我试图使用修改使用2个类（true，false）的原始代码来使用4个类（unacc，acc，good，vgood）。原始代码假设＆＃34; boolean＆＃34;，因此要解决它我需要比较字符串到字符串。
由于某种原因，它返回以下错误对象引用未设置为对象的实例。我已经谷歌了，当我没有启动某些事情时就会发生这种情况。
错误：
Unhandled Exception: System.NullReferenceException: Object reference not set to an instance of an object.
at AA.TreeNode..ctor(Attribute attribute) in c:\VS\Program.cs:line 117
at AA.DecisionTreeID3.internalMountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 574
at AA.DecisionTreeID3.mountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 626
at AA.DecisionTreeID3.internalMountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 607
at AA.DecisionTreeID3.mountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 626
at AA.DecisionTreeID3.internalMountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 607
at AA.DecisionTreeID3.mountTree(DataTable samples, String targetAttribute, Attribute[] attributes) in c:\VS\Program.cs:line 626
at AA.ID3Sample.Main(String[] args) in c:\VS\Program.cs:line 718
原始代码/解决方案可用here
我将代码留在下面。 比较两个字符串可能是我的错误？ 可以创建一个函数作为字符串，最后用字符串返回值吗？如果没有，我可以将功能更改为＆＃34; int＆＃34;并将其作为值0,1,2和3返回。
有人可以帮助我或分享我如何解决这个问题的想法吗？
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Collections;
using System.Data;
namespace AA
{
/// <summary>
/// Classe que representa um atributo utilizado na classe de decisão
/// </summary>
public class Attribute
{
ArrayList mValues;
string mName;
object mLabel;
/// <summary>
/// Inicializa uma nova instância de uma classe Atribute
/// </summary>
/// <param name="name">Indica o nome do atributo</param>
/// <param name="values">Indica os valores possíveis para o atributo</param>
public Attribute(string name, string[] values)
{
mName = name;
mValues = new ArrayList(values);
mValues.Sort();
}
public Attribute(object Label)
{
mLabel = Label;
mName = string.Empty;
mValues = null;
}
/// <summary>
/// Indica o nome do atributo
/// </summary>
public string AttributeName
{
get
{
return mName;
}
}
/// <summary>
/// Retorna um array com os valores do atributo
/// </summary>
public string[] values
{
get
{
if (mValues != null)
return (string[])mValues.ToArray(typeof(string));
else
return null;
}
}
/// <summary>
/// Indica se um valor é permitido para este atributo
/// </summary>
/// <param name="value"></param>
/// <returns></returns>
public bool isValidValue(string value)
{
return indexValue(value) >= 0;
}
/// <summary>
/// Retorna o índice de um valor
/// </summary>
/// <param name="value">Valor a ser retornado</param>
/// <returns>O valor do índice na qual a posição do valor se encontra</returns>
public int indexValue(string value)
{
if (mValues != null)
return mValues.BinarySearch(value);
else
return -1;
}
/// <summary>
///
/// </summary>
/// <returns></returns>
public override string ToString()
{
if (mName != string.Empty)
{
return mName;
}
else
{
return mLabel.ToString();
}
}
}
/// <summary>
/// Classe que representará a arvore de decisão montada;
/// </summary>
public class TreeNode
{
private ArrayList mChilds = null;
private Attribute mAttribute;
/// <summary>
/// Inicializa uma nova instância de TreeNode
/// </summary>
/// <param name="attribute">Atributo ao qual o node está ligado</param>
public TreeNode(Attribute attribute)
{
if (attribute.values != null)
{
mChilds = new ArrayList(attribute.values.Length);
for (int i = 0; i < attribute.values.Length; i++)
mChilds.Add(null);
}
else
{
mChilds = new ArrayList(1);
mChilds.Add(null);
}
mAttribute = attribute;
}
/// <summary>
/// Adiciona um TreeNode filho a este treenode no galho de nome indicicado pelo ValueName
/// </summary>
/// <param name="treeNode">TreeNode filho a ser adicionado</param>
/// <param name="ValueName">Nome do galho onde o treeNode é criado</param>
public void AddTreeNode(TreeNode treeNode, string ValueName)
{
int index = mAttribute.indexValue(ValueName);
mChilds[index] = treeNode;
}
/// <summary>
