private void savePlotsToolStripMenuItem_Click(object sender, EventArgs e)
{
var sfd = new SaveFileDialog();
sfd.InitialDirectory = @"H: \Users\Jack\Source\Repos\BraitenbergSimulator\BraitenbergSimulator\BraitenbergSimulator";
sfd.Filter = "PNG files (*.png)|*.png|All files (.*)|*.*";
sfd.FilterIndex = 0;
sfd.RestoreDirectory = true;
if (sfd.ShowDialog() == DialogResult.OK)
{
var pngExporter = new PngExporter { Width = 500, Height = 500, Background = OxyColors.White };
pngExporter.ExportToFile(plotView1.Model, sfd.FileName);
}
}
private void saveButton_Click(object sender, EventArgs e)
{
var sfd = new SaveFileDialog();
sfd.InitialDirectory = @"H: \Users\Jack\Source\Repos\BraitenbergSimulator\BraitenbergSimulator\BraitenbergSimulator";
sfd.Filter = "PNG files (*.png)|*.png|All files (.*)|*.*";
sfd.FilterIndex = 0;
sfd.RestoreDirectory = true;
if (sfd.ShowDialog() == DialogResult.OK)
{
var pngExporter = new PngExporter { Width = 500, Height = 500, Background = OxyColors.White };
pngExporter.ExportToFile(plotView1.Model, sfd.FileName);
string subName = Path.Combine(Path.GetDirectoryName(sfd.FileName), Path.GetFileNameWithoutExtension(sfd.FileName) + "pip" + Path.GetExtension(sfd.FileName));
pngExporter.ExportToFile(plotView2.Model, subName);
using (var writer = File.CreateText(Path.ChangeExtension(sfd.FileName, "yaml")))
{
Serializer s = new Serializer(namingConvention: new CamelCaseNamingConvention());
s.Serialize(writer, stats);
}
}
}
private void btn_CreateResultTable_Click(object sender, EventArgs e)
{
OxyColor EEGColor = OxyColors.Cyan;
OxyColor EDAColor = OxyColors.LightGreen;
OxyColor HRColor = OxyColors.Salmon;
//var tester = new List<double>
//{
// 0.01
//};
//var tester2 = new List<double>
//{
// 0.01, 0.01
//};
//var tester3 = new List<double>
//{
// 0.01, 0.01, 0.01
//};
//MessageBox.Show(tester.FisherCombineP().ToString("0.00000000") + "\n" + tester2.FisherCombineP().ToString("0.00000000") + "\n" + tester3.FisherCombineP().ToString("0.00000000"));
//var res1 = FisherCompare1(0.4, 10, 0.3, 12);
//var res2 = FisherCompare2(0.4, 10, 0.3, 12);
//MessageBox.Show("1)\n" + res1.Item1 + "\n" + res1.Item2 + "\n" + res1.Item3 + "\n\n2)\n" + res2.Item1 + "\n" + res2.Item2 + "\n" + res2.Item3);
string corrType = "Pearson";
//string corrType = "Kendall";
//string corrType = "Spearman";
double minMilliseconds = 10000;
FolderBrowserDialog fbd = new FolderBrowserDialog();
if (fbd.ShowDialog() == DialogResult.OK)
{
//sensor is first string
var timeTable = new Dictionary<string, Dictionary<int, List<Tuple<double, double>>>>();
var stimuliTable = new Dictionary<string, Dictionary<string, List<Tuple<double, double>>>>();
var totalList = new Dictionary<string, List<Tuple<double, double>>>();
var big5List = new Dictionary<string, List<Dictionary<Big5, int>>>();
List<string> sensors = new List<string>();
List<int> times = new List<int>();
List<string> stimulis = new List<string>();
List<string> resultFiles = new List<string>();
foreach (var folder in Directory.GetDirectories(fbd.SelectedPath))
{
if (folder.Contains("Stimuli high") ||
folder.Contains("Stimuli low") ||
folder.Contains("Time 0") ||
folder.Contains("Time 1") ||
folder.Contains("Time 2") ||
folder.Contains(".git") ||
folder.Split('\\').Last() == "3" ||
folder.Split('\\').Last() == "6" ||
folder.Split('\\').Last() == "13")
{
continue;
}
string subject = folder.Split('\\').Last();
var metaLines = File.ReadAllLines($"{folder}/meta.txt");
var big5 = GetBig5(metaLines);
int time = int.Parse(metaLines[0].Split('=').Last());
string stimuli = metaLines[1].Split('=').Last();
stimuli = stimuli == "neu" ? "low" : "high";
if (!times.Contains(time)) times.Add(time);
if (!stimulis.Contains(stimuli)) stimulis.Add(stimuli);
List<string> foldersToExamine = new List<string>();
foldersToExamine.Add(fbd.SelectedPath + "\\Time " + time);
if (time > 0)
{
foldersToExamine.Add(fbd.SelectedPath + "\\Stimuli " + stimuli);
}
if (!big5List.ContainsKey("time" + time))
{
big5List.Add("time" + time, new List<Dictionary<Big5, int>>());
}
if (!big5List.ContainsKey("stim" + stimuli))
{
big5List.Add("stim" + stimuli, new List<Dictionary<Big5, int>>());
}
if (!big5List.ContainsKey("total"))
{
big5List.Add("total", new List<Dictionary<Big5, int>>());
}
if (!big5List.ContainsKey("corr"))
{
big5List.Add("corr", new List<Dictionary<Big5, int>>());
}
if (!big5List.ContainsKey("revCorr"))
{
big5List.Add("revCorr", new List<Dictionary<Big5, int>>());
}
big5List["time" + time].Add(big5);
if (time != 0)
{
big5List["stim" + stimuli].Add(big5);
}
big5List["total"].Add(big5);
foreach (var folderToExamine in foldersToExamine)
{
foreach (var resultFile in Directory.GetFiles(folderToExamine).Where(f => f.Split('\\').Last().StartsWith(subject) && f.Split('\\').Last().EndsWith(".txt")))
{
if (resultFiles.Contains(resultFile.Split('\\').Last()) || !folderToExamine.Contains("Time") && !foldersToExamine.Contains("Stimuli") || resultFile.Contains("dtw"))
{
continue;
}
resultFiles.Add(resultFile.Split('\\').Last());
string sensor = new String(resultFile.Split('.').First().SkipWhile(x => x != '_').Skip(1).SkipWhile(x => x != '_').Skip(1).ToArray());
if (!sensors.Contains(sensor)) sensors.Add(sensor);
