neural network - How to fit data from InceptionV3 to ImageDataGenerator -

how fit data inceptionv3 imagedatagenerator?

the examples found fitting data imagedatagenerator mnist or cifar10, this:

(x_train, y_train), (x_test, y_test) = mnist.load_data() # fit parameters data datagen.fit(x_train) 

but can fit data inceptionv3 model imagedatagenerator?

i load inception v3 model like:

base_model = inceptionv3(weights='imagenet', include_top=true) datagen = imagedatagenerator(...) datagen.fit(base_model.get_layer('avg_pool').output) 

but error saying 'valueerror: setting array element sequence.'

i assume need in 2 steps. first feed in data inceptionv3 model , saving output numpy array. feeding numpy array second model.

first step (taken here):

generator = datagen.flow_from_directory(     'data/train',     target_size=(150, 150),     batch_size=batch_size,     class_mode=none,  # means our generator yield batches of data, no labels     shuffle=false)  # our data in order  bottleneck_features_train = model.predict_generator(generator, 2000)  np.save(open('bottleneck_features_train.npy', 'w'),  

bottleneck_features_train)