Updated the comments

Author: Riccardo231

src/nf/nf_conv1d_layer.f90

@@ -17,8 +17,8 @@ module nf_conv1d_layer
       integer :: filters
 
       real, allocatable :: biases(:) ! size(filters)
-      real, allocatable :: kernel(:,:,:) ! filters x channels x window x window
-      real, allocatable :: output(:,:) ! filters x output_width * output_height
+      real, allocatable :: kernel(:,:,:) ! filters x channels x window 
+      real, allocatable :: output(:,:) ! filters x output_width 
       real, allocatable :: z(:,:) ! kernel .dot. input + bias
 
       real, allocatable :: dw(:,:,:) ! weight (kernel) gradients

src/nf/nf_layer_constructors.f90

@@ -154,22 +154,21 @@ module function flatten() result(res)
     end function flatten
 
     module function conv1d(filters, kernel_size, activation) result(res)
-      !! CHANGE THE COMMENTS
-      !! 2-d convolutional layer constructor.
+      !! 1-d convolutional layer constructor.
       !!
-      !! This layer is for building 2-d convolutional network.
-      !! Although the established convention is to call these layers 2-d,
-      !! the shape of the data is actuall 3-d: image width, image height,
+      !! This layer is for building 1-d convolutional network.
+      !! Although the established convention is to call these layers 1-d,
+      !! the shape of the data is actually 2-d: image width
       !! and the number of channels.
-      !! A conv2d layer must not be the first layer in the network.
+      !! A conv1d layer must not be the first layer in the network.
       !!
       !! Example:
       !!
       !! ```
-      !! use nf, only :: conv2d, layer
-      !! type(layer) :: conv2d_layer
-      !! conv2d_layer = dense(filters=32, kernel_size=3)
-      !! conv2d_layer = dense(filters=32, kernel_size=3, activation='relu')
+      !! use nf, only :: conv1d, layer
+      !! type(layer) :: conv1d_layer
+      !! conv1d_layer = dense(filters=32, kernel_size=3)
+      !! conv1d_layer = dense(filters=32, kernel_size=3, activation='relu')
       !! ```
       integer, intent(in) :: filters
         !! Number of filters in the output of the layer

src/nf/nf_layer_submodule.f90

@@ -27,7 +27,7 @@ pure module subroutine backward_1d(self, previous, gradient)
     real, intent(in) :: gradient(:)
 
     ! Backward pass from a 1-d layer downstream currently implemented
-    ! only for dense and flatten layers
+    ! only for dense, dropout and flatten layers
     select type(this_layer => self % p)
 
       type is(dense_layer)
@@ -50,7 +50,7 @@ pure module subroutine backward_1d(self, previous, gradient)
 
       type is(flatten_layer)
 
-        ! Upstream layers permitted: input2d, input3d, conv2d, locally_connected_1d, maxpool1d, maxpool2d
+        ! Upstream layers permitted: input2d, input3d, conv1d, conv2d, locally_connected_1d, maxpool1d, maxpool2d
         select type(prev_layer => previous % p)
           type is(input2d_layer)
             call this_layer % backward(prev_layer % output, gradient)

test/test_conv1d_network.f90

@@ -41,7 +41,7 @@ program test_conv1d_network
       real :: y(1)
       real :: tolerance = 1e-4
       integer :: n
-      integer, parameter :: num_iterations = 1000
+      integer, parameter :: num_iterations = 1500
 
       ! Test training of a minimal constant mapping
       allocate(sample_input(1, 5))
@@ -79,7 +79,7 @@ program test_conv1d_network
       real :: y(1)
       real :: tolerance = 1e-4
       integer :: n
-      integer, parameter :: num_iterations = 1000
+      integer, parameter :: num_iterations = 1500
 
       call random_number(x)
       y = [0.1234567]