/// Retorna o nro total de filhos do nó
/// </summary>
public int totalChilds
{
get
{
return mChilds.Count;
}
}
/// <summary>
/// Retorna o nó filho de um nó
/// </summary>
/// <param name="index">Indice do nó filho</param>
/// <returns>Um objeto da classe TreeNode representando o nó</returns>
public TreeNode getChild(int index)
{
return (TreeNode)mChilds[index];
}
/// <summary>
/// Atributo que está conectado ao Nó
/// </summary>
public Attribute attribute
{
get
{
return mAttribute;
}
}
/// <summary>
/// Retorna o filho de um nó pelo nome do galho que leva até ele
/// </summary>
/// <param name="branchName">Nome do galho</param>
/// <returns>O nó</returns>
public TreeNode getChildByBranchName(string branchName)
{
int index = mAttribute.indexValue(branchName);
return (TreeNode)mChilds[index];
}
}
/// <summary>
/// Classe que implementa uma árvore de Decisão usando o algoritmo ID3
/// </summary>
public class DecisionTreeID3
{
private DataTable mSamples;
private int mTotalUnacc = 0;
private int mTotalAcc = 0;
private int mTotalGood = 0;
private int mTotalVgood = 0;
private int mTotal = 0;
private string mTargetAttribute = "result";
private double mEntropySet = 0.0;
/// <summary>
/// Retorna o total de amostras positivas em uma tabela de amostras
/// </summary>
/// <param name="samples">DataTable com as amostras</param>
/// <returns>O nro total de amostras positivas</returns>
///
private int countTotalUnacc(DataTable samples)
{
int result = 0;
foreach (DataRow aRow in samples.Rows)
{
if (aRow[mTargetAttribute].ToString().ToUpper().Trim() == "UNACC")
result++;
}
return result;
}
private int countTotalAcc(DataTable samples)
{
int result = 0;
foreach (DataRow aRow in samples.Rows)
{
if (aRow[mTargetAttribute].ToString().ToUpper().Trim() == "ACC")
result++;
}
return result;
}
private int countTotalGood(DataTable samples)
{
int result = 0;
foreach (DataRow aRow in samples.Rows)
{
if (aRow[mTargetAttribute].ToString().ToUpper().Trim() == "GOOD")
result++;
}
return result;
}
private int countTotalVgood(DataTable samples)
{
int result = 0;
foreach (DataRow aRow in samples.Rows)
{
if (aRow[mTargetAttribute].ToString().ToUpper().Trim() == "VGOOD")
result++;
}
return result;
}
/// <summary>
/// Calcula a entropia dada a seguinte fórmula
/// -p+log2p+ - p-log2p-
///
/// onde: p+ é a proporção de valores positivos
/// p- é a proporção de valores negativos
/// </summary>
/// <param name="positives">Quantidade de valores positivos</param>
/// <param name="negatives">Quantidade de valores negativos</param>
/// <returns>Retorna o valor da Entropia</returns>
private double calcEntropy(int unacc, int acc,int good,int vgood)
{
//int total = positives + negatives;
int total = unacc + acc + good + vgood;
double ratioUnacc = (double)unacc / total;
double ratioAcc = (double)acc / total;
double ratioGood = (double)good / total;
double ratioVgood = (double)vgood / total;
if (ratioUnacc != 0)
ratioUnacc = -(ratioUnacc) * System.Math.Log(ratioUnacc, 2);
if (ratioAcc != 0)
ratioAcc = -(ratioAcc) * System.Math.Log(ratioAcc, 2);
if (ratioGood != 0)
ratioGood = -(ratioGood) * System.Math.Log(ratioGood, 2);
if (ratioVgood != 0)
ratioVgood = -(ratioVgood) * System.Math.Log(ratioVgood, 2);
double result = ratioUnacc + ratioAcc + ratioGood + ratioVgood;
return result;
}
/// <summary>
/// Varre tabela de amostras verificando um atributo e se o resultado é positivo ou negativo
/// </summary>
/// <param name="samples">DataTable com as amostras</param>
/// <param name="attribute">Atributo a ser pesquisado</param>
/// <param name="value">valor permitido para o atributo</param>
/// <param name="positives">Conterá o nro de todos os atributos com o valor determinado com resultado positivo</param>
/// <param name="negatives">Conterá o nro de todos os atributos com o valor determinado com resultado negativo</param>
