Changeset 11229 for branches/HeuristicLab.BottomUpTreeDistance/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SimilarityCalculators/BottomUpSimilarityCalculator.cs

Timestamp:

07/29/14 16:20:08 (9 years ago)

Author:

bburlacu

Message:

#2215: Refactored and simplified DirectedGraph and related components API, simplified the BottomUpSimilarityCalculator by not using a directed graph and vertices but a simpler object so that the similarity calculator is self-contained.

File:

• branches/HeuristicLab.BottomUpTreeDistance/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SimilarityCalculators/BottomUpSimilarityCalculator.cs (modified) (8 diffs)

branches/HeuristicLab.BottomUpTreeDistance/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/SimilarityCalculators/BottomUpSimilarityCalculator.cs

@@ -22 +22 @@
 using System;
 using System.Collections.Generic;
-using System.Drawing;
-using System.Globalization;
 using System.Linq;
-using System.Text;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
 using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
-using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Views;
 using HeuristicLab.Optimization.Operators;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
@@ -78 +74 @@
 
       // visit nodes in order of decreasing height to ensure correct mapping
-      foreach (var v in n1.IterateNodesPrefix().OrderByDescending(x => compactedGraph[x].Weight)) {
+      foreach (var v in n1.IterateNodesPrefix().OrderByDescending(x => compactedGraph[x].Depth)) {
         if (forwardMap.ContainsKey(v))
           continue;
@@ -115 +111 @@
     /// Creates a compact representation of the two trees as a directed acyclic graph
     /// </summary>
-    /// <param name="t1">The first tree</param>
-    /// <param name="t2">The second tree</param>
+    /// <param name="n1">The root of the first tree</param>
+    /// <param name="n2">The root of the second tree</param>
     /// <returns>The compacted DAG representing the two trees</returns>
-    private Dictionary<ISymbolicExpressionTreeNode, IVertex> Compact(ISymbolicExpressionTreeNode n1, ISymbolicExpressionTreeNode n2) {
-      var nodesToVertices = new Dictionary<ISymbolicExpressionTreeNode, IVertex>(); // K
-      var labelsToVertices = new Dictionary<string, IVertex>(); // L
+    private Dictionary<ISymbolicExpressionTreeNode, GraphNode> Compact(ISymbolicExpressionTreeNode n1, ISymbolicExpressionTreeNode n2) {
+      var nodeMap = new Dictionary<ISymbolicExpressionTreeNode, GraphNode>(); // K
+      var labelMap = new Dictionary<string, GraphNode>(); // L
       var childrenCount = new Dictionary<ISymbolicExpressionTreeNode, int>(); // Children
-      var vertices = new List<IVertex>(); // G
 
       var nodes = n1.IterateNodesPostfix().Concat(n2.IterateNodesPostfix()); // the disjoint union F
+      var list = new List<GraphNode>();
       var queue = new Queue<ISymbolicExpressionTreeNode>();
 
@@ -130 +126 @@
         if (n.SubtreeCount == 0) {
           var label = n.ToString();
-          if (!labelsToVertices.ContainsKey(label)) {
-            var z = new Vertex { Content = n, Label = label };
-            labelsToVertices[z.Label] = z;
-            vertices.Add(z);
-          }
-          nodesToVertices[n] = labelsToVertices[label];
+          if (!labelMap.ContainsKey(label)) {
+            var z = new GraphNode { SymbolicExpressionTreeNode = n, Label = label };
+            labelMap[z.Label] = z;
+            list.Add(z);
+          }
+          nodeMap[n] = labelMap[label];
           queue.Enqueue(n);
         } else {
@@ -141 +137 @@
         }
       }
-
       while (queue.Any()) {
-        var v = queue.Dequeue();
-
-        if (v.SubtreeCount > 0) {
-          var label = v.Symbol.Name;
+        var n = queue.Dequeue();
+
+        if (n.SubtreeCount > 0) {
+          var label = n.Symbol.Name;
           bool found = false;
-          var height = v.GetDepth();
+          var depth = n.GetDepth();
 
