Class: MachineLearningWorkbench::NeuralNetwork::Base

Inherits:
Object
  • Object
show all
Defined in:
lib/machine_learning_workbench/neural_network/base.rb

Overview

Neural Network base class

Direct Known Subclasses

FeedForward, Recurrent

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(struct, act_fn: nil, dtype: :float32) ⇒ Base

Returns a new instance of Base.

Parameters:

  • struct (Array<Integer>) —

    list of layer sizes

  • act_fn (Symbol) (defaults to: nil) —

    choice of activation function for the neurons

  • dtype (NMatrix dtype) (defaults to: :float32) —

    NMatrix dtype for weights and states



28
29
30
31
32
33
34
35
36
37
38
39
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 28

def initialize struct, act_fn: nil, dtype: :float32
  @struct = struct
  @act_fn = self.get_act_fn(act_fn || :sigmoid)
  # @state holds both inputs, possibly recurrency, and bias
  # it is a complete input for the next layer, hence size from layer sizes
  @state = layer_row_sizes.collect do |size|
    NMatrix.zeros([1, size], dtype: dtype)
  end
  # to this, append a matrix to hold the final network output
  @state.push NMatrix.zeros([1, nneurs(-1)], dtype: dtype)
  reset_state
end

Instance Attribute Details

#act_fn ⇒ #call (readonly)

activation function, common to all neurons (for now)

Returns:

  • (#call) —

    activation function



20
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 20

attr_reader :layers, :state, :act_fn, :struct, :dtype

#dtype ⇒ Object (readonly)

Returns the value of attribute dtype.



20
21
22
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 20

def dtype
  @dtype
end

#layers ⇒ Array<NMatrix> (readonly)

List of matrices, each being the weights connecting a layer’s inputs (rows) to a layer’s neurons (columns), hence its shape is ‘[ninputs, nneurs]`

Returns:

  • (Array<NMatrix>) —

    list of weight matrices, each uniquely describing a layer



20
21
22
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 20

def layers
  @layers
end

#state ⇒ Array<NMatrix> (readonly)

It’s a list of one-dimensional matrices, each an input to a layer, plus the output layer’s output. The first element is the input to the first layer of the network, which is composed of the network’s input, possibly the first layer’s activation on the last input (recursion), and a bias (fixed ‘1`). The second to but-last entries follow the same structure, but with the previous layer’s output in place of the network’s input. The last entry is the activation of the output layer, without additions since it’s not used as an input by anyone.

Returns:

  • (Array<NMatrix>) —

    current state of the network.



20
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 20

attr_reader :layers, :state, :act_fn, :struct, :dtype

#struct ⇒ Array<Integer> (readonly)

list of number of (inputs or) neurons in each layer

Returns:

  • (Array<Integer>) —

    structure of the network



20
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 20

attr_reader :layers, :state, :act_fn, :struct, :dtype

Instance Method Details

#activate(input) ⇒ Array

Activate the network on a given input

Parameters:

  • input (Array<Float>) —

    the given input

Returns:

  • (Array) —

    the activation of the output layer

Raises:

  • (ArgumentError) 


150
151
152
153
154
155
156
157
158
159
160
161
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 150

def activate input
  raise ArgumentError unless input.size == struct.first
  raise ArgumentError unless input.is_a? Array
  # load input in first state
  @state[0][0, 0..-2] = input
  # activate layers in sequence
  (0...nlayers).each do |i|
    act = activate_layer i
    @state[i+1][0,0...act.size] = act
  end
  return out
end

#bias ⇒ Object

The “fixed ‘1`” used in the layer’s input



143
144
145
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 143

def bias
  @bias ||= NMatrix[[1], dtype: dtype]
end

#deep_reset ⇒ Object

Resets memoization: needed to play with structure modification



61
62
63
64
65
66
67
68
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 61

def deep_reset
  # reset memoization
  [:@layer_row_sizes, :@layer_col_sizes, :@nlayers, :@layer_shapes,
   :@nweights_per_layer, :@nweights].each do |sym|
     instance_variable_set sym, nil
  end
  reset_state
end

#get_act_fn(type, *args) ⇒ NMatrix

Activation function caller. Allows to cleanly define the activation function as one-dimensional, by calling it over the inputs and building a NMatrix to return.

Returns:

  • (NMatrix) —

    activations for one layer



175
176
177
178
179
180
181
182
183
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 175

def get_act_fn type, *args
  fn = send(type,*args)
  lambda do |inputs|
    NMatrix.new([1, inputs.size], dtype: dtype) do |_,i|
      # single-row matrix, indices are columns
      fn.call inputs[i]
    end
  end
end

#init_random ⇒ Object

Initialize the network with random weights



52
53
54
55
56
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 52

def init_random
  # Will only be used for testing, no sense optimizing it (NMatrix#rand)
  # Reusing #load_weights instead helps catching bugs
  load_weights nweights.times.collect { rand(-1.0..1.0) }
end

#interface_methods ⇒ Object

Declaring interface methods - implement in child class!



