Class: DNN::Layers::Conv2D

Inherits:
Connection show all
Includes:
Conv2DModule
Defined in:
lib/dnn/core/cnn_layers.rb

Instance Attribute Summary collapse

Attributes inherited from Connection

#l1_lambda, #l2_lambda

Attributes inherited from HasParamLayer

#params, #trainable

Class Method Summary collapse

Instance Method Summary collapse

Methods inherited from Connection

#dlasso, #dridge, #lasso, #ridge

Methods inherited from HasParamLayer

#update

Methods inherited from Layer

#built?, #prev_layer

Constructor Details

#initialize(num_filters, filter_size, weight_initializer: Initializers::RandomNormal.new, bias_initializer: Initializers::RandomNormal.new, strides: 1, padding: false, l1_lambda: 0, l2_lambda: 0) ⇒ Conv2D

Returns a new instance of Conv2D.



70
71
72
73
74
75
76
77
78
79
80
81
82
83
 
# File 'lib/dnn/core/cnn_layers.rb', line 70

def initialize(num_filters, filter_size,
               weight_initializer: Initializers::RandomNormal.new,
               bias_initializer: Initializers::RandomNormal.new,
               strides: 1,
               padding: false,
               l1_lambda: 0,
               l2_lambda: 0)
  super(weight_initializer: weight_initializer, bias_initializer: bias_initializer,
        l1_lambda: l1_lambda, l2_lambda: l2_lambda)
  @num_filters = num_filters
  @filter_size = filter_size.is_a?(Integer) ? [filter_size, filter_size] : filter_size
  @strides = strides.is_a?(Integer) ? [strides, strides] : strides
  @padding = padding
end

Instance Attribute Details

#filter_size ⇒ Object (readonly)

Returns the value of attribute filter_size.



67
68
69
 
# File 'lib/dnn/core/cnn_layers.rb', line 67

def filter_size
  @filter_size
end

#num_filters ⇒ Object (readonly)

Returns the value of attribute num_filters.



66
67
68
 
# File 'lib/dnn/core/cnn_layers.rb', line 66

def num_filters
  @num_filters
end

#strides ⇒ Object (readonly)

Returns the value of attribute strides.



68
69
70
 
# File 'lib/dnn/core/cnn_layers.rb', line 68

def strides
  @strides
end

Class Method Details

.load_hash(hash) ⇒ Object 



85
86
87
88
89
90
91
92
93
 
# File 'lib/dnn/core/cnn_layers.rb', line 85

def self.load_hash(hash)
  Conv2D.new(hash[:num_filters], hash[:filter_size],
             weight_initializer: Util.load_hash(hash[:weight_initializer]),
             bias_initializer: Util.load_hash(hash[:bias_initializer]),
             strides: hash[:strides],
             padding: hash[:padding],
             l1_lambda: hash[:l1_lambda],
             l2_lambda: hash[:l2_lambda])
end

Instance Method Details

#backward(dout) ⇒ Object 



114
115
116
117
118
119
120
121
 
# File 'lib/dnn/core/cnn_layers.rb', line 114

def backward(dout)
  dout = dout.reshape(dout.shape[0..2].reduce(:*), dout.shape[3])
  @weight.grad = @col.transpose.dot(dout)
  @bias.grad = dout.sum(0)
  dcol = dout.dot(@weight.data.transpose)
  dx = col2im(dcol, @x_shape, *@out_size, *@filter_size, @strides)
  @padding ? back_padding(dx, @pad) : dx
end

#build(model) ⇒ Object 



95
96
97
98
99
100
101
102
103
104
 
# File 'lib/dnn/core/cnn_layers.rb', line 95

def build(model)
  super
  prev_h, prev_w = prev_layer.shape[0..1]
  @out_size = out_size(prev_h, prev_w, *@filter_size, @strides)
  out_w, out_h = @out_size
  if @padding
    @pad = [prev_h - out_h, prev_w - out_w]
    @out_size = [prev_h, prev_w]
  end
end

#forward(x) ⇒ Object 



106
107
108
109
110
111
112
 
# File 'lib/dnn/core/cnn_layers.rb', line 106

def forward(x)
  x = padding(x, @pad) if @padding
  @x_shape = x.shape
  @col = im2col(x, *@out_size, *@filter_size, @strides)
  out = @col.dot(@weight.data) + @bias.data
  out.reshape(x.shape[0], *@out_size, out.shape[3])
end

#shape ⇒ Object 



123
124
125
 
# File 'lib/dnn/core/cnn_layers.rb', line 123

def shape
  [*@out_size, @num_filters]
end

#to_hash ⇒ Object 



127
128
129
130
131
132
 
# File 'lib/dnn/core/cnn_layers.rb', line 127

def to_hash
  super({num_filters: @num_filters,
         filter_size: @filter_size,
         strides: @strides,
         padding: @padding})
end