Fix

src/SharpLearning.DecisionTrees/ImpurityCalculators/ClassificationImpurityCalculator.cs

@@ -33,9 +33,9 @@ public abstract class ClassificationImpurityCalculator
     /// </summary>
     protected double m_weightedRight = 0.0;
 
-    internal TargetCounts m_weightedTargetCount = new();
-    internal TargetCounts m_weightedTargetCountLeft = new();
-    internal TargetCounts m_weightedTargetCountRight = new();
+    internal TargetCounts WeightedTargetCount = new();
+    internal TargetCounts WeightedTargetCountLeft = new();
+    internal TargetCounts WeightedTargetCountRight = new();
 
     /// <summary>
     /// 
@@ -96,9 +96,9 @@ public void Init(double[] targetNames, double[] targets, double[] weights, Inter
             m_targetIndexOffSet *= -1;
         }
 
-        m_weightedTargetCount.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
-        m_weightedTargetCountLeft.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
-        m_weightedTargetCountRight.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
+        WeightedTargetCount.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
+        WeightedTargetCountLeft.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
+        WeightedTargetCountRight.Reset(m_maxTargetNameIndex, m_targetIndexOffSet);
 
         var w = 1.0;
         var weightsPresent = m_weights.Length != 0;
@@ -113,7 +113,7 @@ public void Init(double[] targetNames, double[] targets, double[] weights, Inter
                 w = weights[i];
 
             var targetIndex = (int)targets[i];
-            m_weightedTargetCount[targetIndex] += w;
+            WeightedTargetCount[targetIndex] += w;
 
             m_weightedTotal += w;
         }
@@ -132,8 +132,8 @@ public void Reset()
         m_weightedLeft = 0.0;
         m_weightedRight = m_weightedTotal;
 
-        m_weightedTargetCountLeft.Clear();
-        m_weightedTargetCountRight.SetCounts(m_weightedTargetCount);
+        WeightedTargetCountLeft.Clear();
+        WeightedTargetCountRight.SetCounts(WeightedTargetCount);
     }
 
     /// <summary>
@@ -168,8 +168,8 @@ public void UpdateIndex(int newPosition)
                 w = m_weights[i];
 
             var targetIndex = (int)m_targets[i];
-            m_weightedTargetCountLeft[targetIndex] += w;
-            m_weightedTargetCountRight[targetIndex] -= w;
+            WeightedTargetCountLeft[targetIndex] += w;
+            WeightedTargetCountRight[targetIndex] -= w;
 
             w_diff += w;
         }

src/SharpLearning.DecisionTrees/ImpurityCalculators/GiniClassificationImpurityCalculator.cs

@@ -29,8 +29,8 @@ public override ChildImpurities ChildImpurities()
         foreach (var targetValue in m_targetNames)
         {
             var targetIndex = (int)targetValue;
-            var leftCount = m_weightedTargetCountLeft[targetIndex];
-            var rightCount = m_weightedTargetCountRight[targetIndex];
+            var leftCount = WeightedTargetCountLeft[targetIndex];
+            var rightCount = WeightedTargetCountRight[targetIndex];
 
             giniLeft += leftCount * leftCount;
             giniRight += rightCount * rightCount;
@@ -52,7 +52,7 @@ public override double NodeImpurity()
 
         foreach (var targetValue in m_targetNames)
         {
-            var value = m_weightedTargetCount[(int)targetValue];
+            var value = WeightedTargetCount[(int)targetValue];
             gini += value * value;
         }
 
@@ -86,7 +86,7 @@ public override double LeafValue()
 
         foreach (var targetValue in m_targetNames)
         {
-            var value = m_weightedTargetCount[(int)targetValue];
+            var value = WeightedTargetCount[(int)targetValue];
             if (value > maxWeight)
             {
                 maxWeight = value;
@@ -110,7 +110,7 @@ public override double[] LeafProbabilities()
         for (var i = 0; i < m_targetNames.Length; i++)
         {
             var targetValue = (int)m_targetNames[i];
-            var targetProbability = (m_weightedTargetCount[targetValue] + 1) * probabilityFactor;
+            var targetProbability = (WeightedTargetCount[targetValue] + 1) * probabilityFactor;
             probabilities[i] = targetProbability;
         }
 

src/SharpLearning.GradientBoost/Models/RegressionGradientBoostModel.cs

@@ -91,10 +91,10 @@ public double[] Predict(F64Matrix observations)
     /// <returns></returns>
     public Dictionary<string, double> GetVariableImportance(Dictionary<string, int> featureNameToIndex)
     {
-        var m_rawVariableImportance = GetRawVariableImportance();
-        var max = m_rawVariableImportance.Max();
+        var rawVariableImportance = GetRawVariableImportance();
+        var max = rawVariableImportance.Max();
 
-        var scaledVariableImportance = m_rawVariableImportance
+        var scaledVariableImportance = rawVariableImportance
             .Select(v => (v / max) * 100.0)
             .ToArray();
 

src/SharpLearning.Neural/Layers/BatchNormalizationLayer.cs

@@ -98,6 +98,7 @@ public Matrix<float> Backward(Matrix<float> delta)
         var diff_src = m_delta;
         var diff_scaleshift = m_scaleGradients;
 
+#pragma warning disable IDE1006 // Naming Styles
         var N = diff_src.RowCount;
         var C = Depth;
         var H = Height;
@@ -136,7 +137,7 @@ public Matrix<float> Backward(Matrix<float> delta)
                         diff_src.Data()[diffSrcIndex] *= gamma * variance;
                     }
         });
-
+#pragma warning restore IDE1006 // Naming Styles
         return m_delta;
     }
 
@@ -151,7 +152,7 @@ public Matrix<float> Forward(Matrix<float> input)
 
         var src = input;
         var dst = OutputActivations.Data();
-
+#pragma warning disable IDE1006 // Naming Styles
         var N = input.RowCount; // number of items in mini batch
         var C = Depth;
         var H = Height;
@@ -197,7 +198,7 @@ public Matrix<float> Forward(Matrix<float> input)
                         var bias = Bias[c];
                         dst[d_off] = scale * (src.Data()[d_off] - mean) * variance + bias;
                     }
-
+#pragma warning restore IDE1006 // Naming Styles
             if (is_training)
             {
                 MovingAverageMeans[c] = MovingAverage(MovingAverageMeans[c], mean);