Training Keras model with multiple inputs

Question

I want to train a model with 3 different inputs using keras. The training data - x_train , left_train , right_train is of shape (10000,83,12). Here is part of the code.

from keras.layers import Dense, Input, LSTM
...
x = Input(shape = (83,12), dtype = "float32")
left = Input(shape = (83,12), dtype = "float32")
right = Input(shape = (83,12), dtype = "float32")
...
model = Model(inputs = [x, left, right], outputs = output)
model.compile(optimizer = "adadelta", loss = "categorical_crossentropy", metrics = ["accuracy"])
model.fit([x_train, left_train, right_train], y_train, validation_data=(x_test, y_test), epochs=20, batch_size=128)
...

I am getting the following error while training:

ValueError                       Traceback (most recent call last)
<ipython-input-17-261d36872e91> in <module>()
     51 
     52 
---> 53 model.fit([x_train, left, right], y_train, validation_data= 
(x_test, y_test), epochs=20, batch_size=128)
     54 
     55 scores = model.evaluate(x_test, y_test)
     ...
     ValueError: Error when checking model input: the list of 
Numpy arrays that you are passing to your model is not the size the 
model expected. Expected to see 3 array(s), but instead got the 
following list of 1 arrays: [array(...

I do pass a list of 3 inputs when calling fit method. What's the problem?

Answer 1

The validation_data and model.evaluate needs to be multi-input as well. In your case you only provide a single array (x_test, y_test) and just x_test where it would be something similar to ([x_test, left_test, right_test], y_test) . Essentially, the validation data needs to have same number of inputs / outputs as the training data.