var resultLines = File.ReadAllLines(resultFile);
string correlationLine = resultLines.First(x => x.Contains("|" + corrType));
int corrId = resultLines.ToList().IndexOf(correlationLine);
int sigId = corrId + 2;
string significanceLine = resultLines[sigId];
string N = resultLines[sigId + 2];
double highPassThreshold = minMilliseconds / 1000;
if (sensor.Contains("EEG"))
{
highPassThreshold *= 128;
}
else if (sensor.Contains("GSR"))
{
highPassThreshold *= 20;
}
else if (sensor.Contains("HR"))
{
highPassThreshold *= 1;
}
string[] Nsplit = N.Split(new char[] { '|', ' ' }, StringSplitOptions.RemoveEmptyEntries);
if (correlationLine.Contains(".a") || significanceLine.Contains(".a") || int.Parse(Nsplit[2]) < highPassThreshold)
{
if (int.Parse(Nsplit[2]) < highPassThreshold)
{
Log.LogMessage("Removing - " + Nsplit[2] + ": " + resultFile);
}
continue;
}
int splitIndex = (corrType == "Pearson") ? 3 : 4;
double pearsCorrelation = double.Parse(correlationLine.Split(new char[] { '|', '*' }, StringSplitOptions.RemoveEmptyEntries)[splitIndex].Replace(',', '.'), System.Globalization.CultureInfo.InvariantCulture);
double pearsSignificance = double.Parse(significanceLine.Split(new char[] { '|', '*' }, StringSplitOptions.RemoveEmptyEntries)[splitIndex].Replace(',', '.'), System.Globalization.CultureInfo.InvariantCulture);
var result = Tuple.Create(pearsCorrelation, pearsSignificance);
if (!timeTable.ContainsKey(sensor))
{
timeTable.Add(sensor, new Dictionary<int, List<Tuple<double, double>>>());
stimuliTable.Add(sensor, new Dictionary<string, List<Tuple<double, double>>>());
totalList.Add(sensor, new List<Tuple<double, double>>());
}
if (!timeTable[sensor].ContainsKey(time))
{
timeTable[sensor].Add(time, new List<Tuple<double, double>>());
}
if (!stimuliTable[sensor].ContainsKey(stimuli))
{
stimuliTable[sensor].Add(stimuli, new List<Tuple<double, double>>());
}
timeTable[sensor][time].Add(result);
totalList[sensor].Add(result);
if (time != 0)
{
stimuliTable[sensor][stimuli].Add(result);
}
if (pearsCorrelation > 0)
{
big5List["corr"].Add(big5);
}
else
{
big5List["revCorr"].Add(big5);
}
}
}
}
//done gathering results
List<string> totalToWrite = new List<string>();
totalToWrite.Add("Sensor&Avg Corr&Avg Sig. \\\\");
foreach (var sensor in sensors)
{
double avgCorrelation = totalList[sensor].Average(x => x.Item1);
double stdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(totalList[sensor].Select(x => x.Item1).ToArray());
double avgSignificance = totalList[sensor].Average(x => x.Item2);
double stdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(totalList[sensor].Select(x => x.Item2).ToArray());
totalToWrite.Add($"{sensor}&{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})&{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")}) \\\\");
}
Dictionary<Big5, List<string>> big5Anova = new Dictionary<Big5, List<string>>();
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
big5Anova.Add(item, new List<string>());
totalToWrite.Add(item + " Mean: " + big5List["total"].Average(x => x[item]).ToString("0.00") + ", SD: " + MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(big5List["total"].Select(x => x[item]).ToArray()).ToString("0.00") + ".");
big5List["time0"].ForEach(x => big5Anova[item].Add("0;" + x[item]));
big5List["time1"].ForEach(x => big5Anova[item].Add("1;" + x[item]));
big5List["time2"].ForEach(x => big5Anova[item].Add("2;" + x[item]));
big5List["stimlow"].ForEach(x => big5Anova[item].Add("3;" + x[item]));
big5List["stimhigh"].ForEach(x => big5Anova[item].Add("4;" + x[item]));
}
foreach (var big5group in big5Anova)
{
File.WriteAllLines(fbd.SelectedPath + "/" + big5group.Key + "_anova.csv", big5group.Value);
}
File.WriteAllLines(fbd.SelectedPath + "/" + corrType + "_totals.txt", totalToWrite);
double width = 1 / (sensors.Count * 1.4);
double widthTime = 0.3;
var timeModel = new PlotModel() { Title = $"Time Groups Box Plot" };
var avgLineSeries = new OxyPlot.Series.LineSeries() { };
List<OxyColor> colors = new List<OxyColor>()
{
};
int small = sensors.Count / 3;
int mid = small * 2;
int stepsize = 255 / small;
for (int i = 0; i < sensors.Count; i++)
{
byte increaser = (byte)((i % small) * stepsize);
byte decreaser = (byte)(255 - increaser);
if (i < small)
{
colors.Add(OxyColor.FromRgb(increaser, decreaser, decreaser));
}
else if (i < mid)
{
colors.Add(OxyColor.FromRgb(decreaser, increaser, decreaser));
}
else
{
colors.Add(OxyColor.FromRgb(decreaser, decreaser, increaser));
}
}
List<string> sensorsAdded = new List<string>();
foreach (var sensor in sensors)
{
List<string> timeAnova = new List<string>();
foreach (var time in times)
{
timeTable[sensor][time].ForEach(x => timeAnova.Add(time + ";" + x.Item1));
}
File.WriteAllLines(fbd.SelectedPath + "/" + sensor + ".csv", timeAnova);
}
Dictionary<string, int[]> significantAmount = new Dictionary<string, int[]>();
Dictionary<string, int[]> significantAmountMax = new Dictionary<string, int[]>();
var significantCorr = new Dictionary<string, Tuple<double, double, double, double>[]>();
foreach (var sensor in sensors)
{
significantAmount.Add(sensor, new int[5]);
significantAmountMax.Add(sensor, new int[5]);
//significantCorr.Add(sensor, new Tuple<double, double>[5]);
}
