Fixed cuDNN Unit test project compile error

Author: Sergio0694

NeuralNetwork.NET/SupervisedLearning/Optimization/NetworkTrainer.cs

@@ -85,17 +85,7 @@ public static TrainingSessionResult TrainNetwork(
         /// <summary>
         /// Gets whether or not a neural network is currently processing the training samples through backpropagation (as opposed to evaluating them)
         /// </summary>
-        public static bool BackpropagationInProgress
-        {
-            get;
-
-            // Switch from private to internal in DEBUG mode to allow for external handling in the Unit tests
-            #if DEBUG
-            set;
-            #else
-            private set;
-            #endif
-        }
+        public static bool BackpropagationInProgress { get; private set; }
 
         /// <summary>
         /// Trains the target <see cref="SequentialNetwork"/> using the input algorithm

Unit/NeuralNetwork.NET.Cuda.Unit/CuDnnLayersTest.cs

@@ -1,4 +1,5 @@
 using System;
+using System.Reflection;
 using JetBrains.Annotations;
 using Microsoft.VisualStudio.TestTools.UnitTesting;
 using NeuralNetworkNET.APIs.Enums;
@@ -46,9 +47,17 @@ private static void TestForward(NetworkLayerBase cpu, NetworkLayerBase gpu, int
             Tensor.Free(x, z_cpu, a_cpu, z_gpu, a_gpu);
         }
 
+        // Sets the static property that signals whenever the backpropagation pass is being executed (needed for some layer types)
+        private static void SetBackpropagationProperty(bool value)
+        {
+            PropertyInfo property = typeof(NetworkTrainer).GetProperty(nameof(NetworkTrainer.BackpropagationInProgress), BindingFlags.Static | BindingFlags.Public);
+            if (property == null) throw new InvalidOperationException("Couldn't find the target property");
+            property.SetValue(null, value);
+        }
+
         private static void TestBackward(WeightedLayerBase cpu, WeightedLayerBase gpu, int samples)
         {
-            NetworkTrainer.BackpropagationInProgress = true;
+            SetBackpropagationProperty(true);
             Tensor
                 x = CreateRandomTensor(samples, cpu.InputInfo.Size),
                 dy = CreateRandomTensor(samples, cpu.OutputInfo.Size);
@@ -62,12 +71,12 @@ private static void TestBackward(WeightedLayerBase cpu, WeightedLayerBase gpu, i
             Assert.IsTrue(dJdw_cpu.ContentEquals(dJdw_gpu, 1e-4f, 1e-5f));
             Assert.IsTrue(dJdb_cpu.ContentEquals(dJdb_gpu, 1e-4f, 1e-5f)); // The cuDNN ConvolutionBackwardBias is not always as precise as the CPU version
             Tensor.Free(x, dy, dx1, dx2, z_cpu, a_cpu, z_gpu, a_gpu, dJdw_cpu, dJdb_cpu, dJdw_gpu, dJdb_gpu);
-            NetworkTrainer.BackpropagationInProgress = false;
+            SetBackpropagationProperty(false);
         }
 
         private static unsafe void TestBackward(OutputLayerBase cpu, OutputLayerBase gpu, float[,] y)
         {
-            NetworkTrainer.BackpropagationInProgress = true;
+            SetBackpropagationProperty(true);
             int n = y.GetLength(0);
             fixed (float* p = y)
             {
@@ -86,7 +95,7 @@ private static unsafe void TestBackward(OutputLayerBase cpu, OutputLayerBase gpu
                 Assert.IsTrue(dJdb_cpu.ContentEquals(dJdb_gpu, 1e-4f, 1e-5f));
                 Tensor.Free(x, dy, dx1, dx2, z_cpu, a_cpu, z_gpu, a_gpu, dJdw_cpu, dJdw_gpu, dJdb_cpu, dJdb_gpu);
             }
-            NetworkTrainer.BackpropagationInProgress = false;
+            SetBackpropagationProperty(false);
         }
 
         #endregion