private void getValuesToAttribute(DataTable samples, Attribute attribute, string value, out int unacc, out int acc, out int good, out int vgood)
{
unacc = 0;
acc = 0;
good = 0;
vgood = 0;
foreach (DataRow aRow in samples.Rows)
{
if (
((string)aRow[attribute.AttributeName] == value))
if ((string)aRow[mTargetAttribute] == "unacc")
unacc++;
else if ((string)aRow[mTargetAttribute] == "acc")
acc++;
else if ((string)aRow[mTargetAttribute] == "good")
good++;
else
vgood++;
}
}
/// <summary>
/// Calcula o ganho de um atributo
/// </summary>
/// <param name="attribute">Atributo a ser calculado</param>
/// <returns>O ganho do atributo</returns>
private double gain(DataTable samples, Attribute attribute)
{
string[] values = attribute.values;
double sum = 0.0;
for (int i = 0; i < values.Length; i++)
{
int unacc, acc, good, vgood;
unacc = acc = good = vgood = 0;
getValuesToAttribute(samples, attribute, values[i], out unacc, out acc, out good, out vgood);
double entropy = calcEntropy(unacc, acc,good,vgood);
sum += -(double)(unacc + acc + good + vgood) / mTotal * entropy;
}
return mEntropySet + sum;
}
/// <summary>
/// Retorna o melhor atributo.
/// </summary>
/// <param name="attributes">Um vetor com os atributos</param>
/// <returns>Retorna o que tiver maior ganho</returns>
private Attribute getBestAttribute(DataTable samples, Attribute[] attributes)
{
double maxGain = 0.0;
Attribute result = null;
foreach (Attribute attribute in attributes)
{
double aux = gain(samples, attribute);
if (aux > maxGain)
{
maxGain = aux;
result = attribute;
}
}
return result;
}
/// <summary>
/// Retorna true caso todos os exemplos da amostragem são positivos
/// </summary>
/// <param name="samples">DataTable com as amostras</param>
/// <param name="targetAttribute">Atributo (coluna) da tabela a qual será verificado</param>
/// <returns>True caso todos os exemplos da amostragem são positivos</returns>
private string allSamplesUnacc(DataTable samples, string targetAttribute)
{
foreach (DataRow row in samples.Rows)
{ //alterar
if (row[targetAttribute].ToString() == "acc")
return "acc";
if (row[targetAttribute].ToString() == "good")
return "good";
if (row[targetAttribute].ToString() == "vgood")
return "vgood";
}
return "unacc";
}
private string allSamplesAcc(DataTable samples, string targetAttribute)
{
foreach (DataRow row in samples.Rows)
{ //alterar
if (row[targetAttribute].ToString() == "unacc")
return "unacc";
if (row[targetAttribute].ToString() == "good")
return "good";
if (row[targetAttribute].ToString() == "vgood")
return "vgood";
}
return "acc";
}
private string allSamplesGood(DataTable samples, string targetAttribute)
{
foreach (DataRow row in samples.Rows)
{ //alterar
if (row[targetAttribute].ToString() == "unacc")
return "unacc";
if (row[targetAttribute].ToString() == "acc")
return "acc";
if (row[targetAttribute].ToString() == "vgood")
return "vgood";
}
return "good";
}
private string allSamplesVgood(DataTable samples, string targetAttribute)
{
foreach (DataRow row in samples.Rows)
{ //alterar
if (row[targetAttribute].ToString() == "unacc")
return "unacc";
if (row[targetAttribute].ToString() == "acc")
return "acc";
if (row[targetAttribute].ToString() == "good")
return "good";
}
return "vgood";
}
/// <summary>
/// Retorna uma lista com todos os valores distintos de uma tabela de amostragem
/// </summary>
/// <param name="samples">DataTable com as amostras</param>
/// <param name="targetAttribute">Atributo (coluna) da tabela a qual será verificado</param>
/// <returns>Um ArrayList com os valores distintos</returns>
private ArrayList getDistinctValues(DataTable samples, string targetAttribute)
{
ArrayList distinctValues = new ArrayList(samples.Rows.Count);
foreach (DataRow row in samples.Rows)
{
if (distinctValues.IndexOf(row[targetAttribute]) == -1)
distinctValues.Add(row[targetAttribute]);
}