significantCorr.Add("EEG", new Tuple<double, double, double, double>[5]);
significantCorr.Add("EDA", new Tuple<double, double, double, double>[5]);
significantCorr.Add("HR", new Tuple<double, double, double, double>[5]);
//significantCorr.Add("AVG", new Tuple<double, double, double, double>[5]);
Action<string, int, List<Tuple<double, double>>> AddCorrelation = (sens, id, correl) =>
{
//old average + sd
//significantCorr[sens][id] = Tuple.Create(correl.Average(x => x.Item1), correl.Select(x => x.Item1).STDEV(), correl.Average(x => x.Item2), correl.Select(x => x.Item2).STDEV());
//new fisher algorithms
significantCorr[sens][id] = Tuple.Create(FisherInverse(correl.Average(x => Fisher(x.Item1))), Math.Round((double)correl.Count), correl.Select(x => x.Item2).FisherCombineP(), Math.Round((double)correl.Count));
};
List<string> amountTimeSignificant = new List<string>();
List<double> timeSignificantPoints = new List<double>();
var TimeErrorModel = new PlotModel() { Title = $"Time Error Model" };
var timeErrorSeries = new OxyPlot.Series.ErrorColumnSeries() { };
TimeErrorModel.Series.Add(timeErrorSeries);
TimeErrorModel.Axes.Add(new OxyPlot.Axes.LinearAxis { Position = OxyPlot.Axes.AxisPosition.Left });
var axis = new OxyPlot.Axes.CategoryAxis { Position = OxyPlot.Axes.AxisPosition.Bottom };
axis.Labels.Add("Time 0");
axis.Labels.Add("Time 1");
axis.Labels.Add("Time 2");
TimeErrorModel.Axes.Add(axis);
TimeErrorModel.Annotations.Add(new OxyPlot.Annotations.LineAnnotation() { Y = 0, Type = OxyPlot.Annotations.LineAnnotationType.Horizontal });
var TimeErrorModel2 = new PlotModel() { Title = $"Time Error Model" };
TimeErrorModel2.Axes.Add(new OxyPlot.Axes.LinearAxis { Position = OxyPlot.Axes.AxisPosition.Left });
var axis2 = new OxyPlot.Axes.CategoryAxis { Position = OxyPlot.Axes.AxisPosition.Bottom };
axis2.Labels.Add("Time 0");
axis2.Labels.Add("Time 1");
axis2.Labels.Add("Time 2");
TimeErrorModel2.Axes.Add(axis2);
List<string> AnovaIndividual = new List<string>();
List<string> AnovaAvg = new List<string>();
int anovaIndividualId = 0;
int anovaAvgId = 0;
List<string> AnovaIndividualLegend = new List<string>();
List<string> AnovaAvgLegend = new List<string>();
foreach (var time in times)
{
int sensorId = 0;
List<string> timeToWrite = new List<string>();
timeToWrite.Add("Sensor&Avg Corr&Avg Sig. \\\\");
List<double> avgs = new List<double>();
List<double> sigPoints = new List<double>();
var errorSeries = new OxyPlot.Series.ErrorColumnSeries();
TimeErrorModel2.Series.Add(errorSeries);
var EEGAllCorrelations = new List<Tuple<double, double>>();
var GSRAllCorrelations = timeTable["GSR"][time];
var HRAllCorrelations = timeTable["HR"][time];
foreach (var sensor in sensors)
{
if (sensor.Contains("EEG"))
{
EEGAllCorrelations.AddRange(timeTable[sensor][time]);
}
double avgCorrelation = timeTable[sensor][time].Average(x => x.Item1);
double stdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Select(x => x.Item1).ToArray());
double avgSignificance = timeTable[sensor][time].Average(x => x.Item2);
double stdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Select(x => x.Item2).ToArray());
var orderedAll = timeTable[sensor][time].OrderBy(x => x.Item1).ToList();//timeTable[sensor][time].Where(x => x.Item2 * 100 < (int)5).OrderBy(x => x.Item1).ToList();
amountTimeSignificant.Add(time + " & " + sensor + " & " + orderedAll.Count);
significantAmount[sensor][time] = orderedAll.Count;
significantAmountMax[sensor][time] = timeTable[sensor][time].Count;
//significantCorr[sensor][time] = Tuple.Create(orderedAll.Average(x => x.Item1), MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(orderedAll.Select(x => x.Item1).ToArray()));
//var boxItem = new OxyPlot.Series.BoxPlotItem(time + sensorId * widthTime - (0.5 * widthTime * sensors.Count), orderedAll[0].Item1, orderedAll[(int)(orderedAll.Count * 0.25)].Item1, orderedAll[orderedAll.Count / 2].Item1, orderedAll[(int)(orderedAll.Count * 0.75)].Item1, orderedAll.Last().Item1);
//var boxItem = new OxyPlot.Series.BoxPlotItem(sensorId + time * widthTime - (0.5 * widthTime * (times.Count - 1)), orderedAll[0].Item1, orderedAll[(int)(orderedAll.Count * 0.25)].Item1, orderedAll[orderedAll.Count / 2].Item1, orderedAll[(int)(orderedAll.Count * 0.75)].Item1, orderedAll.Last().Item1);
//var boxSeries = new OxyPlot.Series.BoxPlotSeries() { };
//boxSeries.BoxWidth = widthTime;
//boxSeries.WhiskerWidth = widthTime;
//boxSeries.Items.Add(boxItem);
//boxSeries.Fill = colors[sensorId];
//timeModel.Series.Add(boxSeries);
errorSeries.Items.Add(new OxyPlot.Series.ErrorColumnItem(orderedAll.Average(x => x.Item1), MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(orderedAll.Select(x => x.Item1).ToArray()), time) { Color = colors[sensorId] });
//if (!sensorsAdded.Contains(sensor))
//{
// sensorsAdded.Add(sensor);
// boxSeries.Title = sensor;
//}
avgs.Add(orderedAll.Average(x => x.Item1));
sigPoints.AddRange(orderedAll.Select(x => x.Item1));
if (avgSignificance * 100 < (int)5)
{
timeToWrite.Add($"\\textbf{{{sensor}}}&\\textbf{{{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})}}&\\textbf{{{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")})}} \\\\");
}
else
{
