nasnet: NASNet-A model

In stschn/deepANN: Neural Network Toolbox

NASNet-A model

Description

NASNet-A model

Usage

nasnet(
  include_top = TRUE,
  weights = "imagenet",
  input_tensor = NULL,
  input_shape = NULL,
  classes = 1000,
  classifier_activation = "softmax",
  default_size = NULL,
  penultimate_filters = 4032,
  num_blocks = 6,
  stem_block_filters = 96,
  skip_reduction = TRUE,
  filter_multiplier = 2
)

Arguments

include_top	Whether to include the fully-connected layer at the top of the network. A model without a top will output activations from the last convolutional or pooling layer directly.
weights	One of NULL (random initialization), 'imagenet' (pre-trained weights), an array, or the path to the weights file to be loaded.
input_tensor	Optional tensor to use as image input for the model.
input_shape	Dimensionality of the input not including the samples axis.
classes	Optional number of classes or labels to classify images into, only to be specified if include_top = TRUE.
classifier_activation	A string or callable for the activation function to use on top layer, only if include_top = TRUE.
default_size	Specifies the default image size of the model. If no value is specified (default) the size is set equal to 331 for NASNetLarge. For NASNetMobile the default size is 224.
penultimate_filters	Number of filters in the penultimate layer.
num_blocks	Number of repeated blocks of the NASNet model.
stem_block_filters	Number of filters in the initial stem block.
skip_reduction	Whether to skip the reduction step at the tail end of the network.
filter_multiplier	Controls the width of the network. if filter_multiplier < 1.0, proportionally decreases the number of filters in each layer. if filter_multiplier > 1.0, proportionally increases the number of filters in each layer. if filter_multiplier = 1.0, default number of filters from the paper are used at each layer.

Details

The input shape is usually c(height, width, channels) for a 2D image. If no input shape is specified the default_size is used.
The number of classes can be computed in three steps. First, build a factor of the labels (classes). Second, use as_CNN_image_Y to one-hot encode the outcome created in the first step. Third, use nunits to get the number of classes. The result is equal to nlevels used on the result of the first step.

For a n-ary classification problem with single-label associations, the output is either one-hot encoded with categorical_crossentropy loss function or binary encoded (0,1) with sparse_categorical_crossentropy loss function. In both cases, the output activation function is softmax.
For a n-ary classification problem with multi-label associations, the output is one-hot encoded with sigmoid activation function and binary_crossentropy loss function.

NASNet models use the notation NASNet (N @ P) where N is the number of blocks and P is the number of penultimate filters.

The current parameter defaults are the values for the large NASNet model type. The parameter values for the the mobile NASNet model type are:
penultimate_filters = 1056
num_blocks = 4
stem_block_filters = 32
skip_reduction = FALSE

For a task with another top layer block, e.g. a regression problem, use the following code template:

base_model <- nasnet(include_top = FALSE)
base_model$trainable <- FALSE
outputs <- base_model$output %>%
layer_flatten()
layer_dense(units = 1, activation = "linear")
model <- keras_model(inputs = base_model$input, outputs = outputs)

inputs <- layer_input(shape = c(256, 256, 3))
blocks <- inputs %>%
layer_conv_2d_transpose(filters = 3, kernel_size = c(1, 1)) %>%
layer_max_pooling_2d()
model <- nasnet(input_tensor = blocks)

Value

A CNN model object from type NASNet-A.

References

Zoph, B., Vasudevan, V., Shlens, J., Le, Q. V. (2017). Learning Transferable Architectures for Scalable Image Recognition. arXiv:1707.07012 cs. https://arxiv.org/abs/1707.07012.
https://arxiv.org/pdf/1707.07012.pdf

see also https://github.com/keras-team/keras-applications/blob/master/keras_applications/nasnet.py

See Also

Other Convolutional Neural Network (CNN): alexnet(), as_CNN_image_X(), as_CNN_image_Y(), as_CNN_temp_X(), as_CNN_temp_Y(), as_images_array(), as_images_tensor(), images_load(), images_resize(), inception_resnet_v2(), inception_v3(), lenet5(), mobilenet(), mobilenet_v2(), mobilenet_v3(), resnet, unet(), vgg, xception(), zfnet()

stschn/deepANN documentation built on June 25, 2024, 7:27 a.m.