Merge pull request #92 from milancurcic/rename-output-to-predict

Rename output -> predict for consistency with Keras

Author: milancurcic

example/cnn.f90

@@ -27,6 +27,6 @@ program cnn
   allocate(x(3,32,32))
   call random_number(x)
 
-  print *, 'Output:', net % output(x)
+  print *, 'Output:', net % predict(x)
 
 end program cnn

example/cnn_from_keras.f90

@@ -48,7 +48,7 @@ real function accuracy(net, x, y)
     integer :: i, good
     good = 0
     do i = 1, size(x, dim=4)
-      if (all(maxloc(net % output(x(:,:,:,i))) == maxloc(y(:,i)))) then
+      if (all(maxloc(net % predict(x(:,:,:,i))) == maxloc(y(:,i)))) then
         good = good + 1
       end if
     end do

example/dense_from_keras.f90

@@ -42,7 +42,7 @@ real function accuracy(net, x, y)
     integer :: i, good
     good = 0
     do i = 1, size(x, dim=2)
-      if (all(maxloc(net % output(x(:,i))) == maxloc(y(:,i)))) then
+      if (all(maxloc(net % predict(x(:,i))) == maxloc(y(:,i)))) then
         good = good + 1
       end if
     end do

example/mnist.f90

@@ -52,7 +52,7 @@ real function accuracy(net, x, y)
     integer :: i, good
     good = 0
     do i = 1, size(x, dim=2)
-      if (all(maxloc(net % output(x(:,i))) == maxloc(y(:,i)))) then
+      if (all(maxloc(net % predict(x(:,i))) == maxloc(y(:,i)))) then
         good = good + 1
       end if
     end do

example/simple.f90

@@ -27,7 +27,7 @@ program simple
     call net % update(1.)
 
     if (mod(n, 50) == 0) &
-      print '(i4,2(3x,f8.6))', n, net % output(x)
+      print '(i4,2(3x,f8.6))', n, net % predict(x)
 
   end do
 

example/sine.f90

@@ -34,7 +34,7 @@ program sine
     call net % update(1.)
 
     if (mod(n, 10000) == 0) then
-      ypred = [(net % output([xtest(i)]), i = 1, test_size)]
+      ypred = [(net % predict([xtest(i)]), i = 1, test_size)]
       print '(i0,1x,f9.6)', n, sum((ypred - ytest)**2) / size(ypred)
     end if
 

src/nf/nf_network.f90

@@ -23,13 +23,13 @@ module nf_network
 
     procedure, private :: forward_1d
     procedure, private :: forward_3d
-    procedure, private :: output_1d
-    procedure, private :: output_3d
-    procedure, private :: output_batch_1d
-    procedure, private :: output_batch_3d
+    procedure, private :: predict_1d
+    procedure, private :: predict_3d
+    procedure, private :: predict_batch_1d
+    procedure, private :: predict_batch_3d
 
     generic :: forward => forward_1d, forward_3d
-    generic :: output => output_1d, output_3d, output_batch_1d, output_batch_3d
+    generic :: predict => predict_1d, predict_3d, predict_batch_1d, predict_batch_3d
 
   end type network
 
@@ -89,45 +89,45 @@ end subroutine forward_3d
 
   interface output
 
-    module function output_1d(self, input) result(res)
+    module function predict_1d(self, input) result(res)
       !! Return the output of the network given the input 1-d array.
       class(network), intent(in out) :: self
         !! Network instance
       real, intent(in) :: input(:)
         !! Input data
       real, allocatable :: res(:)
         !! Output of the network
-    end function output_1d
+    end function predict_1d
 
-    module function output_3d(self, input) result(res)
+    module function predict_3d(self, input) result(res)
       !! Return the output of the network given the input 3-d array.
       class(network), intent(in out) :: self
         !! Network instance
       real, intent(in) :: input(:,:,:)
         !! Input data
       real, allocatable :: res(:)
         !! Output of the network
-    end function output_3d
+    end function predict_3d
 
-    module function output_batch_1d(self, input) result(res)
+    module function predict_batch_1d(self, input) result(res)
       !! Return the output of the network given an input batch of 3-d data.
       class(network), intent(in out) :: self
         !! Network instance
       real, intent(in) :: input(:,:)
         !! Input data; the last dimension is the batch
       real, allocatable :: res(:,:)
         !! Output of the network; the last dimension is the batch
-    end function output_batch_1d
+    end function predict_batch_1d
 