timeToWrite.Add($"{sensor}&{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})&{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")}) \\\\");
}
sensorId++;
}
double boxWidth = 0.3;
//eeg
EEGAllCorrelations = EEGAllCorrelations.OrderBy(x => x.Item1).ToList();
var EEGSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = EEGColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (time == 0) EEGSeries.Title = "EEG";
var EEGItem = CreateBoxItem(EEGAllCorrelations);
EEGItem.X = time - EEGSeries.BoxWidth * 1;
EEGSeries.Items.Add(EEGItem);
timeModel.Series.Add(EEGSeries);
AddCorrelation("EEG", time, EEGAllCorrelations);
foreach (var cor in EEGAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=time_" + time + "_EEG");
//gsr
GSRAllCorrelations = GSRAllCorrelations.OrderBy(x => x.Item1).ToList();
var GSRSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = EDAColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (time == 0) GSRSeries.Title = "EDA";
var GSRItem = CreateBoxItem(GSRAllCorrelations);
GSRItem.X = time;
GSRSeries.Items.Add(GSRItem);
timeModel.Series.Add(GSRSeries);
AddCorrelation("EDA", time, GSRAllCorrelations);
foreach (var cor in GSRAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=time_" + time + "_GSR");
//hr
HRAllCorrelations = HRAllCorrelations.OrderBy(x => x.Item1).ToList();
var HRSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = HRColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (time == 0) HRSeries.Title = "HR";
var HRItem = CreateBoxItem(HRAllCorrelations);
HRItem.X = time + HRSeries.BoxWidth * 1;
HRSeries.Items.Add(HRItem);
timeModel.Series.Add(HRSeries);
AddCorrelation("HR", time, HRAllCorrelations);
foreach (var cor in HRAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=time_" + time + "_HR");
//avg
var AVGAllCorrelations = EEGAllCorrelations.Concat(GSRAllCorrelations.Concat(HRAllCorrelations)).ToList();
//AddCorrelation("AVG", time, AVGAllCorrelations);
foreach (var cor in AVGAllCorrelations)
{
AnovaAvg.Add(anovaAvgId + ";" + cor.Item1);
}
AnovaAvgLegend.Add(anovaAvgId++ + "=time_" + time);
double totalAvg = AVGAllCorrelations.Average(x => x.Item1);
var txtAvg = new OxyPlot.Annotations.TextAnnotation() { TextPosition = new OxyPlot.DataPoint(time, -1), Text = "Avg " + totalAvg.ToString(".000").Replace(",", "."), Stroke = OxyColors.White };
timeModel.Annotations.Add(txtAvg);
timeErrorSeries.Items.Add(new OxyPlot.Series.ErrorColumnItem(sigPoints.Average(), MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(sigPoints.ToArray()), time));
avgLineSeries.Points.Add(new OxyPlot.DataPoint(time, avgs.Average()));
File.WriteAllLines(fbd.SelectedPath + "/" + corrType + "_time" + time + ".txt", timeToWrite);
}
File.WriteAllLines(fbd.SelectedPath + "/significantTime.tex", amountTimeSignificant);
timeModel.LegendPlacement = LegendPlacement.Outside;
timeModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Left, Maximum = 1, Minimum = -1, Title = "Pearson's r" });
timeModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = 2.5, Minimum = -0.5, MajorStep = 1, Title = "Time", MinorTickSize = 0 });
//timeModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = sensors.Count - 0.5, Minimum = -0.5, MajorStep = 1, Title = "Sensors", MinorTickSize = 0 });
//boxModel.Series.Add(avgLineSeries);
PngExporter pnger = new PngExporter();
pnger.ExportToFile(timeModel, fbd.SelectedPath + "/timeBox.png");
pnger.ExportToFile(TimeErrorModel, fbd.SelectedPath + "/errorPlotTest.png");
pnger.ExportToFile(TimeErrorModel2, fbd.SelectedPath + "/errorPlotTest2.png");
/*
//correlation and reverse correlation
foreach (var time in times)
{
//Correlation
List<string> correlationTimeToWrite = new List<string>();
correlationTimeToWrite.Add("Sensor&Avg Corr&Avg Sig. \\\\");
//Reverse correlation
List<string> reverseCorrelationTimeToWrite = new List<string>();
reverseCorrelationTimeToWrite.Add("Sensor & Avg Corr & Avg Sig. \\\\");
foreach (var sensor in sensors)
{
double correlationAvgCorrelation = timeTable[sensor][time].Where(x => x.Item1 >= 0).Average(x => x.Item1);
double correlationStdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Where(x => x.Item1 >= 0).Select(x => x.Item1).ToArray());
double correlationAvgSignificance = timeTable[sensor][time].Where(x => x.Item1 >= 0).Average(x => x.Item2);
double correlationStdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Where(x => x.Item1 >= 0).Select(x => x.Item2).ToArray());
double reverseCorrelationAvgCorrelation = timeTable[sensor][time].Where(x => x.Item1 < 0).Average(x => x.Item1);
double reverseCorrelationStdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Where(x => x.Item1 < 0).Select(x => x.Item1).ToArray());
double reverseCorrelationAvgSignificance = timeTable[sensor][time].Where(x => x.Item1 < 0).Average(x => x.Item2);
double reverseCorrelationStdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Where(x => x.Item1 < 0).Select(x => x.Item2).ToArray());
correlationTimeToWrite.Add($"{sensor}&{correlationAvgCorrelation.ToString("0.000")}({correlationStdevCorrelation.ToString("0.000")})&{correlationAvgSignificance.ToString("0.000")}({correlationStdevSignificance.ToString("0.000")}) \\\\");