return distinctValues;
}
/// <summary>
/// Retorna o valor mais comum dentro de uma amostragem
/// </summary>
/// <param name="samples">DataTable com as amostras</param>
/// <param name="targetAttribute">Atributo (coluna) da tabela a qual será verificado</param>
/// <returns>Retorna o objeto com maior incidência dentro da tabela de amostras</returns>
private object getMostCommonValue(DataTable samples, string targetAttribute)
{
ArrayList distinctValues = getDistinctValues(samples, targetAttribute);
int[] count = new int[distinctValues.Count];
foreach (DataRow row in samples.Rows)
{
int index = distinctValues.IndexOf(row[targetAttribute]);
count[index]++;
}
int MaxIndex = 0;
int MaxCount = 0;
for (int i = 0; i < count.Length; i++)
{
if (count[i] > MaxCount)
{
MaxCount = count[i];
MaxIndex = i;
}
}
return distinctValues[MaxIndex];
}
/// <summary>
/// Monta uma árvore de decisão baseado nas amostragens apresentadas
/// </summary>
/// <param name="samples">Tabela com as amostragens que serão apresentadas para a montagem da árvore</param>
/// <param name="targetAttribute">Nome da coluna da tabela que possue o valor true ou false para
/// validar ou não uma amostragem</param>
/// <returns>A raiz da árvore de decisão montada</returns></returns?>
private TreeNode internalMountTree(DataTable samples, string targetAttribute, Attribute[] attributes)
{
//alterar
if (allSamplesUnacc(samples, targetAttribute) == "unacc")
return new TreeNode(new Attribute("unacc"));
if (allSamplesAcc(samples, targetAttribute) == "acc")
return new TreeNode(new Attribute("acc"));
if (allSamplesGood(samples, targetAttribute) == "good")
return new TreeNode(new Attribute("good"));
if (allSamplesVgood(samples, targetAttribute) == "vgood")
return new TreeNode(new Attribute("vgood"));
if (attributes.Length == 0)
return new TreeNode(new Attribute(getMostCommonValue(samples, targetAttribute)));
mTotal = samples.Rows.Count;
mTargetAttribute = targetAttribute;
mTotalUnacc = countTotalUnacc(samples);
mTotalAcc = countTotalAcc(samples);
mTotalGood = countTotalGood(samples);
mTotalVgood = countTotalVgood(samples);
mEntropySet = calcEntropy(mTotalUnacc, mTotalAcc, mTotalGood, mTotalVgood);
Attribute bestAttribute = getBestAttribute(samples, attributes);
TreeNode root = new TreeNode(bestAttribute);
DataTable aSample = samples.Clone();
foreach (string value in bestAttribute.values)
{
// Seleciona todas os elementos com o valor deste atributo
aSample.Rows.Clear();
DataRow[] rows = samples.Select(bestAttribute.AttributeName + " = " + "'" + value + "'");
foreach (DataRow row in rows)
{
aSample.Rows.Add(row.ItemArray);
}
// Seleciona todas os elementos com o valor deste atributo
// Cria uma nova lista de atributos menos o atributo corrente que é o melhor atributo
ArrayList aAttributes = new ArrayList(attributes.Length - 1);
for (int i = 0; i < attributes.Length; i++)
{
if (attributes[i].AttributeName != bestAttribute.AttributeName)
aAttributes.Add(attributes[i]);
}
// Cria uma nova lista de atributos menos o atributo corrente que é o melhor atributo
if (aSample.Rows.Count == 0)
{
return new TreeNode(new Attribute(getMostCommonValue(aSample, targetAttribute)));
}
else
{
DecisionTreeID3 dc3 = new DecisionTreeID3();
TreeNode ChildNode = dc3.mountTree(aSample, targetAttribute, (Attribute[])aAttributes.ToArray(typeof(Attribute)));
root.AddTreeNode(ChildNode, value);
}
}
return root;
}
/// <summary>
/// Monta uma árvore de decisão baseado nas amostragens apresentadas
/// </summary>
/// <param name="samples">Tabela com as amostragens que serão apresentadas para a montagem da árvore</param>
/// <param name="targetAttribute">Nome da coluna da tabela que possue o valor true ou false para
/// validar ou não uma amostragem</param>
/// <returns>A raiz da árvore de decisão montada</returns></returns?>
