/// <summary>
/// Exports data as a CSV file with the first column being the date/time encoded as a double.
/// Useful for loading data in Matlab.
/// </summary>
/// <param name="FullFileName"></param>
/// <param name="GCC"></param>
public static void ExportMatlabData(string FullFileName, GrowthCurveCollection GCC)
{
StreamWriter SW = new StreamWriter(FullFileName);
string TitleLine = "Time,";//this will hold the titles for everything below
HashSet <double> Times = new HashSet <double>();
foreach (GrowthCurve GC in GCC)
{
foreach (double d in GC.Select((x) => x.time_as_double))
{
Times.Add(d);
}
TitleLine += GC.ToString() + ",";
}
//Below assumes the time is the same for all of them
SW.WriteLine(TitleLine);
List <double> times = Times.ToList();
times.Sort();
foreach (double time in times)
{
string line = time.ToString() + ",";
foreach (GrowthCurve GR in GCC)
{
int indexPos = -1;
double[] timesGC = GR.TimeValues_As_Double;
if (SimpleFunctions.ValueInArray(timesGC, time, ref indexPos))
{
//decide if this timepoint was included
line += GR.ODValues[indexPos].ToString() + ",";
double DateX = GR[indexPos].time_as_double;
}
else
{
line += "-999,";
}
}
SW.WriteLine(line);
}
SW.Close();
}
public static void ExportDataDEPRECATED(string FullFileName, GrowthCurveCollection GCC)
{
bool LagData = false;
//FullFileName = "C:\\FullName.csv";
StreamWriter SW = new StreamWriter(FullFileName);
SW.WriteLine("Fitted Data Results");
SW.WriteLine("Name, Doubling Time(Hrs),Growth Rate, How Determined?,NumPoints,R2,RMSE, Maximum GrowthRate,MaxOD,Notes,Linear-Fit Slope,Reduction in absolute error from ExpFit,LagTime,Reduction in Sum of Squares from Exp Fit,TimeTill_OD_0.02,EndOD,TreatmentGroup");
string TitleLine = "Time,";//this will hold the titles for everything below
foreach (GrowthCurve GR in GCC)
{
LagData = false;
TitleLine += GR.ToString() + " OD," + "Flag,";
string newline = GR.ToString() + ",";
double ActualGrowth = Math.Log(2) / GR.GrowthRate.GrowthRate;
if (GR.ValidDataSet)
{
newline += ActualGrowth.ToString("n5") + "," + GR.GrowthRate.GrowthRate.ToString() + "," + GR.GrowthRate.FittingUsed + "," + GR.GrowthRate.NumPoints + "," + GR.GrowthRate.R2.ToString("n4") + ","
+ GR.GrowthRate.RMSE.ToString("n5") + "," + GR.MaxGrowthRate.MaxGrowthRate.ToString("n5") + "," + GR.ODValues.Max().ToString("n4") + "," + GR.GrowthRate.Notes;
if (GR.LinearModelFitted && GR.ExpModelFitted)
{
double dif = GR.LinFit.AbsError - GR.ExpFit.AbsError;
double dif2 = GR.LinFit.calculateResidualSumofSquares() - GR.ExpFit.calculateResidualSumofSquares();
newline += "," + GR.LinFit.Parameters[1].ToString("n5") + "," + dif.ToString("n5") + ",DEPRECATED," + dif2.ToString("n5") + ",";
}//+","+RMSEdiff.ToString("n5")+","+GR.LinFit.RMSE.ToString()+","+GR.ExpFit.RMSE.ToString(); }//report the linear fitted slope if possible
else
{
newline += ",No Exp Fit to Compare Against,";
}
newline += GR.HoursTillODReached(.02).ToString() + ",";
newline += GR.ODValues.Last().ToString() + ",";
}
else
{
newline += ",,Weird Data:Blank??,,,,,,,,,,,,,";
}
SW.WriteLine(newline);
}
//Below assumes the time is the same for all of them
string Intermissionline = "Complete Data Listing Below";
if (LagData)
{
Intermissionline += "-Initial OD Present";
}
;
SW.WriteLine(Intermissionline);
SW.WriteLine(TitleLine);
HashSet <DateTime> dtimes = new HashSet <DateTime>();
foreach (GrowthCurve gc in GCC)
{
foreach (DateTime dt in gc.Select((x) => x.time))
{
dtimes.Add(dt);
}
}
List <DateTime> DateTimesinFile = dtimes.Select((x) => x).ToList();
DateTimesinFile.Sort();
foreach (DateTime DT in DateTimesinFile)
{
string line = DT.ToString() + ",";
foreach (GrowthCurve GR in GCC)
{
int indexPos = -1;
DateTime[] timeValues = GR.Times;
if (SimpleFunctions.ValueInArray(timeValues, DT, ref indexPos))
{
//decide if this timepoint was included
line += GR.ODValues[indexPos].ToString() + ",";
double DateX = GR[indexPos].time_as_double;
if (GR.FittedXValues != null && SimpleFunctions.ValueInArray(GR.FittedXValues, DateX)) //now decide if it made it into the fit
{
line += "0,";
}
else
{
line += "1,";
}
}
else
{
line += "-999,-999,";
}
}
SW.WriteLine(line);
}
SW.Close();
}
public LinearFit(double[] XGOOD, double[] YGOOD)
{
name = "Linear";
if ((XGOOD.Length < 2 || YGOOD.Length < 2) || (XGOOD.Length != YGOOD.Length))
{
throw new Exception("Linear fit can't work with less then 2 points or unequal matrices");
}
if (XGOOD.Count(n => !SimpleFunctions.IsARealNumber(n)) > 0 || YGOOD.Count(n => !SimpleFunctions.IsARealNumber(n)) > 0)
{
throw new Exception("Linear fit can't work on data that contains non real numbers");
}
x = XGOOD;
y = YGOOD;
pParameters = new double[2];//linear model
FitModel();
}
