Merge pull request #147 from DrDub/master

Leaky ReLUs with test case
libfann · Apr 16, 2024 · 518eb6b · 518eb6b
2 parents b591110 + 23987ef
commit 518eb6b
Show file tree

Hide file tree

Showing 7 changed files with 80 additions and 8 deletions.
diff --git a/src/fann.c b/src/fann.c
@@ -683,13 +683,20 @@ FANN_EXTERNAL fann_type *FANN_API fann_run(struct fann *ann, fann_type *input) {
           neuron_it->value = neuron_sum;
           break;
         case FANN_LINEAR_PIECE:
-          neuron_it->value = (fann_type)(
-              (neuron_sum < 0) ? 0 : (neuron_sum > multiplier) ? multiplier : neuron_sum);
+          neuron_it->value = (fann_type)((neuron_sum < 0)            ? 0
+                                         : (neuron_sum > multiplier) ? multiplier
+                                                                     : neuron_sum);
           break;
         case FANN_LINEAR_PIECE_SYMMETRIC:
-          neuron_it->value = (fann_type)((neuron_sum < -multiplier)
-                                             ? -multiplier
-                                             : (neuron_sum > multiplier) ? multiplier : neuron_sum);
+          neuron_it->value = (fann_type)((neuron_sum < -multiplier)  ? -multiplier
+                                         : (neuron_sum > multiplier) ? multiplier
+                                                                     : neuron_sum);
+          break;
+        case FANN_LINEAR_PIECE_RECT:
+          neuron_it->value = (fann_type)((neuron_sum < 0) ? 0 : neuron_sum);
+          break;
+        case FANN_LINEAR_PIECE_RECT_LEAKY:
+          neuron_it->value = (fann_type)((neuron_sum < 0) ? 0.01 * neuron_sum : neuron_sum);
           break;
         case FANN_ELLIOT:
         case FANN_ELLIOT_SYMMETRIC:

diff --git a/src/fann_cascade.c b/src/fann_cascade.c
@@ -680,6 +680,8 @@ fann_type fann_train_candidates_epoch(struct fann *ann, struct fann_train_data *
         case FANN_GAUSSIAN_STEPWISE:
         case FANN_ELLIOT:
         case FANN_LINEAR_PIECE:
+        case FANN_LINEAR_PIECE_RECT:
+        case FANN_LINEAR_PIECE_RECT_LEAKY:
         case FANN_SIN:
         case FANN_COS:
           break;

diff --git a/src/fann_train.c b/src/fann_train.c
@@ -40,6 +40,10 @@ fann_type fann_activation_derived(unsigned int activation_function, fann_type st
     case FANN_LINEAR_PIECE:
     case FANN_LINEAR_PIECE_SYMMETRIC:
       return (fann_type)fann_linear_derive(steepness, value);
+    case FANN_LINEAR_PIECE_RECT:
+      return (fann_type)((value < 0) ? 0 : steepness);
+    case FANN_LINEAR_PIECE_RECT_LEAKY:
+      return (fann_type)((value < 0) ? steepness * 0.01 : steepness);
     case FANN_SIGMOID:
     case FANN_SIGMOID_STEPWISE:
       value = fann_clip(value, 0.01f, 0.99f);
@@ -125,6 +129,8 @@ fann_type fann_update_MSE(struct fann *ann, struct fann_neuron *neuron, fann_typ
     case FANN_GAUSSIAN_STEPWISE:
     case FANN_ELLIOT:
     case FANN_LINEAR_PIECE:
+    case FANN_LINEAR_PIECE_RECT:
+    case FANN_LINEAR_PIECE_RECT_LEAKY:
     case FANN_SIN:
     case FANN_COS:
       break;

diff --git a/src/include/fann_activation.h b/src/include/fann_activation.h
@@ -178,6 +178,12 @@ __doublefann_h__ is not defined
     case FANN_GAUSSIAN_STEPWISE:                                                                   \
       result = 0;                                                                                  \
       break;                                                                                       \
+    case FANN_LINEAR_PIECE_RECT:                                                                   \
+      result = (fann_type)((value < 0) ? 0 : value);                                               \
+      break;                                                                                       \
+    case FANN_LINEAR_PIECE_RECT_LEAKY:                                                             \
+      result = (fann_type)((value < 0) ? value * 0.01 : value);                                    \
+      break;                                                                                       \
   }
 
 #endif
diff --git a/src/include/fann_data.h b/src/include/fann_data.h
@@ -196,6 +196,16 @@ static char const *const FANN_TRAIN_NAMES[] = {"FANN_TRAIN_INCREMENTAL", "FANN_T
          * y = cos(x*s)/2+0.5
          * d = s*-sin(x*s)/2
 
