Class-based activation functions (#126)

* Implementation of activation_function class for 1d activations

* 3d activations implemented using activation_function type

* get_name function added to the activation_function type

* Activation_function instances are now passed to contructors

* removal of redundant use statements

* Small fix to make the test build

* Tidy up and formatting

* Formatting

* Set alpha defaults from Keras

* Enable leaky ReLU

* Add tests for setting alpha values to parametric activations (ELU and leaky ReLU)

* Bump version

---------

Co-authored-by: milancurcic <caomaco@gmail.com>

Author: ggoyman

CMakeLists.txt

@@ -24,8 +24,7 @@ include(cmake/json.cmake)
 # library to archive (libneural.a)
 add_library(neural
   src/nf.f90
-  src/nf/nf_activation_1d.f90
-  src/nf/nf_activation_3d.f90
+  src/nf/nf_activation.f90
   src/nf/nf_base_layer.f90
   src/nf/nf_conv2d_layer.f90
   src/nf/nf_conv2d_layer_submodule.f90

example/cnn_mnist.f90

@@ -2,7 +2,7 @@ program cnn_mnist
 
   use nf, only: network, sgd, &
     input, conv2d, maxpool2d, flatten, dense, reshape, &
-    load_mnist, label_digits
+    load_mnist, label_digits, softmax, relu
 
   implicit none
 
@@ -24,11 +24,11 @@ program cnn_mnist
   net = network([ &
     input(784), &
     reshape([1,28,28]), &
-    conv2d(filters=8, kernel_size=3, activation='relu'), &
+    conv2d(filters=8, kernel_size=3, activation=relu()), &
     maxpool2d(pool_size=2), &
-    conv2d(filters=16, kernel_size=3, activation='relu'), &
+    conv2d(filters=16, kernel_size=3, activation=relu()), &
     maxpool2d(pool_size=2), &
-    dense(10, activation='softmax') &
+    dense(10, activation=softmax()) &
   ])
 
   call net % print_info()
@@ -67,4 +67,4 @@ real function accuracy(net, x, y)
     accuracy = real(good) / size(x, dim=2)
   end function accuracy
 
-end program cnn_mnist
+end program cnn_mnist

fpm.toml

@@ -1,5 +1,5 @@
 name = "neural-fortran"
-version = "0.11.0"
+version = "0.12.0"
 license = "MIT"
 author = "Milan Curcic"
 maintainer = "milancurcic@hey.com"

src/nf.f90

@@ -6,4 +6,7 @@ module nf
     conv2d, dense, flatten, input, maxpool2d, reshape
   use nf_network, only: network
   use nf_optimizers, only: sgd
+  use nf_activation, only: activation_function, elu, exponential,  &
+                           gaussian, linear, relu, leaky_relu,     &
+                           sigmoid, softmax, softplus, step, tanhf
 end module nf