BotSharp/BotSharp.NLP.UnitTest/Tokenize/TreebankTokenizerTest.cs

53 lines
1.4 KiB
C#
Raw Normal View History

2018-09-05 18:53:58 +00:00
using BotSharp.NLP.Tokenize;
using Microsoft.VisualStudio.TestTools.UnitTesting;
using System;
using System.Collections.Generic;
using System.Text;
namespace BotSharp.NLP.UnitTest.Tokenize
{
[TestClass]
public class TreebankTokenizerTest
{
[TestMethod]
public void ReplaceStartingQuoting()
{
var tokenizer = new TokenizerFactory<TreebankTokenizer>(new TokenizationOptions
{
}, SupportedLanguage.English);
var tokens = tokenizer.Tokenize("(\"«Hello World.");
}
[TestMethod]
public void ReplacePunctuation()
{
var tokenizer = new TokenizerFactory<TreebankTokenizer>(new TokenizationOptions
{
}, SupportedLanguage.English);
var tokens = tokenizer.Tokenize("Hello World...");
}
[TestMethod]
public void ReplaceBrackets()
{
var tokenizer = new TokenizerFactory<TreebankTokenizer>(new TokenizationOptions
{
}, SupportedLanguage.English);
var tokens = tokenizer.Tokenize("<Hello World.>");
}
[TestMethod]
public void ReplaceConventions()
{
var tokenizer = new TokenizerFactory<TreebankTokenizer>(new TokenizationOptions
{
}, SupportedLanguage.English);
var tokens = tokenizer.Tokenize("I cannot jump.");
}
}
}