reverseCorrelationTimeToWrite.Add($"{sensor}&{reverseCorrelationAvgCorrelation.ToString("0.000")}({reverseCorrelationStdevCorrelation.ToString("0.000")})&{reverseCorrelationAvgSignificance.ToString("0.000")}({reverseCorrelationStdevSignificance.ToString("0.000")}) \\\\");
}
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
correlationTimeToWrite.Add(item + " Mean: " + big5List["corr"].Average(x => x[item]).ToString("0.00") + ", SD: " + MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(big5List["corr"].Select(x => x[item]).ToArray()).ToString("0.00") + ".");
reverseCorrelationTimeToWrite.Add(item + " Mean: " + big5List["revCorr"].Average(x => x[item]).ToString("0.00") + ", SD: " + MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(big5List["revCorr"].Select(x => x[item]).ToArray()).ToString("0.00") + ".");
}
File.WriteAllLines(fbd.SelectedPath + "/correlationTime" + time + ".txt", correlationTimeToWrite);
File.WriteAllLines(fbd.SelectedPath + "/reverseCorrelationTime" + time + ".txt", reverseCorrelationTimeToWrite);
}
*/
var Big5timeBox = new PlotModel() { Title = "Big5 Time Box Plots", LegendPlacement = LegendPlacement.Outside };
Dictionary<Big5, OxyPlot.Series.BoxPlotSeries> big5timeSeries = new Dictionary<Big5, OxyPlot.Series.BoxPlotSeries>();
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
big5timeSeries.Add(item, new OxyPlot.Series.BoxPlotSeries() { Fill = colors[(int)item * 2], Title = item.ToString(), BoxWidth = 0.1, WhiskerWidth = 0.1 });
Big5timeBox.Series.Add(big5timeSeries[item]);
}
Big5timeBox.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Left, Maximum = 50, Minimum = 10, Title = "Score" });
Big5timeBox.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = 2.5, Minimum = -0.5, MajorStep = 1, Title = "Time", MinorTickSize = 0 });
foreach (var time in times)
{
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
var orderino = big5List["time" + time].Select(x => x[item]).OrderBy(x => x).ToList();
big5timeSeries[item].Items.Add(new OxyPlot.Series.BoxPlotItem(time - 0.25 + (int)item * 0.1, orderino[0], orderino[(int)(orderino.Count * 0.25)], orderino[orderino.Count / 2], orderino[(int)(orderino.Count * 0.75)], orderino.Last()));
}
}
pnger.ExportToFile(Big5timeBox, fbd.SelectedPath + "/timeBoxBig5.png");
foreach (var time in times)
{
List<string> timeToWrite = new List<string>();
timeToWrite.Add("\\begin{table}");
timeToWrite.Add("\\centering");
timeToWrite.Add("{\\large \\textbf{Time " + time + "}}\\vspace{1pt}");
timeToWrite.Add("\\begin{tabular}{ccc}");
timeToWrite.Add("\\toprule");
timeToWrite.Add("Sensor&Avg Corr&Avg Sig. \\\\");
timeToWrite.Add("\\midrule");
foreach (var sensor in sensors)
{
double avgCorrelation = timeTable[sensor][time].Average(x => x.Item1);
double stdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Select(x => x.Item1).ToArray());
double avgSignificance = timeTable[sensor][time].Average(x => x.Item2);
double stdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(timeTable[sensor][time].Select(x => x.Item2).ToArray());
if (avgSignificance < 0.05)
{
timeToWrite.Add($"\\textbf{{{sensor}}}&\\textbf{{{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})}}&\\textbf{{{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")})}} \\\\");
}
else
{
timeToWrite.Add($"{sensor}&{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})&{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")}) \\\\");
}
}
timeToWrite.Add("\\bottomrule");
timeToWrite.Add("\\end{tabular}");
timeToWrite.Add("\\caption{Results from time " + time + ".");
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
timeToWrite.Add(item + " Mean: " + big5List["time" + time].Average(x => x[item]).ToString("0.00") + ", SD: " + MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(big5List["time" + time].Select(x => x[item]).ToArray()).ToString("0.00") + ".");
}
timeToWrite.Add("}");
timeToWrite.Add("\\label{[TABLE] res time" + time + "}");
timeToWrite.Add("\\end{table}");
File.WriteAllLines(fbd.SelectedPath + "/" + corrType + "_time" + time + ".txt", timeToWrite);
}
var stimModel = new PlotModel() { Title = "Stimuli Groups Box Plot" };
int stimId = 0;
sensorsAdded.Clear();
avgLineSeries.Points.Clear();
var Big5StimBox = new PlotModel() { Title = "Big5 Stimuli Box Plots", LegendPlacement = LegendPlacement.Outside };
Dictionary<Big5, OxyPlot.Series.BoxPlotSeries> big5Series = new Dictionary<Big5, OxyPlot.Series.BoxPlotSeries>();
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
big5Series.Add(item, new OxyPlot.Series.BoxPlotSeries() { Fill = colors[(int)item * 2], Title = item.ToString(), BoxWidth = 0.1, WhiskerWidth = 0.1 });
Big5StimBox.Series.Add(big5Series[item]);
}
Big5StimBox.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Left, Maximum = 50, Minimum = 10, Title = "Score" });
Big5StimBox.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = 1.5, Minimum = -0.5, MajorStep = 1, Title = "Category", MinorTickSize = 0 });
List<string> amountStimSignificant = new List<string>();
foreach (var stimuli in stimulis)
{
List<string> stimuliToWrite = new List<string>();
stimuliToWrite.Add("\\begin{table}");
stimuliToWrite.Add("\\centering");
stimuliToWrite.Add("{\\large \\textbf{Stimuli " + stimuli + "}}\\vspace{1pt}");
stimuliToWrite.Add("\\begin{tabular}{ccc}");
