public void MAETest()
{
var actual = new Matrix(4, 1);
var expected = new Matrix(4, 1);
actual.InRandomize();
expected.InRandomize();
var autoErr = new MAE().Evaluate(actual, expected);
var error = 0.0;
for (var i = 0; i < actual.Rows; i++)
{
for (var j = 0; j < actual.Columns; j++)
{
error += Math.Abs(expected[i, j] - actual[i, j]);
}
}
error /= actual.Rows * actual.Columns;
Assert.IsTrue(Math.Abs(error - autoErr) < 0.01, new MAE().Type().ToString() + " Forward!");
var autoDErr = new MAE().Backward(actual, expected);
var dErr = new Matrix(actual.Rows, actual.Columns);
for (var i = 0; i < actual.Rows; i++)
{
for (var j = 0; j < actual.Columns; j++)
{
dErr[i, j] = Math.Sign(actual[i, j] - expected[i, j]);
}
}
Assert.IsTrue(Math.Abs(dErr.FrobeniusNorm() - autoDErr.FrobeniusNorm()) < 0.01, new MAE().Type().ToString() + " Backward!");
}
public string RunNMFbasedOMF(int maxEpoch, double learnRate, double regularization, int factorCount,
List <double> quantizer, int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("NMF based OMF"));
// NMF Prediction
// Get ratings from scorer, for both train and test
// R_all contains indexes of all ratings both train and test
DataMatrix R_all = new DataMatrix(R_unknown.UserCount, R_unknown.ItemCount);
R_all.MergeNonOverlap(R_unknown);
R_all.MergeNonOverlap(R_train.IndexesOfNonZeroElements());
Utils.StartTimer();
DataMatrix R_predictedByNMF = NMF.PredictRatings(R_train, R_all, maxEpoch,
learnRate, regularization, factorCount);
log.AppendLine(Utils.StopTimer());
// OMF Prediction
log.AppendLine(Utils.PrintHeading("Ordinal Matrix Factorization with NMF as scorer"));
Utils.StartTimer();
Dictionary <Tuple <int, int>, List <double> > OMFDistributionByUserItem;
DataMatrix R_predicted;
log.AppendLine(OMF.PredictRatings(R_train.Matrix, R_unknown.Matrix, R_predictedByNMF.Matrix,
quantizer, out R_predicted, out OMFDistributionByUserItem));
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
// Save OMFDistribution to file
if (!File.Exists(GetDataFileName("RatingOMF_")))
{
Utils.IO <Dictionary <Tuple <int, int>, List <double> > > .SaveObject(OMFDistributionByUserItem, GetDataFileName("RatingOMF_"));
}
return(log.ToString());
}
/// <summary>
/// Rating based Non-negative Matrix Factorization
/// </summary>
public string RunNMF(int maxEpoch, double learnRate, double regularization,
int factorCount, int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("NMF"));
// Prediction
Utils.StartTimer();
DataMatrix R_predicted = NMF.PredictRatings(R_train, R_unknown, maxEpoch,
learnRate, regularization, factorCount);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
return(log.ToString());
}
/// <summary>
/// Predict all unknown values as global mean rating.
/// </summary>
public string RunGlobalMean()
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("Global Mean"));
// Prediction
Utils.StartTimer();
double globalMean = R_train.GetGlobalMean();
DataMatrix R_predicted = R_unknown.Multiply(globalMean);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
return(log.ToString());
}
public string RunUserKNN(int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("UserKNN"));
// Prediction
Utils.StartTimer();
DataMatrix R_predicted = Numerical.UserKNN.PredictRatings(R_train, R_unknown, UserSimilaritiesOfRating, MaxCountOfNeighbors);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000"));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
return(log.ToString());
}
// 运行一次算法
private void button1_Click(object sender, EventArgs e)
{
// 记录当前时间
DateTime dt_1 = DateTime.Now;
this.Rec_Items_num = int.Parse(this.textBox13.Text); // Top-N推荐个数