public TreeNode mountTree(DataTable samples, string targetAttribute, Attribute[] attributes)
{
mSamples = samples;
return internalMountTree(mSamples, targetAttribute, attributes);
}
}
/// <summary>
/// Classe que exemplifica a utilização do ID3
/// </summary>
class ID3Sample
{
public static void printNode(TreeNode root, string tabs)
{
Console.WriteLine(tabs + '|' + root.attribute + '|');
if (root.attribute.values != null)
{
for (int i = 0; i < root.attribute.values.Length; i++)
{
Console.WriteLine(tabs + "\t" + "<" + root.attribute.values[i] + ">");
TreeNode childNode = root.getChildByBranchName(root.attribute.values[i]);
printNode(childNode, "\t" + tabs);
}
}
}
static DataTable getDataTable()
{
DataTable result = new DataTable("samples");
DataColumn column = result.Columns.Add("buying");
column.DataType = typeof(string);
column = result.Columns.Add("maint");
column.DataType = typeof(string);
column = result.Columns.Add("doors");
column.DataType = typeof(string);
column = result.Columns.Add("persons");
column.DataType = typeof(string);
column = result.Columns.Add("lugboot");
column.DataType = typeof(string);
column = result.Columns.Add("safety");
column.DataType = typeof(string);
column = result.Columns.Add("result");
column.DataType = typeof(string);
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "small", "high", "acc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "med", "low", "unacc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "med", "med", "unacc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "med", "high", "acc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "big", "low", "unacc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "big", "med", "acc" });
result.Rows.Add(new object[] { "vhigh", "med", "2", "4", "big", "high", "acc" });
result.Rows.Add(new object[] { "med", "low", "2", "4", "big", "high", "vgood" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "small", "low", "unacc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "small", "med", "unacc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "small", "high", "unacc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "med", "low", "unacc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "med", "med", "acc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "med", "high", "good" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "big", "low", "unacc" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "big", "med", "good" });
result.Rows.Add(new object[] { "med", "low", "2", "more", "big", "high", "vgood" });
return result;
}
/// <summary>
/// The main entry point for the application.
/// </summary>
///
[STAThread]
static void Main(string[] args)
{
Attribute buying = new Attribute("buying", new string[] { "vhigh", "high", "med", "low" });
Attribute maint = new Attribute("maint", new string[] { "vhigh", "high", "med", "low" });
Attribute doors = new Attribute("doors", new string[] { "2", "3", "4", "5more" });
Attribute persons = new Attribute("persons", new string[] { "2", "4", "more" });
Attribute lugboot = new Attribute("lugboot", new string[] { "small", "med", "big" });
Attribute safety = new Attribute("safety", new string[] { "low", "med", "high" });
Attribute[] attributes = new Attribute[] { buying, maint, doors, persons, lugboot, safety };
DataTable samples = getDataTable();
DecisionTreeID3 id3 = new DecisionTreeID3();
TreeNode root = id3.mountTree(samples, "result", attributes);
printNode(root, "");
}
}
}
答案 0 :(得分：0)
您的堆栈跟踪包含您需要的信息：
AA.TreeNode..ctor(Attribute attribute) in c:\VS\Program.cs:line 117
'ctor'是构造函数的花哨编译器俚语，因此我们将看一下TreeNode（Attribute）。
这可能意味着attribute（构造函数的参数）为null。那是怎么发生的？由于我们已经将所有代码放在了我们面前，让我们专注于一个没有显式创建新的Attribute对象以传递给TreeNode构造函数的调用：
Attribute bestAttribute = getBestAttribute(samples, attributes);
TreeNode root = new TreeNode(bestAttribute);
getBestAttribute将返回null。要修复异常，您需要修复该方法的输入或逻辑。