1 type derived from SentencePieceTokenizer
Microsoft.ML.Tokenizers (1)
Model\LlamaTokenizer.cs (1)
17public sealed class LlamaTokenizer : SentencePieceTokenizer
11 references to SentencePieceTokenizer
Microsoft.ML.GenAI.Core (2)
Pipeline\CausalLMPipeline.cs (2)
271SentencePieceTokenizer bpeTokenizer => bpeTokenizer.Decode(tokenIds.Concat(tokenIds), considerSpecialTokens: true) ?? throw new InvalidOperationException("Failed to decode token ids"), 277SentencePieceTokenizer bpeTokenizer => bpeTokenizer.Decode(tokenIds, considerSpecialTokens: true) ?? throw new InvalidOperationException("Failed to decode token ids"),
Microsoft.ML.Tokenizers (2)
Model\SentencePieceTokenizer.cs (2)
1738static void TryDecodeAsSpecialToken(SentencePieceTokenizer tokenizer, int id, bool considerSpecialTokens, ref ValueStringBuilder sb) 1982static OperationStatus TryDecodeAsSpecialToken(SentencePieceTokenizer tokenizer, int id, bool considerSpecialTokens, Span<char> buffer, ref int charsWritten)
Microsoft.ML.Tokenizers.Tests (7)
LlamaTests.cs (5)
69PropertyInfo? propertyInfo = typeof(SentencePieceTokenizer).GetProperty("TreatWhitespaceAsSuffix", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.Public); 511SentencePieceTokenizer sentencePieceBpe = (tokenizer as SentencePieceTokenizer)!; 565SentencePieceTokenizer sentencePieceBpe = (tokenizer as SentencePieceTokenizer)!;
TokenizerTests.cs (2)
155if (tokenizer is SentencePieceTokenizer) 173if (tokenizer is SentencePieceTokenizer)