bool LoadData(string path, FusionData fd)
{
currentPath = path;
Log.LogMessage("Selected folder: " + path);
//load fusion data
samData = SAMData.LoadFromPath(path + @"\SAM.json");
string temp = samData.ShouldSkip();
if (!(temp == ""))
{
Log.LogMessage(temp);
return false;
}
shouldRun = fd.LoadFromFile(new string[] { path + @"\EEG.dat", path + @"\GSR.dat", path + @"\HR.dat", path + @"\KINECT.dat" }, samData.startTime);
//Slicing
List<SAMDataPoint> throwaway = new List<SAMDataPoint>();
foreach (SAMDataPoint samD in samData.dataPoints)
{
if (FeatureCreator.EEGDataSlice(fd.eegData.ToList<DataReading>(), samD).Count == 0 ||
FeatureCreator.GSRDataSlice(fd.gsrData.ToList<DataReading>(), samD).Count == 0 ||
FeatureCreator.HRDataSlice(fd.hrData.ToList<DataReading>(), samD).Count == 0 ||
FeatureCreator.FaceDataSlice(fd.faceData.ToList<DataReading>(), samD).Count == 0)
{
throwaway.Add(samD);
}
}
if (throwaway.Count > 5)
{
return false;
}
for (int i = 0; i < throwaway.Count; i++)
{
Log.LogMessage("Threw away a sam data point");
samData.dataPoints.Remove(throwaway[i]);
}
if (throwaway.Count > 5 && samData.ShouldSkip() == "")
{
Log.LogMessage("Too many data points thrown away (" + throwaway.Count + ")");
return false;
}
Log.LogMessage("Fusion Data loaded!");
Log.LogMessage("Applying data to features..");
FeatureCreator.GSRArousalOptimizationFeatures.ForEach(x => x.SetData(fd.gsrData.ToList<DataReading>()));
FeatureCreator.HRArousalOptimizationFeatures.ForEach(x => x.SetData(fd.hrData.ToList<DataReading>()));
FeatureCreator.HRValenceOptimizationFeatures.ForEach(x => x.SetData(fd.hrData.ToList<DataReading>()));
FeatureCreator.EEGArousalOptimizationFeatures.ForEach(x => x.SetData(fd.eegData.ToList<DataReading>()));
FeatureCreator.EEGValenceOptimizationFeatures.ForEach(x => x.SetData(fd.eegData.ToList<DataReading>()));
FeatureCreator.FACEArousalOptimizationFeatures.ForEach(x => x.SetData(fd.faceData.ToList<DataReading>()));
FeatureCreator.FACEValenceOptimizationFeatures.ForEach(x => x.SetData(fd.faceData.ToList<DataReading>()));
Log.LogMessage("Looking for configurations...");
svmConfs.Clear();
if (Directory.Exists(path + @"\STD"))
{
var files = Directory.GetFiles(path + @"\STD");
Log.LogMessage("Found STD! Contains " + files.Length + " configurations.");
foreach (var item in files)
{
svmConfs.Add(SVMConfiguration.Deserialize(File.ReadAllText(item)));
}
}
if (Directory.Exists(path + @"\META"))
{
var files = Directory.GetFiles(path + @"\META");
Log.LogMessage("Found META! Contains " + files.Length + " configurations.");
/* same procedure?? */
foreach (var item in files)
{
metaConfs.Add(MetaSVMConfiguration.Deserialize(File.ReadAllText(item)));
}
}
if (svmConfs.Count == 0 && metaConfs.Count == 0)
{
Log.LogMessage("No configurations found, maybe you should run some optimizations on some features.");
}
return true;
}
private void btn_ExportDataCompare_Click(object sender, EventArgs e)
{
FolderBrowserDialog fbd = new FolderBrowserDialog();
if (fbd.ShowDialog() == DialogResult.OK)
{
int counterino = 1;
var dirs = Directory.GetDirectories(fbd.SelectedPath);
foreach (var dirPath in dirs)
{
if (dirPath == "results") continue;
List<string> files = new List<string>()
{
"EEG.dat",
//"GSR.dat",
//"HR.dat",
//"KINECT.dat"
};
bool runIndividualTasks = true;
files.RemoveAll(f => !File.Exists($"{dirPath}/test/{f}") || !File.Exists($"{dirPath}/recall/{f}"));