stimuliToWrite.Add("\\toprule");
stimuliToWrite.Add("Sensor&Avg Corr&Avg Sig. \\\\");
stimuliToWrite.Add("\\midrule");
List<double> avgs = new List<double>();
int sensorId = 0;
var EEGAllCorrelations = new List<Tuple<double, double>>();
var GSRAllCorrelations = stimuliTable["GSR"][stimuli];
var HRAllCorrelations = stimuliTable["HR"][stimuli];
foreach (var sensor in sensors)
{
if (sensor.Contains("EEG"))
{
EEGAllCorrelations.AddRange(stimuliTable[sensor][stimuli]);
}
double avgCorrelation = stimuliTable[sensor][stimuli].Average(x => x.Item1);
double stdevCorrelation = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(stimuliTable[sensor][stimuli].Select(x => x.Item1).ToArray());
double avgSignificance = stimuliTable[sensor][stimuli].Average(x => x.Item2);
double stdevSignificance = MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(stimuliTable[sensor][stimuli].Select(x => x.Item2).ToArray());
var orderedAll = stimuliTable[sensor][stimuli].Where(x => x.Item2 * 100 < (int)5).OrderBy(x => x.Item1).ToList();
amountStimSignificant.Add(stimuli + " & " + sensor + " & " + orderedAll.Count);
significantAmount[sensor][stimuli == "low" ? 3 : 4] = orderedAll.Count;
significantAmountMax[sensor][stimuli == "low" ? 3 : 4] = stimuliTable[sensor][stimuli].Count;
//significantCorr[sensor][stimuli == "low" ? 3 : 4] = Tuple.Create(orderedAll.Average(x => x.Item1), MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(orderedAll.Select(x => x.Item1).ToArray()));
var boxItem = new OxyPlot.Series.BoxPlotItem(((1 + stimId) % 2) + sensorId * width - (0.5 * width * sensors.Count), orderedAll[0].Item1, orderedAll[(int)(orderedAll.Count * 0.25)].Item1, orderedAll[orderedAll.Count / 2].Item1, orderedAll[(int)(orderedAll.Count * 0.75)].Item1, orderedAll.Last().Item1);
//var boxItem = new OxyPlot.Series.BoxPlotItem(sensorId + ((1 + stimId) % 2) * widthTime - (0.5 * widthTime), orderedAll[0].Item1, orderedAll[(int)(orderedAll.Count * 0.25)].Item1, orderedAll[orderedAll.Count / 2].Item1, orderedAll[(int)(orderedAll.Count * 0.75)].Item1, orderedAll.Last().Item1);
var boxSeries = new OxyPlot.Series.BoxPlotSeries() { };
boxSeries.BoxWidth = width;
boxSeries.WhiskerWidth = width;
//boxSeries.BoxWidth = widthTime;
//boxSeries.WhiskerWidth = widthTime;
boxSeries.Items.Add(boxItem);
boxSeries.Fill = colors[sensorId];
//stimModel.Series.Add(boxSeries);
avgs.Add(orderedAll.Average(x => x.Item1));
if (!sensorsAdded.Contains(sensor))
{
sensorsAdded.Add(sensor);
boxSeries.Title = sensor;
}
if (avgSignificance < 0.05)
{
stimuliToWrite.Add($"\\textbf{{{sensor}}}&\\textbf{{{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})}}&\\textbf{{{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")})}} \\\\");
}
else
{
stimuliToWrite.Add($"{sensor}&{avgCorrelation.ToString("0.000")}({stdevCorrelation.ToString("0.000")})&{avgSignificance.ToString("0.000")}({stdevSignificance.ToString("0.000")}) \\\\");
}
sensorId++;
}
double boxWidth = 0.3;
//eeg
EEGAllCorrelations = EEGAllCorrelations.OrderBy(x => x.Item1).ToList();
var EEGSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = EEGColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (stimuli == "low") EEGSeries.Title = "EEG";
var EEGItem = CreateBoxItem(EEGAllCorrelations);
EEGItem.X = (stimuli == "low" ? 0 : 1) - EEGSeries.BoxWidth * 1;
EEGSeries.Items.Add(EEGItem);
stimModel.Series.Add(EEGSeries);
AddCorrelation("EEG", stimuli == "low" ? 3 : 4, EEGAllCorrelations);
foreach (var cor in EEGAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=stimuli_" + stimuli + "_EEG");
//gsr
GSRAllCorrelations = GSRAllCorrelations.OrderBy(x => x.Item1).ToList();
var GSRSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = EDAColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (stimuli == "low") GSRSeries.Title = "EDA";
var GSRItem = CreateBoxItem(GSRAllCorrelations);
GSRItem.X = (stimuli == "low" ? 0 : 1);
GSRSeries.Items.Add(GSRItem);
stimModel.Series.Add(GSRSeries);
AddCorrelation("EDA", stimuli == "low" ? 3 : 4, GSRAllCorrelations);
foreach (var cor in GSRAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=stimuli_" + stimuli + "_GSR");
//hr
HRAllCorrelations = HRAllCorrelations.OrderBy(x => x.Item1).ToList();
var HRSeries = new OxyPlot.Series.BoxPlotSeries() { Fill = HRColor, BoxWidth = boxWidth, WhiskerWidth = boxWidth };
if (stimuli == "low") HRSeries.Title = "HR";
var HRItem = CreateBoxItem(HRAllCorrelations);
HRItem.X = (stimuli == "low" ? 0 : 1) + HRSeries.BoxWidth * 1;
HRSeries.Items.Add(HRItem);
stimModel.Series.Add(HRSeries);
AddCorrelation("HR", stimuli == "low" ? 3 : 4, HRAllCorrelations);
foreach (var cor in HRAllCorrelations)
{
AnovaIndividual.Add(anovaIndividualId + ";" + cor.Item1);
}
AnovaIndividualLegend.Add(anovaIndividualId++ + "=stimuli_" + stimuli + "_HR");
//avg
var AVGAllCorrelations = EEGAllCorrelations.Concat(GSRAllCorrelations.Concat(HRAllCorrelations)).ToList();
//AddCorrelation("AVG", stimuli == "low" ? 3 : 4, AVGAllCorrelations);
foreach (var cor in AVGAllCorrelations)
{
AnovaAvg.Add(anovaAvgId + ";" + cor.Item1);
}
AnovaAvgLegend.Add(anovaAvgId++ + "=stimuli_" + stimuli);
avgLineSeries.Points.Add(new OxyPlot.DataPoint(0, avgs.Average()));
