source: branches/2886_SymRegGrammarEnumeration/ExpressionClustering/Program.cs @ 15866

using System;
using System.Collections.Generic;
using System.Drawing;
using System.IO;
using System.Linq;
using HeuristicLab.Analysis;
using HeuristicLab.Analysis.Views;
using System.Windows.Forms;

namespace ExpressionClustering {
  class Program {
    private static readonly string folder = Environment.GetFolderPath(Environment.SpecialFolder.DesktopDirectory);
    private static readonly string clusterFolder = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.DesktopDirectory), "clusters");
    private static readonly string distinctSentencesFileName = Path.Combine(folder, @"distinctSentences.csv");
    private static readonly string allSentencesFileName = Path.Combine(folder, "allSentences.csv");
    private static readonly string outputFileName = Path.Combine(folder, "evaluations.csv.gz");
    private static int N = 100;
    private static double[] evalBuf = new double[N];

    // pagie-1 (univariate)
    private static double min = -5.0;
    private static double max = +5.0;
    private static double[] xs = Enumerable.Range(1, N).Select(xi => ((double)xi / N) * (max - min) + min).ToArray(); // input
    private static double[] ys = xs.Select(xi => 1.0 / (1 + Math.Pow(xi, -4))).ToArray(); // target (necessary for scaling and clustering 


    public static int MAX_STACK = 20;
    public static double[][] stack = new double[MAX_STACK][];
    static Program() {
      for (int i = 0; i < MAX_STACK; i++)
        stack[i] = new double[N];
    }

    // loads symbolic expressions in postfix notation from a stream and identifies clusters of expressions
    static void Main(string[] args) {

      var hash2Sentences = new Dictionary<string, List<string>>();
      // for debugging only
      var postfix2infix = new Dictionary<string, string>();



      // read all sentences and determine shortest sentences
      using (var reader = new StreamReader(allSentencesFileName)) {
        // read header
        reader.ReadLine();
        int nSentences = 0;
        while (!reader.EndOfStream) {
          var line = reader.ReadLine();
          var toks = line.Split(';');
          var hash = toks[2];
          List<string> ls;
          if (!hash2Sentences.TryGetValue(hash, out ls)) {
            ls = new List<string>(1);
            hash2Sentences.Add(hash, ls);
          }
          ls.Add(toks[1]);
          postfix2infix.Add(toks[1], toks[0]);
          nSentences++;
        }

        Console.WriteLine("{0} {1}", nSentences, hash2Sentences.Count);
        //Evaluate(toks[1], xs, evalBuf);
      }

      List<double[]> functions = new List<double[]>();
      List<string> sentences = new List<string>();
      List<double> qualities = new List<double>();

      foreach (var kvp in hash2Sentences) {
        var ls = kvp.Value;
        var sentence = FindShortest(ls);
        Evaluate(sentence, xs, evalBuf);
        if (evalBuf.Any(ei => float.IsInfinity((float)ei) || float.IsNaN((float)ei))) {
          Console.WriteLine("skipping {0} {1}", evalBuf.Average(), sentence);
        } else {
          try {
            Scale(evalBuf, ys);
            functions.Add((double[])evalBuf.Clone());
            sentences.Add(sentence);
            HeuristicLab.Problems.DataAnalysis.OnlineCalculatorError error;
            qualities.Add(HeuristicLab.Problems.DataAnalysis.OnlinePearsonsRSquaredCalculator.Calculate(evalBuf, ys, out error));
          } catch (ArgumentException e) {
            // scaling failed
          }
        }
      }

      List<int> clusters;
      List<double> distances;
      Flann.FindClusters(functions, out clusters, out distances, 100);
      
      // output all clusters and functions
      using (var writer = new StreamWriter(new System.IO.Compression.GZipStream(new FileStream(outputFileName, FileMode.OpenOrCreate), System.IO.Compression.CompressionMode.Compress))) {
        for (int i = 0; i < functions.Count; i++) {
          writer.WriteLine("{0};{1};{2};{3};{4};{5}", clusters[i], distances[i], qualities[i], sentences[i], postfix2infix[sentences[i]], string.Join(";", functions[i].Select(fi => fi.ToString())));
        }
      }

      var funClusters = functions.Zip(clusters, (f, c) => Tuple.Create(f, c)).GroupBy(t => t.Item2);
      var dtView = new DataTableView();
      dtView.Size = new Size(800, 600);

