SparseNN/SparseBlockTemporalMaxPooling.lua at master · IBPA/SparseNN · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
local SparseBlockTemporalMaxPooling, parent = torch.class('nn.SparseBlockTemporalMaxPooling', 'nn.Module')

function SparseBlockTemporalMaxPooling:__init(kW, dW, isRelax)
   dW = dW or kW
   self.kW = kW
   self.dW = dW
   self.isRelax = isRelax or false
end

function SparseBlockTemporalMaxPooling:pri_ensureOutput(input)
   if self.output ~= nil then
      return
   end

   self.output = { nBatchSize = input.nBatchSize, taData = {} }

   local nColumns = table.getn(input.taData)
   for i=1, nColumns do
      local taInputCurr = input.taData[i]

      taOutputCurr = { teValue = torch.Tensor(),
                       teRowIdx = taInputCurr.teRowIdx }

      table.insert(self.output.taData, taOutputCurr)
   end
end

function SparseBlockTemporalMaxPooling:pri_ensureTaIndices(input)
   if self.taIndices ~= nil then
      return
   end

   self.taIndices = {}

   local nColumns = table.getn(input.taData)
   for i=1, nColumns do
      local taInputCurr = input.taData[i]
      table.insert(self.taIndices, torch.Tensor()) -- Important Note: due to a bug in torch+nn, this is reverted back (instead of torch.LongTensor()). For this to work, need to be using torch @ commit: 1e5a315d03c91286d859512574d3b0b25e12d512, and nn @ commit: 1443cd7c2becd793b3d954144dcf4a1bf9947771
   end
end

function SparseBlockTemporalMaxPooling:pri_updateOutput_column(taInput, taOutput, teIndices)
   local input = taInput.teValue
   local output = taOutput.teValue

   local kW = self.kW
   if self.isRelax  then
      kW = math.min(input:size(2), kW)
   end

   input.THNN.TemporalMaxPooling_updateOutput(
       input:cdata(), output:cdata(),
       teIndices:cdata(), kW, self.dW
   )

end

function SparseBlockTemporalMaxPooling:updateOutput(input)
   self:pri_ensureTaIndices(input)
   self:pri_ensureOutput(input)

   local nColumns = table.getn(self.output.taData)
   for i=1, nColumns do
      self:pri_updateOutput_column(input.taData[i],
                                   self.output.taData[i],
                                   self.taIndices[i])
   end

   return self.output
end

function SparseBlockTemporalMaxPooling:pri_ensureGradInput(input)
   if self.gradInput ~= nil then
      return
   end

   self.gradInput = { nBatchSize = input.nBatchSize, taData = {} }

   local nColumns = table.getn(input.taData)
   for i=1, nColumns do
      local taInputCurr = input.taData[i]

      taGradInputCurr = { teValue = torch.Tensor(),
                          teRowIdx = taInputCurr.teRowIdx }
      table.insert(self.gradInput.taData, taGradInputCurr)
   end
end

function SparseBlockTemporalMaxPooling:pri_updateGradInput_column(taInput, taGradOutput, taGradInput, teIndices)
   local input = taInput.teValue
   local gradOutput = taGradOutput.teValue
   local gradInput = taGradInput.teValue

   local kW = self.kW
   if self.isRelax  then
      kW = math.min(input:size(2), kW)
   end


   input.THNN.TemporalMaxPooling_updateGradInput(
       input:cdata(), gradOutput:cdata(),
       gradInput:cdata(), teIndices:cdata(),
       kW, self.dW
   )

end

function SparseBlockTemporalMaxPooling:updateGradInput(input, gradOutput)
   self:pri_ensureGradInput(input)

   local nColumns = table.getn(self.gradInput.taData)
   for i=1, nColumns do
      self:pri_updateGradInput_column(input.taData[i],
                                      gradOutput.taData[i],
                                      self.gradInput.taData[i],
                                      self.taIndices[i])
   end

   return self.gradInput
end