/// <summary>
/// Мои привязки
/// </summary>
void InitializeMyBindings()
{
mainVM.PropertyChanged += (sender, e) =>
{
switch (e.PropertyName)
{
case "MatrixValues":
//обнулить кэш
mainUC = null;
normalDataUC = null;
normalizedDataUC = null;
normilizeStatisticsUC = null;
normalDistributionUC = null;
pairCorrelationsUC = null;
particalCorrelationsUC = null;
significanceCorrelationsUC = null;
multipleCorrelationUC = null;
correlationDiagramPageUC = null;
regressionUC = null;
regressionParamsControlUC = null;
break;
case "MatrixHeadersRegression":
//обнулить кэш
regressionUC = null;
break;
}
};
}
/// <summary>
/// Загрузить контент
/// </summary>
/// <param name="type"></param>
public void LoadView(ViewType type)
{
switch (type)
{
case ViewType.Main:
if (mainUC == null) //если требуется перерисовка
{
mainUC = new MainUC();
mainUC.DataContext = mainVM;
}
cpMainContent.Content = mainUC;
break;
case ViewType.Data:
if (normalDataUC == null)
{
normalDataUC = new DataTableUC();
normalDataUC.SetHeader("Исходные данные:");
normalDataUC.Table.SetTable(MatrixOperations.Round(MatrixOperations.Transpose(mainVM.MatrixValues), RoundConverter.Round), mainVM.MatrixHeaders);
}
cpMainContent.Content = normalDataUC;
break;
case ViewType.NormalizedData:
if (normalizedDataUC == null)
{
normalizedDataUC = new DataTableUC();
normalizedDataUC.SetHeader("Нормированные данные:");
normalizedDataUC.Table.SetTable(MatrixOperations.Round(MatrixOperations.Transpose(mainVM.MatrixNormalizedValues), RoundConverter.Round), mainVM.MatrixHeaders);
}
cpMainContent.Content = normalizedDataUC;
break;
case ViewType.NormalizedStatistic:
if (normilizeStatisticsUC == null)
{
normilizeStatisticsUC = new DataTableUC();
normilizeStatisticsUC.SetHeader("Описательная статистика для нормированной выборки:");
normilizeStatisticsUC.Table.SetTable(MatrixOperations.Round(MatrixOperations.Transpose(mainVM.MatrixNormalizedStatisticsValues), RoundConverter.Round), mainVM.MatrixHeaders, mainVM.StatisticsHeaders);
}
cpMainContent.Content = normilizeStatisticsUC;
break;
case ViewType.NormalDistribution:
if (normalDistributionUC == null)
{
normalDistributionUC = new DataTableUC();
normalDistributionUC.SetHeader($"Проверка гипотезы о нормальности распределения выборок (χ2-крит = {mainVM.ChiSquareCrit}):");
normalDistributionUC.Table.SetTable(mainVM.MatrixNormalDistribution, mainVM.MatrixHeaders, mainVM.NormalDistributionHeaders);
}
cpMainContent.Content = normalDistributionUC;
break;
case ViewType.PairCorrelations:
if (pairCorrelationsUC == null)
{
pairCorrelationsUC = new DataTableUC();
pairCorrelationsUC.SetHeader($"Матрица парных корреляций:");
pairCorrelationsUC.Table.SetTable(MatrixOperations.Round(mainVM.PairCorrelationsMatrix, RoundConverter.Round), mainVM.MatrixHeaders, mainVM.MatrixHeaders);
pairCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.2 && Math.Abs(e) < 0.4, Brushes.LightGreen);
pairCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.4 && Math.Abs(e) < 0.6, new SolidColorBrush(Color.FromArgb(0xFF, 0xDF, 0x9B, 0xFF)));
pairCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.6 && Math.Abs(e) < 0.8, Brushes.Orange);
pairCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.8 && Math.Abs(e) <= 1, Brushes.Red);
pairCorrelationsUC.Table.Highlight(e => Math.Abs(e) == 1, Brushes.Black);
pairCorrelationsUC.cpFooter.Content = new CorrelationColor();
}
cpMainContent.Content = pairCorrelationsUC;
break;
case ViewType.ParticalCorrelations:
if (particalCorrelationsUC == null)
{
particalCorrelationsUC = new DataTableUC();
