private static void RunData <T>(string filename) where T : Col, new()
{
var t = new Tbl(new List <Col>());
t.Read(filename);
var splitter = new Div2();
splitter.Split <T>(t, 1, 0);
Console.Out.WriteLine($"----- Data File: {filename} ------ ");
PrintSplit(splitter, 0, t);
}
public void Split(Tbl table, int colToSplitOn, int otherColumn)
{
table.Cols[colToSplitOn].SortNaturally();
before = table.Cols[colToSplitOn].Clone();
var other = table.Cols[otherColumn];
step = (int)Math.Pow(before.cells.Count, min);
start = before.cells.First();
stop = before.cells.Last();
ranges = new List <Col>();
epsilon = ((Num)other).SD * cohen;
SplitInternal(before, 1);
//gain /= len(i._lst);
}
private void Train(Row row, List <Col> cols)
{
string goal = row.GetGoals().First().RawValue;
if (!_classCounts.ContainsKey(goal))
{
List <Col> colsCopy = new List <Col>();
foreach (var col in cols)
{
colsCopy.Add(col.Clone());
}
_classCounts[goal] = new Tbl(colsCopy);
}
Tbl table = _classCounts[goal];
table.AddRow(row);
}
private static void RunData(string filename)
{
var t = new Tbl(new List <Col>());
t.AddPredictor(new ZeroR()
{
BootstrapMin = 3
}, "ZeroR");
t.AddPredictor(new NaiveBayes()
{
BootstrapMin = 4
}, "NaiveBayes");
t.Read(filename);
var results = t.GetResults();
Console.Out.WriteLine($"----- Data File: {filename} ------ ");
foreach (var r in results)
{
Console.Out.WriteLine($"----- Learner: {r.Label} ------ ");
Console.Out.WriteLine(r.ToString());
}
}
private static void PrintSplit(Div2 splitter, int xColumn, Tbl t)
{
StringBuilder b = new StringBuilder();
foreach (var y in splitter.ranges)
{
var indices = y.cells.Select(cell => cell.Index());
Num xCol = new Num(t.Cols[xColumn].Name);
var myXFriends = t.Cols[xColumn].cells.Where(cell1 => indices.Contains(cell1.Index()));
foreach (var xf in myXFriends)
{
xCol.AddCell(xf);
}
xCol.ToTreeString(b, 3, "x");
b.Append(" | ");
y.ToTreeString(b, 3, "y");
b.AppendLine("");
}
// b.AppendLine($"{splitter.before.Variety()} {splitter.gain}");
Console.Out.WriteLine(b.ToString());
}
