows

File "G:/test3/main.py", line 167, in <module>
model.fit([images, captions], next_words, batch_size=128, epochs=50)

File "C:\Users\Acer\Anaconda3\lib\site-packages\keras\engine\training.py", line 950, in fit
batch_size=batch_size)

File "C:\Users\Acer\Anaconda3\lib\site-packages\keras\engine\training.py", line 671, in _standardize_user_data
self._set_inputs(x)

File "C:\Users\Acer\Anaconda3\lib\site-packages\keras\engine\training.py", line 575, in _set_inputs
assert len(inputs) == 1

AssertionError

My code is as follows 

model=Sequential()

model.add(Concatenate([image_model, language_model]))
model.add(LSTM(1000, return_sequences=False))
model.add(Dense(vocab_size))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer=Nadam(), metrics=['accuracy'])

model.fit([images, captions], next_words, batch_size=5, epochs=50)
model.summary()

model.save_weights("./models/vgg16_weights_tf_dim_ordering_tf_kernels.h5")

images has a shape of (18724,1000) and captions has a shape of (18724, 43)

Upvotes: 1

Views: 8683

Answers (1)

sdcbr
sdcbr

Reputation: 7129

You are getting this error because you did not specify any inputs to your model, and Keras is trying to set them on calling model.fit(). The assertion is there because each model wrapped in a Sequential container should take only one input.

To implement what you want, you probably want to go for Keras' Functional API instead of the Sequential API. Something along these lines:

from keras.models import Model
from keras.layers import Concatenate, Input, Dense

# First model
first_input = Input((2, ))
first_dense = Dense(128)(first_input)

# Second model
second_input = Input((10, ))
second_dense = Dense(64)(second_input)

# Concatenate both
merged = Concatenate()([first_dense, second_dense])
output_layer = Dense(1)(merged)

model = Model(inputs=[first_input, second_input], outputs=output_layer)
model.compile(optimizer='sgd', loss='mse')

Upvotes: 1

Related Questions