string subject = dirPath.Split('\\').Last();
statusLabel.Text = $"{counterino++} / {dirs.Length}";
var metaLines = File.ReadAllLines(dirPath + "/meta.txt");
fdTest = new FusionData();
fdRecall = new FusionData();
var fdTestStatus = fdTest.LoadFromFile(files.Select(f => dirPath + "/test/" + f).ToArray());
if (!fdTest.Loaded)
{
//throw new Exception("I crashed because bad loading of fdTest");
Log.LogMessage("ERROR in " + dirPath + " couldn't load test data");
continue;
}
var fdRecallStatus = fdRecall.LoadFromFile(files.Select(x => dirPath + "/recall/" + x).ToArray());
if (!fdRecall.Loaded)
{
//throw new Exception("I crashed because bad loading of fdRecall");
Log.LogMessage("ERROR in " + dirPath + " couldn't load recall data");
continue;
}
var testEvents = File.ReadAllLines(dirPath + "/test/SecondTest.dat");
//fix waiting period offset til at være første event i testevents
int offset = int.Parse(metaLines.ToList().First(x => x.StartsWith("sync")).Split(':', '=').Last());
string stimul = metaLines[1].Split('=').Last();
string time = metaLines[0].Split('=').Last();
int waitPeriodDone = int.Parse(testEvents[0].Split('#')[0]);
//int firstTwoTasksDone = int.Parse(testEvents[2].Split('#')[0]);
int wholePeriodDone = int.Parse(testEvents[testEvents.Length - 1].Split('#')[0]);
//int waitPeriodDone = 180000;
List<TaskStartEnd> tasks = new List<TaskStartEnd>();
if (!runIndividualTasks)
{
tasks.Add(new TaskStartEnd(waitPeriodDone, wholePeriodDone, "full"));
}
else
{
List<string> taskEvents = testEvents.Where(x => x.Contains("TaskWizard - ")).ToList();
int start = waitPeriodDone;
for (int s = 0; s < taskEvents.Count; s++)
{
string[] currentEvent = taskEvents[s].Split(new char[] { '-', ' ', '#' }, StringSplitOptions.RemoveEmptyEntries);
int end = int.Parse(currentEvent[0]);
tasks.Add(new TaskStartEnd(start, end, currentEvent[currentEvent.Length - 1]));
start = end;
}
}
for (int ta = 0; ta < tasks.Count; ta++)
{
string csvTimePath = "csv/Time " + time + "/" + subject + "_" + tasks[ta].filenameAppend + "_";
string csvStimuliPath = "csv/Stimuli " + (stimul == "neu" ? "low" : "high") + "/" + subject + "_" + tasks[ta].filenameAppend + "_";
//string csvPath = "csv/" + subject + "_";
Directory.CreateDirectory("csv/Time " + time);
Directory.CreateDirectory("csv/Stimuli " + (stimul == "neu" ? "low" : "high"));
if (files.Contains("GSR.dat"))
{
Log.LogMessage("Starting GSR");
var fdTestGsr = fdTest.gsrData.SkipWhile(x => x.timestamp < tasks[ta].start).TakeWhile(x => x.timestamp < tasks[ta].end).Select(x => Tuple.Create(x.timestamp, (double)x.resistance)).ToList();
var fdRecallGsr = fdRecall.gsrData.SkipWhile(x => x.timestamp - offset < tasks[ta].start).TakeWhile(x => x.timestamp - offset < tasks[ta].end).Select(x => Tuple.Create(x.timestamp - offset, (double)x.resistance)).ToList();
var gsr = FilterData(
fdTestGsr,
fdRecallGsr
);
if (gsr.Item2.Count != 0 || gsr.Item3.Count != 0)
{
var gsrNorm = NormalizeFilterData(gsr);
var pearsCorr = MathNet.Numerics.Statistics.Correlation.Pearson(gsrNorm.Item1.GetRange(0, Math.Min(gsrNorm.Item1.Count, gsrNorm.Item2.Count)), gsrNorm.Item2.GetRange(0, Math.Min(gsrNorm.Item1.Count, gsrNorm.Item2.Count)));
//var nonTemporal = gsrNorm.Item1.Zip(gsrNorm.Item2, (a, b) => Tuple.Create(a, b)).OrderBy(x => x.Item1);
//var nonTempA = nonTemporal.Select(x => x.Item1).ToList();
//var nonTempB = nonTemporal.Select(x => x.Item2).ToList();
