public Reconciler(TextNormalizerEstimator.CaseNormalizationMode textCase, bool keepDiacritics, bool keepPunctuations, bool keepNumbers) { _textCase = textCase; _keepDiacritics = keepDiacritics; _keepPunctuations = keepPunctuations; _keepNumbers = keepNumbers; }
/// <summary> /// Normalizes input text by changing case, removing diacritical marks, punctuation marks and/or numbers. /// </summary> /// <param name="input">The column to apply to.</param> /// <param name="textCase">Casing text using the rules of the invariant culture.</param> /// <param name="keepDiacritics">Whether to keep diacritical marks or remove them.</param> /// <param name="keepPunctuations">Whether to keep punctuation marks or remove them.</param> /// <param name="keepNumbers">Whether to keep numbers or remove them.</param> public static Scalar <string> NormalizeText(this Scalar <string> input, TextNormalizerEstimator.CaseNormalizationMode textCase = TextNormalizerEstimator.CaseNormalizationMode.Lower, bool keepDiacritics = false, bool keepPunctuations = true, bool keepNumbers = true) => new OutPipelineColumn(input, textCase, keepDiacritics, keepPunctuations, keepNumbers);
public OutPipelineColumn(Scalar <string> input, TextNormalizerEstimator.CaseNormalizationMode textCase, bool keepDiacritics, bool keepPunctuations, bool keepNumbers) : base(new Reconciler(textCase, keepDiacritics, keepPunctuations, keepNumbers), input) { Input = input; }