214
215
216
217
218
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 214

[:layer_row_sizes, :activate_layer].each do |sym|
  define_method sym do
    raise NotImplementedError, "Implement ##{sym} in child class!"
  end
end

#layer_col_sizes ⇒ Array

Number of neurons per layer. Although this implementation includes inputs in the layer counts, this methods correctly ignores the input as not having neurons.

Returns:

  • (Array) —

    list of neurons per each (proper) layer (i.e. no inputs)



100
101
102
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 100

def layer_col_sizes
  @layer_col_sizes ||= struct.drop(1)
end

#layer_shapes ⇒ Array<Array[Integer, Integer]>

Shapes for the weight matrices, each corresponding to a layer

Returns:

  • (Array<Array[Integer, Integer]>) —

    Weight matrix shapes



108
109
110
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 108

def layer_shapes
  @layer_shapes ||= layer_row_sizes.zip layer_col_sizes
end

#lecun_hyperbolic ⇒ Object

LeCun hyperbolic activation

See Also:



202
203
204
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 202

def lecun_hyperbolic
  lambda { |x| 1.7159 * Math.tanh(2.0*x/3.0) + 1e-3*x }
end

#load_weights(weights) ⇒ true

Loads a plain list of weights into the weight matrices (one per layer). Preserves order. Reuses allocated memory if available.

Returns:

  • (true) —

    always true. If something’s wrong it simply fails, and if all goes well there’s nothing to return but a confirmation to the caller.

Raises:

  • (ArgumentError) 


126
127
128
129
130
131
132
133
134
135
136
137
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 126

def load_weights weights
  raise ArgumentError unless weights.size == nweights
  weights_iter = weights.each
  @layers ||= layer_shapes.collect { |shape| NMatrix.new shape, dtype: dtype }
  layers.each do |nmat|
    nmat.each_with_indices do |_val, *idxs|
      nmat[*idxs] = weights_iter.next
    end
  end
  reset_state
  return true
end

#logistic ⇒ Object

Traditional logistic



193
194
195
196
197
198
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 193

def logistic
  lambda { |x|
    exp = Math.exp(x)
    exp.infinite? ? exp : exp / (1.0 + exp)
  }
end

#nlayers ⇒ Integer

Count the layers. This is a computation helper, and for this implementation the inputs are considered as if a layer like the others.

Returns:

  • (Integer) —

    number of layers



85
86
87
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 85

def nlayers
  @nlayers ||= layer_shapes.size
end

#nneurs(nlay = nil) ⇒ Integer

Count the neurons in a particular layer or in the whole network.

Parameters:

  • nlay (Integer, nil) (defaults to: nil) —

    the layer of interest, 1-indexed. ‘0` will return the number of inputs. `nil` will compute the total neurons in the network.

Returns:

  • (Integer) —

    the number of neurons in a given layer, or in all network, or the number of inputs



117
118
119
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 117

def nneurs nlay=nil
  nlay.nil? ? struct.reduce(:+) : struct[nlay]
end

#nweights ⇒ Integer

Total weights in the network

Returns:

  • (Integer) —

    total number of weights



72
73
74
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 72

def nweights
  @nweights ||= nweights_per_layer.reduce(:+)
end

#nweights_per_layer ⇒ Array<Integer>

List of per-layer number of weights

Returns:

  • (Array<Integer>) —

    list of weights per each layer



78
79
80
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 78

def nweights_per_layer
  @nweights_per_layer ||= layer_shapes.collect { |shape| shape.reduce(:*) }
end

#out ⇒ Array

Extract and convert the output layer’s activation

Returns:

  • (Array) —

    the activation of the output layer as 1-dim Array



165
166
167
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 165

def out
  state.last.to_flat_a
end

#relu ⇒ Object

Rectified Linear Unit (ReLU)



207
208
209
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 207

def relu
  lambda { |x| x>0 && x || 0 }
end

#reset_state ⇒ Object

Reset the network to the initial state



42
43
44
45
46
47
48
49
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 42

def reset_state
  @state.each do |m| # state has only single-row matrices
    # reset all to zero
    m[0,0..-1] = 0
    # add bias to all but output
    m[0,-1] = 1 unless m.object_id == @state.last.object_id
  end
end

#sigmoid(k = 0.5) ⇒ Object

Traditional sigmoid with variable steepness



186
187
188
189
190
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 186

def sigmoid k=0.5
  # k is steepness:  0<k<1 is flatter, 1<k is flatter
  # flatter makes activation less sensitive, better with large number of inputs
  lambda { |x| 1.0 / (Math.exp(-k * x) + 1.0) }
end

#weights ⇒ Array

Returns the weight matrix

Returns:

  • (Array) —

    three-dimensional Array of weights: a list of weight matrices, one for each layer.



92
93
94
 
# File 'lib/machine_learning_workbench/neural_network/base.rb', line 92

def weights
  layers.collect(&:to_consistent_a)
end