/// <summary>
/// 计算变量重要性
/// </summary>
/// <param name="xy"></param>
/// <param name="oobavgerror"></param>
private void computeSignificant(double[,] xy, alglib.dfreport rep)
{
int dim1 = xy.GetLength(0); // 测试集大小
int dim2 = xy.GetLength(1); // 自变量个数+1
double[] results = new double[dim2 - 1];
int info;
alglib.decisionforest df;
alglib.dfreport rep1;
for (int i = 0; i < results.Length; i++)
{
double[,] xy1 = (double[, ])xy.Clone();
// 清空列
for (int j = 0; j < dim1; j++)
{
xy1[j, i] = 0;
}
alglib.dfbuildrandomdecisionforest(xy1, dim1, dim2 - 1, this.landInfo.NumOfLandUseTypes, this.NumOfTree, this.SampleRatio, out info, out df, out rep1);
results[i] = rep.oobavgce - rep.oobavgce;
//this.updateConsoleEvent("变量" + (i + 1) + "的重要性分数oobavgce是" + (rep.oobavgce - rep1.oobavgce));
//this.updateConsoleEvent("变量" + (i + 1) + "的重要性分数oobavgerror是" + (rep.oobavgerror - rep1.oobavgerror));
//this.updateConsoleEvent("变量" + (i + 1) + "的重要性分数oobavgrelerror是" + (rep.oobavgrelerror - rep1.oobavgrelerror));
//this.updateConsoleEvent("变量" + (i + 1) + "的重要性分数oobrelclserror是" + (rep.oobrelclserror - rep1.oobrelclserror));
this.updateConsoleEvent("变量" + (i + 1) + this.driveLayerNames[i] + "的重要性分数oobrmserror是" + (rep1.oobrmserror - rep.oobrmserror));
}
}
public void forestText()
{
//-----------------------------текст из поля ввода--------------------------------
string VocabularyTxt = "../../../data/vocabulary.txt";
vocabulary = File.ReadAllLines(VocabularyTxt);
double[] text = new double[vocabulary.Length];
Char[] separators = { ' ', ',', '.', '-', '\n' };
song = richTextBox1.Text.Split(separators);
int i = 0;
double[] textAnalys = { 0, 0 };
foreach (string str in vocabulary)
{
foreach (string str1 in song)
{
if (string.Equals(str, str1))
{
text[i] = text[i] + 1;
}
}
i++;
}
//----------------------------...................................текст из поля ввода
// Чтение CSV --------------------------------------------------------------------------------------------------
char separator = ';';
int flags = 0;
// Случайный лес решений --------------------------------------------------------------------------------------------------
var dForest1 = new alglib.decisionforest();
var rep = new alglib.dfreport();
int info;
alglib.read_csv("../../../data/trainText.csv", separator, flags, out TrainingSet);
// ---------------------------------------------------------------------------------------------------Чтение CSV
alglib.dfbuildrandomdecisionforestx1(TrainingSet, 40, 2945, 2, 200, 1, 0.9, out info, out dForest1, out rep);
alglib.dfprocess(dForest1, text, ref textAnalys);
label1.Text = "Random forest:\n" + "Result: " + info.ToString() + "\n"
+ "Веселый : " + textAnalys[0].ToString() + "\n"
+ "Грустный : " + textAnalys[1].ToString() + "\n";
// + "Романтический : " + textAnalys[2].ToString() + "\n"
// + "Философский : " + textAnalys[3].ToString() + "\n"
// + "Политический : " + textAnalys[4].ToString() + "\n"
// + "Бессмысленный : " + textAnalys[5].ToString() + "\n";
}
public void construire(double[,] lu)
{
rf = new alglib.decisionforest();
int npoints = longue - 1;
int nvars = larg - 1;
int nclasses = 1;
int ntrees = 100;
int info = 0;
double r = 0.66;
alglib.dfreport rep = new alglib.dfreport();
alglib.dfbuildrandomdecisionforest(lu, npoints, nvars, nclasses, ntrees, r, out info, out rf, out rep);
//Entrée, Samples(nombre de sets), nbFeatures20, nclasse4, trees 100, r 0.66,...
}
private static alglib.decisionforest CreateRandomForestModel(int seed, double[,] inputMatrix, int nTrees, double r, double m, int nClasses, out alglib.dfreport rep)
{
AssertParameters(r, m);
AssertInputMatrix(inputMatrix);
int info = 0;
alglib.math.rndobject = new System.Random(seed);
var dForest = new alglib.decisionforest();
rep = new alglib.dfreport();
int nRows = inputMatrix.GetLength(0);
int nColumns = inputMatrix.GetLength(1);
int sampleSize = Math.Max((int)Math.Round(r * nRows), 1);
int nFeatures = Math.Max((int)Math.Round(m * (nColumns - 1)), 1);
alglib.dforest.dfbuildinternal(inputMatrix, nRows, nColumns - 1, nClasses, nTrees, sampleSize, nFeatures, alglib.dforest.dfusestrongsplits + alglib.dforest.dfuseevs, ref info, dForest.innerobj, rep.innerobj);
if (info != 1)
{
throw new ArgumentException("Error in calculation of random forest model");
}
return(dForest);
}
private static alglib.decisionforest CreateRandomForestModel(int seed, double[,] inputMatrix, int nTrees, double r, double m, int nClasses, out alglib.dfreport rep) {
AssertParameters(r, m);
AssertInputMatrix(inputMatrix);
int info = 0;
alglib.math.rndobject = new System.Random(seed);
var dForest = new alglib.decisionforest();
rep = new alglib.dfreport();
int nRows = inputMatrix.GetLength(0);
int nColumns = inputMatrix.GetLength(1);
int sampleSize = Math.Max((int)Math.Round(r * nRows), 1);
int nFeatures = Math.Max((int)Math.Round(m * (nColumns - 1)), 1);
alglib.dforest.dfbuildinternal(inputMatrix, nRows, nColumns - 1, nClasses, nTrees, sampleSize, nFeatures, alglib.dforest.dfusestrongsplits + alglib.dforest.dfuseevs, ref info, dForest.innerobj, rep.innerobj);
if (info != 1) throw new ArgumentException("Error in calculation of random forest model");
return dForest;
}
