Using Pre-Trained Models

Applications

Keras Applications are deep learning models that are made available alongside pre-trained weights. These models can be used for prediction, feature extraction, and fine-tuning.

Weights are downloaded automatically when instantiating a model. They are stored at ~/.keras/models/.

The following image classification models (with weights trained on ImageNet) are available:

All of these architectures are compatible with all the backends (TensorFlow, Theano, and CNTK), and upon instantiation the models will be built according to the image data format set in your Keras configuration file at ~/.keras/keras.json. For instance, if you have set image_data_format=channels_last, then any model loaded from this repository will get built according to the TensorFlow data format convention, “Height-Width-Depth”.

Usage Examples

Classify ImageNet classes with ResNet50

  class_name class_description      score
1  n02504013   Indian_elephant 0.90117526
2  n01871265            tusker 0.08774310
3  n02504458  African_elephant 0.01046011

Extract features with VGG16

Extract features from an arbitrary intermediate layer with VGG19

Fine-tune InceptionV3 on a new set of classes

# create the base pre-trained model
base_model <- application_inception_v3(weights = 'imagenet', include_top = FALSE)

# add our custom layers
predictions <- base_model$output %>% 
  layer_global_average_pooling_2d() %>% 
  layer_dense(units = 1024, activation = 'relu') %>% 
  layer_dense(units = 200, activation = 'softmax')

# this is the model we will train
model <- keras_model(inputs = base_model$input, outputs = predictions)

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

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

# train the model on the new data for a few epochs
model %>% fit_generator(...)

# at this point, the top layers are well trained and we can start fine-tuning
# convolutional layers from inception V3. We will freeze the bottom N layers
# and train the remaining top layers.

# let's visualize layer names and layer indices to see how many layers
# we should freeze:
layers <- base_model$layers
for (i in 1:length(layers))
  cat(i, layers[[i]]$name, "\n")

# we chose to train the top 2 inception blocks, i.e. we will freeze
# the first 172 layers and unfreeze the rest:
freeze_weights(base_model, from = 1, to = 172)
unfreeze_weights(base_model, from = 173)

# we need to recompile the model for these modifications to take effect
# we use SGD with a low learning rate
model %>% compile(
  optimizer = optimizer_sgd(lr = 0.0001, momentum = 0.9), 
  loss = 'categorical_crossentropy'
)

# we train our model again (this time fine-tuning the top 2 inception blocks
# alongside the top Dense layers
model %>% fit_generator(...)

Build InceptionV3 over a custom input tensor

Additional examples

The VGG16 model is the basis for the Deep dream Keras example script.