SavePng(csvTimePath + "GSR.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: ) - Red = test, blue = recall", gsrNorm.Item1, gsrNorm.Item2);
//SavePngScatter(csvTimePath + "GSR_Scatter.png", $"{subject} (Time: {time}, Stim: {stimul})", gsrNorm.Item1, gsrNorm.Item2);
SaveZip(csvTimePath + "GSR.csv", gsrNorm.Item1, gsrNorm.Item2);
//SavePng(csvTimePath + "GSR_nonTemporal.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", nonTempA, nonTempB);
int t;
if (int.TryParse(time, out t) && t != 0)
{
SavePng(csvStimuliPath + "GSR.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: ) - Red = test, blue = recall", gsrNorm.Item1, gsrNorm.Item2);
//SavePngScatter(csvStimuliPath + "GSR_Scatter.png", $"{subject} (Time: {time}, Stim: {stimul})", gsrNorm.Item1, gsrNorm.Item2);
SaveZip(csvStimuliPath + "GSR.csv", gsrNorm.Item1, gsrNorm.Item2);
}
}
Log.LogMessage("GSR done, data filtered: " + gsr.Item1.ToString("0.0") + "%");
}
if (files.Contains("HR.dat"))
{
Log.LogMessage("Starting HR");
var hr = FilterData(
fdTest.hrData.SkipWhile(x => x.timestamp < tasks[ta].start).TakeWhile(x => x.timestamp < tasks[ta].end).Select(x => Tuple.Create(x.timestamp, (double)x.BPM)).ToList(),
fdRecall.hrData.SkipWhile(x => x.timestamp - offset < tasks[ta].start).TakeWhile(x => x.timestamp - offset < tasks[ta].end).Select(x => Tuple.Create(x.timestamp - offset, (double)x.BPM)).ToList(),
20
);
if (hr.Item2.Count != 0 && hr.Item3.Count != 0)
{
var hrNorm = NormalizeFilterData(hr);
var setA = hrNorm.Item1;//.MedianFilter(25);
var setB = hrNorm.Item2;//.MedianFilter(25);
var pearsCorr = MathNet.Numerics.Statistics.Correlation.Pearson(setA.GetRange(0, Math.Min(setA.Count, setB.Count)), setB.GetRange(0, Math.Min(setA.Count, setB.Count)));
SavePng(csvTimePath + "HR.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", hrNorm.Item1, hrNorm.Item2);
//SavePngScatter(csvTimePath + "HR_Scatter.png", $"{subject} (Time: {time}, Stim: {stimul})", hrNorm.Item1, hrNorm.Item2);
SaveZip(csvTimePath + "HR.csv", setA, setB);
int t;
if (int.TryParse(time, out t) && t != 0)
{
SavePng(csvStimuliPath + "HR.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", hrNorm.Item1, hrNorm.Item2);
//SavePngScatter(csvStimuliPath + "HR_Scatter.png", $"{subject} (Time: {time}, Stim: {stimul})", hrNorm.Item1, hrNorm.Item2);
SaveZip(csvStimuliPath + "HR.csv", hrNorm.Item1, hrNorm.Item2);
//SavePng(csvStimuliPath + "GSR_nonTemporal.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", nonTempA, nonTempB);
}
}
Log.LogMessage($"HR done, data filtered: {hr.Item1.ToString("0.0")}%");
}
if (files.Contains("EEG.dat"))
{
Log.LogMessage("Starting EEG");
foreach (var item in Enum.GetNames(typeof(EEGDataReading.ELECTRODE)))
{
var eeg = FilterData(
fdTest.eegData.SkipWhile(x => x.timestamp < tasks[ta].start).TakeWhile(x => x.timestamp < tasks[ta].end).Select(x => Tuple.Create(x.timestamp, (double)x.data[item])).ToList(),
fdRecall.eegData.SkipWhile(x => x.timestamp - offset < tasks[ta].start).TakeWhile(x => x.timestamp - offset < tasks[ta].end).Select(x => Tuple.Create(x.timestamp - offset, (double)x.data[item])).ToList(),
8
);
if (eeg.Item2.Count == 0 || eeg.Item3.Count == 0) continue;
/*
var eegNorm = NormalizeFilterData(eeg);
var setA = eegNorm.Item1.VarianceFilter(64);
var setB = eegNorm.Item2.VarianceFilter(64);
*/
var setA = eeg.Item2.VarianceFilter(64).CalculateNormalized();
