Update. The input/output can now be nested tables.

[dagnn.git] / dagnn.lua

diff --git a/dagnn.lua b/dagnn.lua
index 1ec9b4e..65a30e2 100755 (executable)
--- a/dagnn.lua
+++ b/dagnn.lua
@@ -1,21 +1,17 @@
-#!/usr/bin/env luajit
 
 require 'torch'
 require 'nn'
-require 'image'
-require 'optim'
 
-----------------------------------------------------------------------
+local DAG, parent = torch.class('nn.DAG', 'nn.Container')
 
-local Graph, parent = torch.class('nn.Graph', 'nn.Container')
-
-function Graph:__init()
+function DAG:__init()
    parent.__init(self)
    self.pred = {}
    self.succ = {}
 end
 
-function Graph:addEdge(a, b)
+function DAG:addEdge(a, b)
+   self.sorted = nil
    local pred, succ = self.pred, self.succ
    if not pred[a] and not succ[a] then
       self:add(a)
@@ -29,39 +25,37 @@ function Graph:addEdge(a, b)
    succ[a][#succ[a] + 1] = b
 end
 
-function Graph:setInput(i)
-   if torch.type(i) == 'table' then
-      self.inputModules = i
-      for _, m in ipairs(i) do
-         if not self.pred[m] and not self.succ[m] then
-            self:add(m)
-         end
+function DAG:applyOnModules(f, t1, t2)
+   if torch.type(t1) == 'table' then
+      local result = {}
+      for k, s in pairs(t1) do
+         result[k] = self:applyOnModules(f, s, t2 and t2[k])
       end
+      return result
    else
-      self:setInput({ i })
+      return f(t1, t2)
    end
 end
 
-function Graph:setOutput(o)
-   if torch.type(o) == 'table' then
-      self.outputModules = o
-      for _, m in ipairs(o) do
-         if not self.pred[m] and not self.succ[m] then
-            self:add(m)
-         end
-      end
-   else
-      self:setOutput({ o })
-   end
+function DAG:setInput(i)
+   self.sorted = nil
+   self.inputModules = i
 end
 
-function Graph:order()
-   local distance = {}
+function DAG:setOutput(o)
+   self.sorted = nil
+   self.outputModules = o
+end
 
-   for _, a in pairs(self.inputModules) do
-      distance[a] = 1
+function DAG:sort()
+   if self.sorted then
+      return
    end
 
+   local distance = {}
+
+   self:applyOnModules(function(m) distance[m] = 1 end, self.inputModules)
+
    local nc
 
    repeat
@@ -85,20 +79,18 @@ function Graph:order()
    for i, a in ipairs(self.sorted) do self.sorted[i] = a[2] end
 end
 
-function Graph:print()
+function DAG:print()
+   self:sort()
+
    for i, d in ipairs(self.sorted) do
       print('#' .. i .. ' -> ' .. torch.type(d))
    end
 end
 
-function Graph:updateOutput(input)
-   if #self.inputModules == 1 then
-      self.inputModules[1]:updateOutput(input)
-   else
-      for i, d in ipairs(self.inputModules) do
-         d:updateOutput(input[i])
-      end
-   end
+function DAG:updateOutput(input)
+   self:sort()
+
+   self:applyOnModules(function(m, i) m:updateOutput(i) end, self.inputModules, input)
 
    for _, d in ipairs(self.sorted) do
       if self.pred[d] then
@@ -114,54 +106,13 @@ function Graph:updateOutput(input)
       end
    end
 
-   if #self.outputModules == 1 then
-      self.output = self.outputModules[1].output
-   else
-      self.output = { }
-      for i, d in ipairs(self.outputModules) do
-         self.output[i] = d.output
-      end
-   end
+   self.output = self:applyOnModules(function(m) return m.output end, self.outputModules)
 
    return self.output
 end
 
-----------------------------------------------------------------------
-
-a = nn.Linear(10, 10)
-b = nn.ReLU()
-c = nn.Linear(10, 3)
-d = nn.Linear(10, 3)
-e = nn.CMulTable()
-f = nn.Linear(3, 2)
-
---[[
-
-   a -----> b ---> c ----> e ---
-             \           /
-              \--> d ---/
-                    \
-                     \---> f ---
-]]--
-
-g = Graph:new()
-
-g:setInput(a)
-g:setOutput({ e, f })
-g:addEdge(c, e)
-g:addEdge(a, b)
-g:addEdge(d, e)
-g:addEdge(b, c)
-g:addEdge(b, d)
-g:addEdge(d, f)
-
-g:order()
-
-g:print(graph)
-
-input = torch.Tensor(3, 10):uniform()
-
-output = g:updateOutput(input)
+function DAG:updateGradInput(input, gradOutput)
+   self:sort()
+end
 
-print(output[1])
-print(output[2])
+return DAG