particalCorrelationsUC.SetHeader($"Матрица частных корреляций:");
particalCorrelationsUC.Table.SetTable(MatrixOperations.Round(mainVM.ParticalCorrelationsMatrix, RoundConverter.Round), mainVM.MatrixHeaders, mainVM.MatrixHeaders);
particalCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.2 && Math.Abs(e) < 0.4, Brushes.LightGreen);
particalCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.4 && Math.Abs(e) < 0.6, new SolidColorBrush(Color.FromArgb(0xFF, 0xDF, 0x9B, 0xFF)));
particalCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.6 && Math.Abs(e) < 0.8, Brushes.Orange);
particalCorrelationsUC.Table.Highlight(e => Math.Abs(e) >= 0.8 && Math.Abs(e) < 1, Brushes.Red);
particalCorrelationsUC.Table.Highlight(e => Math.Abs(e) == 1, Brushes.Black);
particalCorrelationsUC.cpFooter.Content = new CorrelationColor();
}
cpMainContent.Content = particalCorrelationsUC;
break;
case ViewType.SignificanceCorrelations:
if (significanceCorrelationsUC == null)
{
significanceCorrelationsUC = new SignificanceUC();
significanceCorrelationsUC.SetHeader1($"Значимость коэффициентов парной корреляции (t-крит = {mainVM.TStudentCritSign}):");
significanceCorrelationsUC.Table1.SetTable(MatrixOperations.Round(mainVM.PairSignificanceCorrelationsMatrix, RoundConverter.Round), mainVM.MatrixHeaders, mainVM.MatrixHeaders);
significanceCorrelationsUC.Table1.Highlight(e => e >= mainVM.TStudentCritSign, Brushes.LightGreen);
significanceCorrelationsUC.SetHeader2($"Значимость коэффициентов частной корреляции (t-крит = {mainVM.TStudentCritSign}):");
significanceCorrelationsUC.Table2.SetTable(MatrixOperations.Round(mainVM.ParticalSignificanceCorrelationsMatrix, RoundConverter.Round), mainVM.MatrixHeaders, mainVM.MatrixHeaders);
significanceCorrelationsUC.Table2.Highlight(e => e >= mainVM.TStudentCritSign, Brushes.LightGreen);
}
cpMainContent.Content = significanceCorrelationsUC;
break;
case ViewType.MultipleCorrelation:
if (multipleCorrelationUC == null)
{
multipleCorrelationUC = new DataTableUC();
multipleCorrelationUC.SetHeader($"Множественная корреляция (F-крит = {mainVM.FCritSignMultiple}):");
double[][] table = mainVM.MultipleCorrelationMatrix;
multipleCorrelationUC.Table.SetTable(MatrixOperations.Round(table, RoundConverter.Round), mainVM.MatrixHeaders, new string[] { "R", "D", "Значимость" });
multipleCorrelationUC.Table.Highlight(e => e >= mainVM.FCritSignMultiple, Brushes.LightGreen, 2);
}
cpMainContent.Content = multipleCorrelationUC;
break;
case ViewType.CorrelativePleiad:
if (correlationDiagramPageUC == null)
{
CorrelationDiagramUC pairDiagram = new CorrelationDiagramUC(MatrixOperations.Round(mainVM.PairCorrelationsMatrix, RoundConverter.Round));
CorrelationDiagramUC particalDiagram = new CorrelationDiagramUC(MatrixOperations.Round(mainVM.ParticalCorrelationsMatrix, RoundConverter.Round));
correlationDiagramPageUC = new CorrelationDiagramMainUC();
correlationDiagramPageUC.SetPairDiagram(pairDiagram);
correlationDiagramPageUC.SetParticalDiagram(particalDiagram);
correlationDiagramPageUC.SetHeader("Диаграммы корреляционных плеяд для коэффициентов парной (слева) и частной (справа) корреляций:");
correlationDiagramPageUC.SetCorrelationColor(new CorrelationColor());
StringBuilder statHeader = new StringBuilder();
//--формирование statHeader--//
statHeader.AppendLine($"Y: {mainVM.MatrixHeaders[0]}");
for (int i = 1; i < mainVM.MatrixHeaders.Length; i++)
{
statHeader.AppendLine($"X{i}: {mainVM.MatrixHeaders[i]}");
}
//--//
correlationDiagramPageUC.SetStatHeader(statHeader.ToString());
correlationDiagramPageUC.SetCorrelationSign(new CorrelationSignDiagramUC());
