public void MapInput(Input input, IntermediateInput intermediate)
{
ReadOnlyMemory <char> text = ReadOnlyMemoryUtils.TrimWhiteSpace(input.Text);
// Handle comments, if any. If no comments are present, the value for that column
// in this row will be empty.
if (!ReadOnlyMemoryUtils.SplitOne(text, '#', out var left, out var right))
{
right = ReadOnlyMemory <char> .Empty;
}
intermediate.Comment = right;
var ator = ReadOnlyMemoryUtils.Split(left, _seps).GetEnumerator();
// Empty lines are filtered in the Input.MapComment step.
var notEmpty = ator.MoveNext();
Contracts.Assert(notEmpty);
ReadOnlyMemory <char> token = ator.Current;
// Parse the label.
if (_tryFloatParse(in token, out intermediate.Label))
{
intermediate.Weight = 1; // Default weight is of course 1.
}
internal static IEnumerable <int> GetColumnSet(this Schema schema, string metadataKind, string value)
{
for (int col = 0; col < schema.Count; col++)
{
var columnType = schema[col].Metadata.Schema.GetColumnOrNull(metadataKind)?.Type;
if (columnType != null && columnType.IsText)
{
ReadOnlyMemory <char> val = default;
schema[col].Metadata.GetValue(metadataKind, ref val);
if (ReadOnlyMemoryUtils.EqualsStr(value, val))
{
yield return(col);
}
}
}
}
/// <summary>
/// Returns the set of column ids which match the value of specified annotation kind.
/// The annotation type should be of type text.
/// </summary>
public static IEnumerable <int> GetColumnSet(this DataViewSchema schema, string annotationKind, string value)
{
for (int col = 0; col < schema.Count; col++)
{
var columnType = schema[col].Annotations.Schema.GetColumnOrNull(annotationKind)?.Type;
if (columnType is TextDataViewType)
{
ReadOnlyMemory <char> val = default;
schema[col].Annotations.GetValue(annotationKind, ref val);
if (ReadOnlyMemoryUtils.EqualsStr(value, val))
{
yield return(col);
}
}
}
}
#pragma warning restore 0649
public static void MapComment(Input input, CommentIndicator output)
{
// We expand a bit on the SVM-light comment strategy. In SVM-light, a comment line
// must have the # as the first character, and a totally whitespace or empty line
// is considered a parse error. However, for the purpose of detecting comments,
// we detect # after trimming whitespace, and also consider totally blank lines
// "comments" instead of whitespace.
ReadOnlyMemory <char> text = ReadOnlyMemoryUtils.TrimWhiteSpace(input.Text);
if (text.IsEmpty || text.Span[0] == '#')
{
output.IsComment = float.NaN;
}
else
{
output.IsComment = 0;
}
}