var setB = eeg.Item3.VarianceFilter(64).CalculateNormalized();
var min = Math.Min(setA.Count, setB.Count);
Log.LogMessage($"{item} done, data filtered: {eeg.Item1.ToString("0.0")}%");
var pearsCorr = MathNet.Numerics.Statistics.Correlation.Pearson(setA.GetRange(0, min), setB.GetRange(0, min));
SavePng(csvTimePath + "EEG_" + item + ".png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", setA, setB);
SaveZip(csvTimePath + "EEG_" + item + ".csv", setA, setB);
int t;
if (int.TryParse(time, out t) && t != 0)
{
SavePng(csvStimuliPath + "EEG_" + item + ".png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", setA, setB);
SaveZip(csvStimuliPath + "EEG_" + item + ".csv", setA, setB);
}
}
Log.LogMessage("EEG done");
}
if (files.Contains("KINECT.dat"))
{
Log.LogMessage("Starting Kinect");
foreach (Microsoft.Kinect.Face.FaceShapeAnimations item in Enum.GetValues(typeof(Microsoft.Kinect.Face.FaceShapeAnimations)))
{
if (item == Microsoft.Kinect.Face.FaceShapeAnimations.Count) continue;
var kinect = FilterData(
fdTest.faceData.SkipWhile(x => x.timestamp < tasks[ta].start).TakeWhile(x => x.timestamp < tasks[ta].end).Select(x => Tuple.Create(x.timestamp, (double)x.data[item])).ToList(),
fdRecall.faceData.SkipWhile(x => x.timestamp - offset < tasks[ta].start).TakeWhile(x => x.timestamp - offset < tasks[ta].end).Select(x => Tuple.Create(x.timestamp - offset, (double)x.data[item])).ToList(),
34
);
if (kinect.Item2.Count == 0 || kinect.Item3.Count == 0) continue;
var kiNorm = NormalizeFilterData(kinect);
var setA = kiNorm.Item1.MovingAverageFilter(25);
var setB = kiNorm.Item2.MovingAverageFilter(25);
var min = Math.Min(setA.Count, setB.Count);
Log.LogMessage($"{item} done, data filtered: {kinect.Item1.ToString("0.0")}%");
var pearsCorr = MathNet.Numerics.Statistics.Correlation.Pearson(setA.GetRange(0, min), setB.GetRange(0, min));
SavePng(csvTimePath + "FACE_" + item + ".png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", setA, setB);
SaveZip(csvTimePath + "FACE_" + item + ".csv", setA, setB);
int t;
if (int.TryParse(time, out t) && t != 0)
{
SavePng(csvStimuliPath + "FACE_" + item + ".png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", setA, setB);
SaveZip(csvStimuliPath + "FACE_" + item + ".csv", setA, setB);
//SavePng(csvStimuliPath + "GSR_nonTemporal.png", $"{subject} (Time: {time}, Stim: {stimul}, Corr: {pearsCorr.ToString("0.000")}) - Red = test, blue = recall", nonTempA, nonTempB);
}
}
Log.LogMessage("Kinect done");
}
}
//var resA = Pearson(result.Item2, result.Item3);//MathNet.Numerics.Statistics.Correlation.Pearson(result.Item2, result.Item3);
//var sigA = Significance(resA, result.Item2.Count);
//Log.LogMessage($"Best case pearson correlation: {resA.ToString("0.000")}");
//var resB = Pearson(result.Item2, result.Item4);//MathNet.Numerics.Statistics.Correlation.Pearson(result.Item2, result.Item4);
//var sigB = Significance(resB, result.Item2.Count);
//Log.LogMessage($"Second best case pearson correlation: {resB.ToString("0.000")}");
//List<string> toWrite = new List<string>();
//toWrite.Add($"[{dirPath.Split('\\').Last()}]");
//toWrite.Add($"Data removed={result.Item1}");
//toWrite.Add($"Pearson closest={resA}");
//toWrite.Add($"Significance={sigA}");
//toWrite.Add($"Pearson second closest={resB}");
//toWrite.Add($"Significance={sigB}");
//toWrite.Add("");
//File.AppendAllLines("results.txt", toWrite);
}
Log.LogMessage("DonnoDK!");
}
}
