TikTokenTokenizer
public string ComponentId { get; }
Language: C#
Return Value: Guid.NewGuid().ToString()
public string ComponentType
Language: C#
public NeuralComponentState State { get; set; }
Language: C#
Return Value: NeuralComponentState.Uninitialized
public INeuralComponentInfo ComponentInfo { get; }
Language: C#
public int VocabularySize { get; }
Language: C#
public ITokenizerSpecialTokens SpecialTokens
Language: C#
public TikTokenTokenizer(string modelName = "cl100k_base", ITokenizerSpecialTokens? specialTokens = null, int vocabularySize = 0)
Language: C#
public static TikTokenTokenizer ForModel(string model)
Language: C#
Создает токенизатор для конкретной модели OpenAI.
public Task InitializeAsync(CancellationToken cancellationToken = default)
Language: C#
public Task ShutdownAsync()
Language: C#
public Task<int[]> EncodeAsync(string text, CancellationToken cancellationToken = default)
Language: C#
public Task<string> DecodeAsync(int[] tokenIds, CancellationToken cancellationToken = default)
Language: C#
public Task<int> CountTokensAsync(string text)
Language: C#
public async Task<int[][]> EncodeBatchAsync(IEnumerable<string> texts, CancellationToken cancellationToken = default)
Language: C#
public Task<float[]?> GetTokenEmbeddingAsync(int tokenId)
Language: C#
public void Dispose()
Language: C#