+        FANN_LINEAR_PIECE_RECT - ReLU
+         * span: -inf < y < inf
+         * y = x<0? 0: x
+         * d = x<0? 0: 1
+
+        FANN_LINEAR_PIECE_RECT_LEAKY - leaky ReLU
+         * span: -inf < y < inf
+         * y = x<0? 0.01*x: x
+         * d = x<0? 0.01: 1
+
         See also:
         <fann_set_activation_function_layer>, <fann_set_activation_function_hidden>,
         <fann_set_activation_function_output>, <fann_set_activation_steepness>,
@@ -223,7 +233,9 @@ enum fann_activationfunc_enum {
   FANN_SIN_SYMMETRIC,
   FANN_COS_SYMMETRIC,
   FANN_SIN,
-  FANN_COS
+  FANN_COS,
+  FANN_LINEAR_PIECE_RECT,
+  FANN_LINEAR_PIECE_RECT_LEAKY
 };
 
 /* Constant: FANN_ACTIVATIONFUNC_NAMES
@@ -254,7 +266,9 @@ static char const *const FANN_ACTIVATIONFUNC_NAMES[] = {"FANN_LINEAR",
                                                         "FANN_SIN_SYMMETRIC",
                                                         "FANN_COS_SYMMETRIC",
                                                         "FANN_SIN",
-                                                        "FANN_COS"};
+                                                        "FANN_COS",
+                                                        "FANN_LINEAR_PIECE_RECT",
+                                                        "FANN_LINEAR_PIECE_RECT_LEAKY"};
 
 /* Enum: fann_errorfunc_enum
         Error function used during training.

diff --git a/src/include/fann_data_cpp.h b/src/include/fann_data_cpp.h
@@ -200,6 +200,16 @@ enum training_algorithm_enum {
      * y = cos(x*s)
      * d = s*-sin(x*s)
 
+    FANN_LINEAR_PIECE_RECT - ReLU
+     * span: -inf < y < inf
+     * y = x<0? 0: x
+     * d = x<0? 0: 1
+
+    FANN_LINEAR_PIECE_RECT_LEAKY - leaky ReLU
+     * span: -inf < y < inf
+     * y = x<0? 0.01*x: x
+     * d = x<0? 0.01: 1
+
         See also:
                 <neural_net::set_activation_function_hidden>,
                 <neural_net::set_activation_function_output>
@@ -220,7 +230,9 @@ enum activation_function_enum {
   LINEAR_PIECE,
   LINEAR_PIECE_SYMMETRIC,
   SIN_SYMMETRIC,
-  COS_SYMMETRIC
+  COS_SYMMETRIC,
+  LINEAR_PIECE_RECT,
+  LINEAR_PIECE_RECT_LEAKY
 };
 
 /* Enum: network_type_enum

diff --git a/tests/fann_test_train.cpp b/tests/fann_test_train.cpp
@@ -21,6 +21,30 @@ TEST_F(FannTestTrain, TrainOnDateSimpleXor) {
     EXPECT_LT(net.test_data(data), 0.001);
 }
 
+TEST_F(FannTestTrain, TrainOnReLUSimpleXor) {
+    neural_net net(LAYER, 3, 2, 3, 1);
+
+    data.set_train_data(4, 2, xorInput, 1, xorOutput);
+    net.set_activation_function_hidden(FANN::LINEAR_PIECE_RECT);
+    net.set_activation_steepness_hidden(1.0);
+    net.train_on_data(data, 100, 100, 0.001);
+
+    EXPECT_LT(net.get_MSE(), 0.001);
+    EXPECT_LT(net.test_data(data), 0.001);
+}
+
+TEST_F(FannTestTrain, TrainOnReLULeakySimpleXor) {
+    neural_net net(LAYER, 3, 2, 3, 1);
+
+    data.set_train_data(4, 2, xorInput, 1, xorOutput);
+    net.set_activation_function_hidden(FANN::LINEAR_PIECE_RECT_LEAKY);
+    net.set_activation_steepness_hidden(1.0);
+    net.train_on_data(data, 100, 100, 0.001);
+
+    EXPECT_LT(net.get_MSE(), 0.001);
+    EXPECT_LT(net.test_data(data), 0.001);
+}
+
 TEST_F(FannTestTrain, TrainSimpleIncrementalXor) {
     neural_net net(LAYER, 3, 2, 3, 1);
 
@@ -41,3 +65,4 @@ TEST_F(FannTestTrain, TrainSimpleIncrementalXor) {
 
     EXPECT_LT(net.get_MSE(), 0.01);
 }
+