           bool sort = commutativeSymbols.Contains(label);
-          var vSubtrees = v.Subtrees.Select(x => nodesToVertices[x]).ToList();
-          if (sort) vSubtrees.Sort((a, b) => String.Compare(a.Label, b.Label, StringComparison.Ordinal));
-
-          // for all nodes w in G in reverse order
-          for (int i = vertices.Count - 1; i >= 0; --i) {
-            var w = vertices[i];
-            var n = (ISymbolicExpressionTreeNode)w.Content;
-            if (v.SubtreeCount != n.SubtreeCount || label != w.Label || height != (int)w.Weight)
+          var nNodes = n.Subtrees.Select(x => nodeMap[x]).ToList();
+          if (sort) nNodes.Sort((a, b) => String.Compare(a.Label, b.Label, StringComparison.Ordinal));
+
+          for (int i = list.Count - 1; i >= 0; --i) {
+            var w = list[i];
+
+            if (!(n.SubtreeCount == w.ChildrenCount && label == w.Label && depth == w.Depth))
               continue;
 
             // sort V and W when the symbol is commutative because we are dealing with unordered trees
-            var wSubtrees = n.Subtrees.Select(x => nodesToVertices[x]).ToList();
-            if (sort) wSubtrees.Sort((a, b) => String.Compare(a.Label, b.Label, StringComparison.Ordinal));
-
-            if (vSubtrees.SequenceEqual(wSubtrees)) {
-              nodesToVertices[v] = w;
+            var m = w.SymbolicExpressionTreeNode;
+            var mNodes = m.Subtrees.Select(x => nodeMap[x]).ToList();
+            if (sort) mNodes.Sort((a, b) => String.Compare(a.Label, b.Label, StringComparison.Ordinal));
+
+            if (nNodes.SequenceEqual(mNodes)) {
+              nodeMap[n] = w;
               found = true;
               break;
             }
-          } // 32: end for
+          }
 
           if (!found) {
-            var w = new Vertex { Content = v, Label = label, Weight = height };
-            vertices.Add(w);
-            nodesToVertices[v] = w;
-
-            foreach (var u in v.Subtrees) {
-              AddArc(w, nodesToVertices[u]);
-            } // 40: end for
-          } // 41: end if
-        } // 42: end if
-
-        var p = v.Parent;
+            var w = new GraphNode { SymbolicExpressionTreeNode = n, Label = label, Depth = depth, ChildrenCount = n.SubtreeCount };
+            list.Add(w);
+            nodeMap[n] = w;
+          }
+        }
+
+        var p = n.Parent;
         if (p == null)
           continue;
@@ -193 +184 @@
       }
 
-      return nodesToVertices;
+      return nodeMap;
     }
 
@@ -202 +193 @@
         var n = list[i];
         if (n.SubtreeCount > 0) {
-          var subtrees = commutativeSymbols.Contains(node.Symbol.Name) ? n.Subtrees.OrderBy(s => s.ToString()) : n.Subtrees;
+          var subtrees = commutativeSymbols.Contains(node.Symbol.Name) ? n.Subtrees.OrderBy(x => x.ToString(), StringComparer.Ordinal) : n.Subtrees;
           list.AddRange(subtrees);
         }
@@ -210 +201 @@
     }
 