      foreach (var funCluster in funClusters) {
        // draw the functions for each cluster into a separate png
        var dtName = string.Format("R² {0}", Enumerable.Range(0, qualities.Count).Where(idx => clusters[idx] == funCluster.Key).Select(idx => qualities[idx]).Average());
        var dt = new DataTable(dtName, dtName);
        var rows = new List<DataRow>();
        int i = 0;
        foreach (var fun in funCluster.Select(t => t.Item1)) {
          var name = i.ToString();
          var dr = new DataRow(name, name, fun);
          rows.Add(dr);
          i++;
        }
        dt.Rows.AddRange(rows);
        dtView.Content = dt;
        using (var bm = new Bitmap(800, 600)) {
          dtView.DrawToBitmap(bm, new Rectangle(0, 0, 800, 600));
          bm.Save(Path.Combine(clusterFolder, string.Format("cluster_{0,3}.png", funCluster.Key)));
        }
      }
    }



    private static string FindShortest(List<string> ls) {
      var minElem = ls.First();
      for (int i = 1; i < ls.Count; i++) {
        if (ls[i].Length < minElem.Length) minElem = ls[i];
      }
      return minElem;
    }



    #region evaluation
    // linear scaling
    private static void Scale(double[] evalBuf, double[] ys) {
      double alpha;
      double beta;
      HeuristicLab.Problems.DataAnalysis.OnlineCalculatorError error;

      HeuristicLab.Problems.DataAnalysis.OnlineLinearScalingParameterCalculator.Calculate(evalBuf, ys, out alpha, out beta, out error);
      if (error != HeuristicLab.Problems.DataAnalysis.OnlineCalculatorError.None) {
        throw new ArgumentException();
      }

      for (int i = 0; i < evalBuf.Length; i++) {
        evalBuf[i] = beta * evalBuf[i] + alpha;
      }
    }

    // evaluates postfix expressions (only for a very specific format)
    private static void Evaluate(string postfixExpr, double[] xs, double[] evalBuf) {
      int topOfStack = -1;
      Evaluate(postfixExpr, 0, xs, ref topOfStack);
      Array.Copy(stack[topOfStack], evalBuf, evalBuf.Length);
    }

    private static void Evaluate(string postfixExpr, int exprPos, double[] xs, ref int topOfStack) {
      while (exprPos < postfixExpr.Length) {
        switch (postfixExpr[exprPos]) {
          case '+': {
              exprPos += 2;
              var a = stack[topOfStack];
              var b = stack[topOfStack - 1];
              for (int i = 0; i < N; i++) {
                b[i] += a[i];
              }
              topOfStack--;
              break;
            }
          case '*': {
              exprPos += 2;
              var a = stack[topOfStack];
              var b = stack[topOfStack - 1];
              for (int i = 0; i < N; i++) {
                b[i] *= a[i];
              }
              topOfStack--;
              break;
            }
          case 'X': {
              exprPos += 2;
              topOfStack++;
              Array.Copy(xs, stack[topOfStack], N);
              break;
            }
          case 'c': {
              // cos
              exprPos += 4;
              var a = stack[topOfStack];
              for (int i = 0; i < N; i++) {
                a[i] = Math.Cos(a[i]);
              }
              break;
            }
          case 's': {
              // sin
              exprPos += 4;
              var a = stack[topOfStack];
              for (int i = 0; i < N; i++) {
                a[i] = Math.Sin(a[i]);
              }
              break;
            }
          case 'l': {
              // log
              exprPos += 4;
              var a = stack[topOfStack];
              for (int i = 0; i < N; i++) {
                a[i] = Math.Log(a[i]);
              }

              break;
            }
          case 'e': {
              // exp
              exprPos += 4;
              var a = stack[topOfStack];
              for (int i = 0; i < N; i++) {
                a[i] = Math.Exp(a[i]);
              }

              break;
            }
          case 'i': {
              // inv
              exprPos += 4;
              var a = stack[topOfStack];
              for (int i = 0; i < N; i++) {
                a[i] = 1.0 / a[i];
              }
              break;
            }
          default: {
              throw new InvalidOperationException(string.Format("Cannot handle {0} in {1}", postfixExpr[exprPos], postfixExpr));
            }
        }
      }
    }
    #endregion
  }
}