this.progressBar1.Maximum = this.comboBox2.SelectedIndex + 15 + 22;
this.progressBar1.Value = 0;
// 读入数据，生成UI矩阵
this.textBox3.Text = "开始读入数据";
this.progressBar1.Value++;
Application.DoEvents();
cReadinData obj_ReadData = new cReadinData(comboBox1.SelectedIndex);
this.textBox3.Text = "读入数据完成 训练数据：" + obj_ReadData.sTrainFileName[comboBox1.SelectedIndex]
+ " 测试数据：" + obj_ReadData.testfileName[comboBox1.SelectedIndex];
this.progressBar1.Value++;
Application.DoEvents();
int number = int.Parse(textBox2.Text);
this.neigh_num = number;
// 相似度算法的选择
if (this.radioButton1.Checked)
{
sim_alg = 1;
}
else if (this.radioButton2.Checked)
{
sim_alg = 2;
}
else if (this.radioButton3.Checked)
{
sim_alg = 3;
}
// 测试用户数目，最少为15
testUserNum = this.comboBox2.SelectedIndex + 15;
this.textBox3.Text = "初始化最大邻居个数";
// 初始化，最近邻居数量最大为200
obj_ItemBased_CF = new cItemBased_CF(200);
this.progressBar1.Value++;
Application.DoEvents();
// 初始化相关数据
for (int i = 1; i <= 1682; i++)
{
this.textBox3.Text = "初始 " + i + " 个项目数据";
if ((i % 100) == 0)
{
this.progressBar1.Value++;
Application.DoEvents();
}
Application.DoEvents();
obj_ItemBased_CF.generateItemNN(i);
}
this.progressBar1.Value++;
Application.DoEvents();
this.textBox3.Text = "读取最近邻居及其相似值文件";
obj_ItemBased_CF.readFile();
this.progressBar1.Value++;
Application.DoEvents();
// 得到测试用户集合
cUser[] testUsers = cReadinData.getTestUser();
this.textBox3.Text = "读取测试用户集合";
this.progressBar1.Value++;
Application.DoEvents();
double MAE, Precison, Recall, F_Measure;
double total_MAE = 0, total_Precison = 0, total_Recall = 0, total_F_Measure = 0;
double average_MAE, average_Precison, average_Recall, average_F_Measure;
// for循环为每个测试用户产生预测评分以及Top-N推荐，并取得算法评价指标
for (int i = 1; i <= testUserNum; i++)
{
obj_AssStrategy = obj_ItemBased_CF.getPredictRating(testUsers[i], this.sim_alg, neigh_num, Rec_Items_num);
MAE = obj_AssStrategy.MAE;
Precison = obj_AssStrategy.Precison;
Recall = obj_AssStrategy.Recall;
F_Measure = obj_AssStrategy.calculateF_Measure();
// 累计各项指标的和
total_MAE += MAE;
total_Precison += Precison;
total_Recall += Recall;
total_F_Measure += F_Measure;
this.textBox3.Text = "第" + i.ToString() + "个用户 MAE:" + MAE.ToString() + " 查准率:" + Precison
+ " 查全率:" + Recall + " F值:" + F_Measure;
this.progressBar1.Value++;
Application.DoEvents();
}
// 计算各项指标的平均值
average_MAE = total_MAE / this.testUserNum;
average_Precison = total_Precison / this.testUserNum;
average_Recall = total_Recall / this.testUserNum;
average_F_Measure = total_F_Measure / this.testUserNum;
DateTime dt_2 = DateTime.Now;
TimeSpan ts = dt_2.Subtract(dt_1);
this.textBox3.Text = "完成 平均MAE:" + average_MAE.ToString() + " 平均查准率:" + average_Precison +
" 平均查全率:" + average_Recall + " 平均F值:" + average_F_Measure + " 总耗时:" + ts.TotalMilliseconds + "ms";
Application.DoEvents();
this.textBox4.Text = average_MAE.ToString();
this.textBox5.Text = ts.TotalMilliseconds + "ms";
this.textBox6.Text = this.sSimAlg[this.sim_alg - 1];
this.textBox7.Text = this.neigh_num.ToString();
this.textBox9.Text = average_Precison.ToString();
this.textBox10.Text = average_Recall.ToString();
this.textBox11.Text = average_F_Measure.ToString();
this.textBox12.Text = Rec_Items_num.ToString();
string log = this.sSimAlg[this.sim_alg - 1] + " 邻居数:" + this.neigh_num.ToString() + " 平均MAE:" +
average_MAE.ToString() + " 平均查准率:" + average_Precison + " 平均查全率:" + average_Recall +
" 平均F值:" + average_F_Measure + " 总耗时:" + ts.TotalMilliseconds + "ms";
this.dataGridView1.Rows.Add(count_dgv++, sSimAlg[sim_alg - 1], this.neigh_num, this.Rec_Items_num, average_MAE,
average_Precison, average_Recall, average_F_Measure, (ts.TotalMilliseconds / this.testUserNum) + " ms");
}