-    private static IArc AddArc(IVertex source, IVertex target) {
-      var arc = new Arc(source, target);
-      source.AddForwardArc(arc);
-      target.AddReverseArc(arc);
-      return arc;
-    }
-
-    // debugging
-    private static string FormatMapping(ISymbolicExpressionTree t1, ISymbolicExpressionTree t2, Dictionary<ISymbolicExpressionTreeNode, ISymbolicExpressionTreeNode> map) {
-      var symbolNameMap = new Dictionary<string, string>
-    {
-      {"ProgramRootSymbol", "Prog"},
-      {"StartSymbol","RPB"},
-      {"Multiplication", "$\\times$"},
-      {"Division", "$\\div$"},
-      {"Addition", "$+$"},
-      {"Subtraction", "$-$"},
-      {"Exponential", "$\\exp$"},
-      {"Logarithm", "$\\log$"}
-    };
-
-      var sb = new StringBuilder();
-      var nodeIds = new Dictionary<ISymbolicExpressionTreeNode, string>();
-      int offset = 0;
-      var layoutEngine = new ReingoldTilfordLayoutEngine<ISymbolicExpressionTreeNode>(x => x.Subtrees);
-      var nodeCoordinates = layoutEngine.CalculateLayout(t1.Root).ToDictionary(n => n.Content, n => new PointF(n.X, n.Y));
-
-      double ws = 0.5;
-      double hs = 0.5;
-
-      var nl = Environment.NewLine;
-      sb.Append("\\documentclass[class=minimal,border=0pt]{standalone}" + nl +
-                 "\\usepackage{tikz}" + nl +
-                 "\\begin{document}" + nl +
-                 "\\begin{tikzpicture}" + nl +
-                 "\\def\\ws{1}" + nl +
-                 "\\def\\hs{0.7}" + nl +
-                 "\\def\\offs{" + offset + "}" + nl);
-
-      foreach (var node in t1.IterateNodesBreadth()) {
-        var id = Guid.NewGuid().ToString();
-        nodeIds[node] = id;
-        var coord = nodeCoordinates[node];
-        var nodeName = symbolNameMap.ContainsKey(node.Symbol.Name) ? symbolNameMap[node.Symbol.Name] : node.ToString();
-        sb.AppendLine(string.Format(CultureInfo.InvariantCulture, "\\node ({0}) at (\\ws*{1} + \\offs,\\hs*{2}) {{{3}}};", nodeIds[node], ws * coord.X, -hs * coord.Y, EscapeLatexString(nodeName)));
-      }
-
-      foreach (ISymbolicExpressionTreeNode t in t1.IterateNodesBreadth()) {
-        var n = t;
-        foreach (var s in t.Subtrees) {
-          sb.AppendLine(string.Format(CultureInfo.InvariantCulture, "\\draw ({0}) -- ({1});", nodeIds[n], nodeIds[s]));
-        }
-      }
-
-      nodeCoordinates = layoutEngine.CalculateLayout(t2.Root).ToDictionary(n => n.Content, n => new PointF(n.X, n.Y));
-
-      offset = 20;
-      sb.Append("\\def\\offs{" + offset + "}" + nl);
-      foreach (var node in t2.IterateNodesBreadth()) {
-        var id = Guid.NewGuid().ToString();
-        nodeIds[node] = id;
-        var coord = nodeCoordinates[node];
-        var nodeName = symbolNameMap.ContainsKey(node.Symbol.Name) ? symbolNameMap[node.Symbol.Name] : node.ToString();
-        sb.AppendLine(string.Format(CultureInfo.InvariantCulture, "\\node ({0}) at (\\ws*{1} + \\offs,\\hs*{2}) {{{3}}};", nodeIds[node], ws * coord.X, -hs * coord.Y, EscapeLatexString(nodeName)));
-      }
-
-      foreach (ISymbolicExpressionTreeNode t in t2.IterateNodesBreadth()) {
-        var n = t;
-        foreach (var s in t.Subtrees) {
-          sb.AppendLine(string.Format(CultureInfo.InvariantCulture, "\\draw ({0}) -- ({1});", nodeIds[n], nodeIds[s]));
-        }
-      }
-
-      foreach (var p in map) {
-        var id1 = nodeIds[p.Key];
-        var id2 = nodeIds[p.Value];
-
-        sb.Append(string.Format(CultureInfo.InvariantCulture, "\\path[draw,->,color=gray] ({0}) edge[bend left,dashed] ({1});" + Environment.NewLine, id1, id2));
-      }
-      sb.Append("\\end{tikzpicture}" + nl +
-                "\\end{document}" + nl);
-      return sb.ToString();
-    }
-
-    private static string EscapeLatexString(string s) {
-      return s.Replace("\\", "\\\\").Replace("{", "\\{").Replace("}", "\\}").Replace("_", "\\_");
+    private class GraphNode {
+      public ISymbolicExpressionTreeNode SymbolicExpressionTreeNode;
+      public string Label;
+      public int Depth;
+      public int ChildrenCount;
     }
   }