-    module function output_batch_3d(self, input) result(res)
+    module function predict_batch_3d(self, input) result(res)
       !! Return the output of the network given an input batch of 3-d data.
       class(network), intent(in out) :: self
         !! Network instance
       real, intent(in) :: input(:,:,:,:)
         !! Input data; the last dimension is the batch
       real, allocatable :: res(:,:)
         !! Output of the network; the last dimension is the batch
-    end function output_batch_3d
+    end function predict_batch_3d
 
   end interface output
 

src/nf/nf_network_submodule.f90

@@ -244,7 +244,7 @@ pure module subroutine forward_3d(self, input)
   end subroutine forward_3d
 
 
-  module function output_1d(self, input) result(res)
+  module function predict_1d(self, input) result(res)
     class(network), intent(in out) :: self
     real, intent(in) :: input(:)
     real, allocatable :: res(:)
@@ -263,10 +263,10 @@ module function output_1d(self, input) result(res)
         error stop 'network % output not implemented for this output layer'
     end select
 
-  end function output_1d
+  end function predict_1d
 
 
-  module function output_3d(self, input) result(res)
+  module function predict_3d(self, input) result(res)
     class(network), intent(in out) :: self
     real, intent(in) :: input(:,:,:)
     real, allocatable :: res(:)
@@ -288,10 +288,10 @@ module function output_3d(self, input) result(res)
         error stop 'network % output not implemented for this output layer'
     end select
 
-  end function output_3d
+  end function predict_3d
 
 
-  module function output_batch_1d(self, input) result(res)
+  module function predict_batch_1d(self, input) result(res)
     class(network), intent(in out) :: self
     real, intent(in) :: input(:,:)
     real, allocatable :: res(:,:)
@@ -318,10 +318,10 @@ module function output_batch_1d(self, input) result(res)
 
     end do batch
 
-  end function output_batch_1d
+  end function predict_batch_1d
 
 
-  module function output_batch_3d(self, input) result(res)
+  module function predict_batch_3d(self, input) result(res)
     class(network), intent(in out) :: self
     real, intent(in) :: input(:,:,:,:)
     real, allocatable :: res(:,:)
@@ -335,23 +335,23 @@ module function output_batch_3d(self, input) result(res)
 
     batch: do concurrent(i = 1:batch_size)
 
-    call self % forward(input(:,:,:,i))
+      call self % forward(input(:,:,:,i))
 
-    select type(output_layer => self % layers(num_layers) % p)
-      type is(conv2d_layer)
-        !FIXME flatten the result for now; find a better solution
-        res(:,i) = pack(output_layer % output, .true.)
-      type is(dense_layer)
-        res(:,i) = output_layer % output
-      type is(flatten_layer)
-        res(:,i) = output_layer % output
-      class default
-        error stop 'network % output not implemented for this output layer'
-    end select
+      select type(output_layer => self % layers(num_layers) % p)
+        type is(conv2d_layer)
+          !FIXME flatten the result for now; find a better solution
+          res(:,i) = pack(output_layer % output, .true.)
+        type is(dense_layer)
+          res(:,i) = output_layer % output
+        type is(flatten_layer)
+          res(:,i) = output_layer % output
+        class default
+          error stop 'network % output not implemented for this output layer'
+      end select
 
     end do batch
 
-  end function output_batch_3d
+  end function predict_batch_3d
 
 
   module subroutine print_info(self)

test/test_cnn_from_keras.f90

@@ -59,7 +59,7 @@ real function accuracy(net, x, y)
     integer :: i, good
     good = 0
     do i = 1, size(x, dim=4)
-      if (all(maxloc(net % output(x(:,:,:,i))) == maxloc(y(:,i)))) then
+      if (all(maxloc(net % predict(x(:,:,:,i))) == maxloc(y(:,i)))) then
         good = good + 1
       end if
     end do

test/test_dense_network.f90

@@ -16,7 +16,7 @@ program test_dense_network
     ok = .false.
   end if
 
-  if (.not. all(net % output([0.]) == 0.5)) then
+  if (.not. all(net % predict([0.]) == 0.5)) then
     write(stderr, '(a)') &
       'dense network should output exactly 0.5 for input 0.. failed'
     ok = .false.
@@ -35,7 +35,7 @@ program test_dense_network
       call net % forward(x)
       call net % backward(y)
       call net % update(1.)
-      if (all(abs(net % output(x) - y) < tolerance)) exit
+      if (all(abs(net % predict(x) - y) < tolerance)) exit
     end do
 
     if (.not. n <= num_iterations) then