This question is asked several times here on SE, but I havent been able to find the right answer. I'm trying to build a network with 1 input and 2 outputs. I don't have a lot of data so I would like to use a generator for augmentation (preferably with imgaug).
My code:

seq = iaa.Sequential([ ....

gen = ImageDataGenerator(preprocessing_function=seq.augment_image)  
batch_size = 64

def generate_data_generator(generator, X, Y1, Y2):
    genX = gen.flow(X, batch_size=batch_size, seed=42)
    genY1 = gen.flow(Y1, batch_size=batch_size, seed=42)
    while True:
            Xi = genX.next()
            Yi1 = genY1.next()
            Yi2 = function(Y2)
            yield Xi, [Yi1, Yi2]

H = model.fit_generator(generate_data_generator(gen, trainX, trainY1, trainY2),
                steps_per_epoch=len(trainX) // batch_size,
                validation_data=(testX, [testY1, testY2]))

With this, I get the error:

'ValueError: ('Input data in `NumpyArrayIterator` should have rank 4. You passed an array with shape', (115, 16))' 

a normal fit like this works just fine, so I don't there is anything wrong with the normal input

H = model.fit(trainX, {"output1": trainY1, "output2": trainY2},
    validation_data=(testX, {"output1": testY1, "output2": testY2}),

This is my input, I have no idea how I get a shape of (115,16) here


(115, 158, 100, 3)

thank you


You should pass X and Y collectively to the ImageDataGenerator.flow() method.

Please refer to this answer in case you are looking for a multi-output classification model using ImageDataGenerator in Keras.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.