stimuliToWrite.Add("\\bottomrule");
stimuliToWrite.Add("\\end{tabular}");
stimuliToWrite.Add("\\caption{Results from stimuli " + stimuli + ".");
foreach (Big5 item in Enum.GetValues(typeof(Big5)))
{
stimuliToWrite.Add(item + " Mean: " + big5List["stim" + stimuli].Average(x => x[item]).ToString("0.00") + ", SD: " + MathNet.Numerics.Statistics.ArrayStatistics.PopulationStandardDeviation(big5List["stim" + stimuli].Select(x => x[item]).ToArray()).ToString("0.00") + ".");
var orderino = big5List["stim" + stimuli].Select(x => x[item]).OrderBy(x => x).ToList();
big5Series[item].Items.Add(new OxyPlot.Series.BoxPlotItem(((1 + stimId) % 2) - 0.25 + (int)item * 0.1, orderino[0], orderino[(int)(orderino.Count * 0.25)], orderino[orderino.Count / 2], orderino[(int)(orderino.Count * 0.75)], orderino.Last()));
}
stimuliToWrite.Add("}");
stimuliToWrite.Add("\\label{[TABLE] res stimuli" + stimuli + "}");
stimuliToWrite.Add("\\end{table}");
File.WriteAllLines(fbd.SelectedPath + "/" + corrType + "_stimuli_" + stimuli + ".txt", stimuliToWrite);
stimId++;
}
File.WriteAllLines(fbd.SelectedPath + "/significantStim.tex", amountStimSignificant);
List<string> sigAmountLines = new List<string>();
foreach (var sensor in sensors)
{
string linerino = sensor;
for (int i = 0; i < 5; i++)
{
linerino += $" & {significantAmount[sensor][i]}/{significantAmountMax[sensor][i]}";
}
sigAmountLines.Add(linerino + "\\\\");
}
File.WriteAllLines(fbd.SelectedPath + "/significantTable.tex", sigAmountLines);
//File.WriteAllLines(fbd.SelectedPath + "/significantTable.tex", significantAmount.Select(x => $"{x.Key} & {x.Value[0]} & {x.Value[1]} & {x.Value[2]} & {x.Value[3]} & {x.Value[4]}").ToList());
File.WriteAllLines(fbd.SelectedPath + "/significantCorrTable.tex", significantCorr.Select(x => $"{x.Key} & {x.Value[0].Item1.ToString(".000")}({x.Value[0].Item2.ToString(".000")}) & {x.Value[1].Item1.ToString(".000")}({x.Value[1].Item2.ToString(".000")}) & {x.Value[2].Item1.ToString(".000")}({x.Value[2].Item2.ToString(".000")}) & {x.Value[3].Item1.ToString(".000")}({x.Value[3].Item2.ToString(".000")}) & {x.Value[4].Item1.ToString(".000")}({x.Value[4].Item2.ToString(".000")}) \\\\"));
File.WriteAllLines(fbd.SelectedPath + "/significantCorrTableTime.tex", significantCorr.Select(x => $"{x.Key} & {x.Value[0].Item1.ToString(".000")} (SD={x.Value[0].Item2.ToString(".000")}, p={x.Value[0].Item3.ToString(".000000")}) & {x.Value[1].Item1.ToString(".000")} (SD={x.Value[1].Item2.ToString(".000")}, p={x.Value[1].Item3.ToString(".000000")}) & {x.Value[2].Item1.ToString(".000")} (SD={x.Value[2].Item2.ToString(".000")}, p={x.Value[2].Item3.ToString(".000000")}) \\\\"));
File.WriteAllLines(fbd.SelectedPath + "/significantCorrTableStimuli.tex", significantCorr.Select(x => $"{x.Key} & {x.Value[3].Item1.ToString(".000")} (SD={x.Value[3].Item2.ToString(".000")}, p={x.Value[3].Item3.ToString(".000")}) & {x.Value[4].Item1.ToString(".000")} (SD={x.Value[4].Item2.ToString(".000")}, p={x.Value[4].Item3.ToString(".000")}) \\\\"));
List<string> timeLines = new List<string>() { "sensor & 0 vs 1 & 1 vs 2 & 0 vs 2" };
List<string> stimLines = new List<string>() { "sensor & 0 vs Low & Low vs High & 0 vs High" };
foreach (var item in significantCorr)
{
var z01 = ZCalc(item.Value[0].Item1, Convert.ToInt32(item.Value[0].Item2), item.Value[1].Item1, Convert.ToInt32(item.Value[1].Item2));
var z12 = ZCalc(item.Value[1].Item1, Convert.ToInt32(item.Value[1].Item2), item.Value[2].Item1, Convert.ToInt32(item.Value[2].Item2));
var z02 = ZCalc(item.Value[0].Item1, Convert.ToInt32(item.Value[0].Item2), item.Value[2].Item1, Convert.ToInt32(item.Value[2].Item2));
var p01 = ZtoP(z01);
var p12 = ZtoP(z12);
var p02 = ZtoP(z02);
timeLines.Add($"{item.Key} & z: {z01} | p: {p01} & z: {z12} | p: {p12} & z: {z02} | p: {p02}");
var z0Low = ZCalc(item.Value[0].Item1, Convert.ToInt32(item.Value[0].Item2), item.Value[3].Item1, Convert.ToInt32(item.Value[3].Item2));
var zLowHigh = ZCalc(item.Value[3].Item1, Convert.ToInt32(item.Value[3].Item2), item.Value[4].Item1, Convert.ToInt32(item.Value[4].Item2));
var z0High = ZCalc(item.Value[0].Item1, Convert.ToInt32(item.Value[0].Item2), item.Value[4].Item1, Convert.ToInt32(item.Value[4].Item2));
var p0Low = ZtoP(z0Low);
var pLowHigh = ZtoP(zLowHigh);
var p0High = ZtoP(z0High);
stimLines.Add($"{item.Key} & z: {z0Low} | p: {p0Low} & z: {zLowHigh} | p: {pLowHigh} & z: {z0High} | p: {p0High}");
}
File.WriteAllLines(fbd.SelectedPath + "/significantCorrCompareTime.tex", timeLines);
File.WriteAllLines(fbd.SelectedPath + "/significantCorrCompareStimuli.tex", stimLines);
pnger.ExportToFile(Big5StimBox, fbd.SelectedPath + "/stimBoxBig5.png");
stimModel.LegendPlacement = LegendPlacement.Outside;
//index 1 = low
var stimTxt0 = new OxyPlot.Annotations.TextAnnotation() { TextPosition = new OxyPlot.DataPoint(0, -1), Text = "Avg " + avgLineSeries.Points[1].Y.ToString(".000").Replace(",", "."), Stroke = OxyColors.White };
//index 0 = high
var stimTxt1 = new OxyPlot.Annotations.TextAnnotation() { TextPosition = new OxyPlot.DataPoint(1, -1), Text = "Avg " + avgLineSeries.Points[0].Y.ToString(".000").Replace(",", "."), Stroke = OxyColors.White };
stimModel.Annotations.Add(stimTxt0);
stimModel.Annotations.Add(stimTxt1);
stimModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Left, Maximum = 1, Minimum = -1, Title = "Pearson's r" });
stimModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = 1.5, Minimum = -0.5, MajorStep = 1, Title = "Stimuli", MinorTickSize = 0 });
//stimModel.Axes.Add(new OxyPlot.Axes.LinearAxis() { Position = OxyPlot.Axes.AxisPosition.Bottom, Maximum = sensors.Count - 0.5, Minimum = -0.5, MajorStep = 1, Title = "Sensors", MinorTickSize = 0 });
pnger.ExportToFile(stimModel, fbd.SelectedPath + "/stimBox.png");
File.WriteAllLines(fbd.SelectedPath + "/anovaIndividual.csv", AnovaIndividual);
File.WriteAllLines(fbd.SelectedPath + "/anovaIndividualLegend.csv", AnovaIndividualLegend);
File.WriteAllLines(fbd.SelectedPath + "/anovaAvg.csv", AnovaAvg);
File.WriteAllLines(fbd.SelectedPath + "/anovaAvgLegend.csv", AnovaAvgLegend);
Log.LogMessage("DonnoDK");
}
}
static void SavePngScatter(string path, string name, List<double> A, List<double> B, double axisMin = 0, double axisMax = 0)
{
PngExporter pngify = new PngExporter();
pngify.Width = 2000;
pngify.Height = 2000;
var model = new PlotModel() { Title = name };
var scatterSeries = new OxyPlot.Series.ScatterSeries()
{
MarkerSize = 0.8f,
MarkerType = MarkerType.Circle,
MarkerFill = OxyColors.Black
};
for (int i = 0; i < A.Count && i < B.Count; i++)
{
scatterSeries.Points.Add(new OxyPlot.Series.ScatterPoint(A[i], B[i]));
}
model.Series.Add(scatterSeries);
model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = axisMin, Maximum = axisMax, Position = OxyPlot.Axes.AxisPosition.Left });
model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = axisMin, Maximum = axisMax, Position = OxyPlot.Axes.AxisPosition.Bottom });
pngify.ExportToFile(model, path);
}
static void SavePng(string path, string name, List<double> A, List<double> B, List<Tuple<int, int>> pairings = null)
{
PngExporter pngify = new PngExporter();
pngify.Width = 3200;
pngify.Height = 1200;
var model = new PlotModel() { Title = name };
var aSeries = new OxyPlot.Series.LineSeries() { Color = OxyColors.Blue };
var bSeries = new OxyPlot.Series.LineSeries() { Color = OxyColors.Red };
for (int i = 0; i < A.Count; i++)
{
aSeries.Points.Add(new OxyPlot.DataPoint(i, A[i]));
}
for (int i = 0; i < B.Count; i++)
{
bSeries.Points.Add(new OxyPlot.DataPoint(i, B[i]));
}
if (pairings != null)
{
for (int i = 0; i < pairings.Count; i += 10)
{
var lineSeries = new OxyPlot.Series.LineSeries() { Color = OxyColors.Gray, StrokeThickness = 0.2 };
lineSeries.Points.Add(aSeries.Points[pairings[i].Item1]);
lineSeries.Points.Add(bSeries.Points[pairings[i].Item2]);
model.Series.Add(lineSeries);
}
}
model.Series.Add(aSeries);
model.Series.Add(bSeries);
model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = 0, Maximum = 1, Position = OxyPlot.Axes.AxisPosition.Left });
//model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = 0, Maximum = 1, Position = OxyPlot.Axes.AxisPosition.Bottom });
pngify.ExportToFile(model, path);
}
private void btn_ExportPNG_Click(object sender, EventArgs e)
{
if (loaded.Count < 2)
{
Log.LogMessage("You must load all data before exporting!");
return;
}
int height = 0;
if (!int.TryParse(txt_height.Text, out height))
{
Log.LogMessage("Height must be an integer");
return;
}
int width = 0;
if (!int.TryParse(txt_width.Text, out width))
{
Log.LogMessage("Width must be an integer");
return;
}
int offset = 0;
if (!int.TryParse(txt_PlotDataOffset.Text, out offset))
{
Log.LogMessage("Offset must be an integer");
return;
}
double pointSize = 0;
if (!double.TryParse(txt_PlotPointSize.Text.Replace(".", ","), out pointSize))
{
Log.LogMessage("Point size must be a double");
return;
}
int from = -1;
if (!int.TryParse(txt_ExportFrom.Text, out from) || txt_ExportFrom.Text == "")
{
Log.LogMessage("Export from not valid integer, using 0");
from = -1;
}
int to = -1;
if (!int.TryParse(txt_ExportTo.Text, out to) || txt_ExportTo.Text == "")
{
Log.LogMessage("Export to not valid integer, using max");
to = -1;
}
SaveFileDialog sfd = new SaveFileDialog();
sfd.DefaultExt = ".png";
sfd.Filter = "png|*.png";
if (sfd.ShowDialog() == DialogResult.OK)
{
PngExporter pngify = new PngExporter();
pngify.Width = width;
pngify.Height = height;
var xy = GetXY(from, to);
var test = xy.Item1;
var recall = xy.Item2;
FitPlot(test, recall);
var model = new PlotModel() { Title = $"Slope:{slope.ToString("0.000")} RSquared:{rsquared.ToString("0.000")}" };
var dataSeries = new OxyPlot.Series.ScatterSeries() { MarkerType = MarkerType.Circle, MarkerStroke = OxyColors.Red };
for (int i = 0; i < test.Count; i++)
{
dataSeries.Points.Add(new OxyPlot.Series.ScatterPoint(test[i], recall[i]) { Size = pointSize });
}
var fitSeries = new OxyPlot.Series.FunctionSeries((x) => intercept + slope * x, test.Min(), test.Max(), 0.001) { Color = OxyColors.Blue };
model.Series.Add(dataSeries);
model.Series.Add(fitSeries);
model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = 0, Maximum = 1, Position = OxyPlot.Axes.AxisPosition.Left });
model.Axes.Add(new OxyPlot.Axes.LinearAxis() { Minimum = 0, Maximum = 1, Position = OxyPlot.Axes.AxisPosition.Bottom });
pngify.ExportToFile(model, sfd.FileName);
Log.LogMessage("Saved " + sfd.FileName + "!");
}
}