How to extract feature vectors in a fine-tuned network in keras

Question

I am trying to extract feature vectors from an added Dense layer after fine tuning the Inception v3 CNN on keras with new data. Basically, I load the network structure and its weights, add two dense layers (my data is for a two class problem) and update weights from only some part of the network as the code below shows:

# create the base pre-trained model
base_model = InceptionV3(weights='imagenet', include_top=False)

# add a global spatial average pooling layer
x = base_model.output
x = GlobalAveragePooling2D()(x)

# let's add a fully-connected layer
x = Dense(64, activation='relu')(x)

# and a logistic layer -- I have 2 classes only
predictions = Dense(2, activation='softmax')(x)

# this is the model to train
model = Model(inputs=base_model.input, outputs=predictions)

# first: train only the top layers (which were randomly initialized)
# i.e. freeze all convolutional InceptionV3 layers

for layer in base_model.layers:
        layer.trainable = False

# compile the model (should be done *after* setting layers to non-trainable)
model.compile(optimizer='rmsprop', loss='categorical_crossentropy')

#load new training data
x_train, x_test, y_train, y_test =load_data(train_data, test_data, train_labels, test_labels)

datagen = ImageDataGenerator()      
datagen.fit(x_train)

epochs=1
batch_size=32

# train the model on the new data for a few epochs
model.fit_generator(datagen.flow(x_train, y_train,
                                 batch_size=batch_size),
                                 steps_per_epoch=x_train.shape[0] // 
                                 batch_size,
                                 epochs=epochs,
                                 validation_data=(x_test, y_test))

# at this point, the top layers are well trained and 
#I can start fine-tuning convolutional layers from inception V3. 
#I will freeze the bottom N layers and train the remaining top layers. 
#I chose to train the top 2 inception blocks, i.e. I will freeze the 
#first 249 layers and unfreeze the rest:

for layer in model.layers[:249]:
    layer.trainable = False
for layer in model.layers[249:]:
    layer.trainable = True

# I need to recompile the model for these modifications to take effect
# I use SGD with a low learning rate
from keras.optimizers import SGD
model.compile(optimizer=SGD(lr=0.0001, momentum=0.9), loss='categorical_crossentropy', metrics=['binary_accuracy'])

# I train our model again (this time fine-tuning the top 2 inception blocks alongside the top Dense layers
model.fit_generator(datagen.flow(x_train, y_train,
                                 batch_size=batch_size),
                                 steps_per_epoch=x_train.shape[0] // 
                                 batch_size,
                                 epochs=epochs,
                                 validation_data=(x_test, y_test))

This code runs perfectly well and its not my problem.

My problem is that, after fine tuning this network I want the output from the last but one layer on my train and test data because I want to use this new network as a feature extractor. I want the output from this part of the network that you can see in the code above:

x = Dense(64, activation='relu')(x)

I tried the following code but it does not work:

 from keras import backend as K
 inputs = [K.learning_phase()] + model.inputs
 _convout1_f = K.function(inputs, model.get_layer(dense_1).output)

The error is the following

 _convout1_f = K.function(inputs, model.get_layer(dense_1).output)
 NameError: global name 'dense_1' is not defined

How can I extract features from the new layer I added after fine tuning a pre-trained network in my new data? what I did wrong here?

How to extract feature vectors in a fine-tuned network in keras

Answers (1)

Related Questions