}
cpMainContent.Content = correlationDiagramPageUC;
break;
case ViewType.Regression:
if (regressionUC == null)
{
regressionUC = new RegressionUC(mainVM);
//формирование коэффициентов уравнения
string[][] regressionCoeffs = new string[4][];
regressionCoeffs[0] = new string[] { "-" }.Concat(mainVM.MatrixHeadersRegression.Skip(1)).ToArray();
regressionCoeffs[1] = mainVM.RegressionCoeffs.Select(e => e.ToString()).ToArray();
regressionCoeffs[2] = Enumerable.Range(0, mainVM.RegressionCoeffs.Length).Select(idx => $"{Math.Round(mainVM.RegressionCoeffs[idx], RoundConverter.Round)} ± {Math.Round(mainVM.IntervalEstimateCoeffs[idx], RoundConverter.Round)}").ToArray();
regressionCoeffs[3] = mainVM.SignificanceEquationCoeffs.Select(e => e.ToString()).ToArray();
string[] rowHeader = { "Название параметра", "Коэффициент регрессии b", "Интервальная оценка β", "Значимость" };
string[] colHeader = new string[mainVM.MatrixHeadersRegression.Length].Select((e, idx) => $"{idx}").ToArray();
regressionUC.CoeffTable.SetTable(MatrixOperations.Round(regressionCoeffs, RoundConverter.Round), colHeader, rowHeader);
regressionUC.SetTKrit(mainVM.TCritEquationCoeffsSign);
regressionUC.CoeffTable.Highlight(e => e >= mainVM.TCritEquationCoeffsSign, Brushes.LightGreen, 3);
//--формирование уравнения регрессии в виде строки--//
string equation = "y = ";
double[] regressionCoeffsVMCopy = mainVM.RegressionCoeffs.Select(e => Math.Round(e, RoundConverter.Round)).ToArray();
for (int i = 1; i < regressionCoeffsVMCopy.Length; i++)
{
equation += regressionCoeffsVMCopy[i] < 0 ? $"({regressionCoeffsVMCopy[i]})⋅x{i} + " : $"{regressionCoeffsVMCopy[i]}⋅x{i} + ";
}
equation += regressionCoeffsVMCopy[0];
//--//
regressionUC.SetRegressionEquation(equation);
//значимость уравнения регрессии
regressionUC.SetSignificanceEquation(mainVM.FСritEquationSign, Math.Round(mainVM.SignificanceEquation, RoundConverter.Round));
//Формирование матрицы оценки точности уравнения (точечная)//
string[][] errors = new string[4][].Select(e => e = new string[mainVM.MatrixNormalizedValuesRegression[0].Length]).ToArray();
errors[0] = mainVM.MatrixNormalizedValuesRegression[0].Select(e => e.ToString()).ToArray();
errors[1] = mainVM.CalculatedY.Select(e => e.ToString()).ToArray();
errors[2] = mainVM.AbsoluteErrorY.Select(e => e.ToString()).ToArray();
errors[3] = Enumerable.Range(0, mainVM.CalculatedY.Length).Select(idx => $"{Math.Round(mainVM.CalculatedY[idx], RoundConverter.Round)} ± {Math.Round(mainVM.IntervalEstimateEquation[idx], RoundConverter.Round)}").ToArray();
//errors[4] = Enumerable.Range(0, mainVM.CalculatedY.Length).Select(idx => $"{Math.Round(mainVM.CalculatedY[idx], RoundConverter.Round)} ± {Math.Round(mainVM.IntervalPredicationEquation[idx], RoundConverter.Round)}").ToArray();
regressionUC.ErrorTable.SetTable(MatrixOperations.Round(errors, RoundConverter.Round), null, new string[] { "Y исходные", "Ŷ расчетные (Ŷ = X*b)", "Абсолютная ошибка (Y - Ŷ)", "Интервальная оценка Ỹ" });
regressionUC.SetApproximationError(Math.Round(mainVM.ApproximationError * 100, RoundConverter.Round).ToString());
//--//
//Формирование прогнозирования//
regressionUC.SetPredication(mainVM.RegressionCoeffs, mainVM.IntervalNormallized);
Label kek = new Label();
}
cpMainContent.Content = regressionUC;
break;
case ViewType.RegressionParamsControl:
if (regressionParamsControlUC == null)
{
regressionParamsControlUC = new RegressionParamsControlUC(mainVM);
regressionParamsControlUC.SetChoiceParamsTable(mainVM.MatrixHeaders);
}
cpMainContent.Content = regressionParamsControlUC;
break;
}
}
