Contents

require "dnn"
require "dnn/lib/cifar10"
#require "numo/linalg/autoloader"

include Numo
include DNN::Layers
include DNN::Activations
include DNN::Optimizers
Model = DNN::Model
CIFAR10 = DNN::CIFAR10

x_train, y_train = CIFAR10.load_train
x_test, y_test = CIFAR10.load_test

x_train = SFloat.cast(x_train).reshape(x_train.shape[0], 32, 32, 3).transpose(0, 3, 2, 1)
x_test = SFloat.cast(x_test).reshape(x_test.shape[0], 32, 32, 3).transpose(0, 3, 2, 1)

x_train /= 255
x_test /= 255

y_train = DNN::Util.to_categorical(y_train, 10)
y_test = DNN::Util.to_categorical(y_test, 10)

model = Model.new

model << InputLayer.new([3, 32, 32])

model << Conv2D.new(16, 5, 5)
model << BatchNormalization.new
model << ReLU.new

model << Conv2D.new(16, 5, 5)
model << BatchNormalization.new
model << ReLU.new

model << MaxPool2D.new(2, 2)

model << Conv2D.new(32, 5, 5)
model << BatchNormalization.new
model << ReLU.new

model << Conv2D.new(32, 5, 5)
model << BatchNormalization.new
model << ReLU.new

model << Flatten.new

model << Dense.new(512)
model << BatchNormalization.new
model << ReLU.new
model << Dropout.new(0.5)

model << Dense.new(10)
model << SoftmaxWithLoss.new

model.compile(Adam.new)

model.train(x_train, y_train, 20, batch_size: 100) do
  model.test(x_test, y_test)
end

model.save("trained_cifar10.marshal")

Version data entries

8 entries across 8 versions & 1 rubygems

Version	Path
ruby-dnn-0.1.8	examples/cifar10_example.rb
ruby-dnn-0.1.7	examples/cifar10_example.rb
ruby-dnn-0.1.6	examples/cifar10_example.rb
ruby-dnn-0.1.5	examples/cifar10_example.rb
ruby-dnn-0.1.4	examples/cifar10_example.rb
ruby-dnn-0.1.3	examples/cifar10_example.rb
ruby-dnn-0.1.2	examples/cifar10_example.rb
ruby-dnn-0.1.1	examples/cifar10_example.rb