1 instantiation of ModelProto
Microsoft.ML.Tokenizers (1)
SentencepieceModel.cs (1)
180
ModelProto result =
new
();
38 references to ModelProto
Microsoft.ML.Tokenizers (38)
Model\LlamaTokenizer.cs (3)
19
internal LlamaTokenizer(
ModelProto
modelProto, bool addBos, bool addEos, IReadOnlyDictionary<string, int>? addedTokens = null) : base(modelProto, addBos, addEos, addedTokens)
40
ModelProto
modelProto =
ModelProto
.Parser.ParseFrom(modelStream);
Model\SentencePieceBaseModel.cs (1)
18
internal SentencePieceBaseModel(
ModelProto
modelProto, bool addBos = false, bool addEos = false, IReadOnlyDictionary<string, int>? specialTokens = null)
Model\SentencePieceBpeModel.cs (11)
26
internal SentencePieceBpeModel(
ModelProto
modelProto, bool addBos, bool addEos, IReadOnlyDictionary<string, int>? specialTokens = null) : base(modelProto, addBos, addEos, specialTokens)
34
if (piece.Type ==
ModelProto
.Types.SentencePiece.Types.Type.Byte)
180
if (type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused)
268
if (id.Type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused ||
458
if (type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused)
573
if (id.Type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused ||
726
if (type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused)
832
if (id.Type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused ||
1002
if (type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused)
1108
if (id.Type != (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused ||
1208
if (leftId.Type == (byte)
ModelProto
.Types.SentencePiece.Types.Type.Unused)
Model\SentencePieceTokenizer.cs (3)
23
internal SentencePieceTokenizer(
ModelProto
modelProto, bool addBos, bool addEos, IReadOnlyDictionary<string, int>? specialTokens = null)
451
ModelProto
modelProto =
ModelProto
.Parser.ParseFrom(modelStream);
Model\SentencePieceUnigramModel.cs (15)
21
private readonly (string Piece, float Score,
ModelProto
.Types.SentencePiece.Types.Type Type)[] _vocabReverse;
27
public SentencePieceUnigramModel(
ModelProto
modelProto, bool addBos, bool addEos, IReadOnlyDictionary<string, int>? specialTokens = null) : base(modelProto, addBos, addEos, specialTokens)
38
_vocabReverse = new (string Piece, float Score,
ModelProto
.Types.SentencePiece.Types.Type Type)[modelProto.Pieces.Count];
45
if (modelProto.Pieces[i].Type ==
ModelProto
.Types.SentencePiece.Types.Type.Normal ||
46
modelProto.Pieces[i].Type ==
ModelProto
.Types.SentencePiece.Types.Type.UserDefined ||
47
modelProto.Pieces[i].Type ==
ModelProto
.Types.SentencePiece.Types.Type.Unused)
56
else if (modelProto.Pieces[i].Type ==
ModelProto
.Types.SentencePiece.Types.Type.Byte)
60
else if (modelProto.Pieces[i].Type ==
ModelProto
.Types.SentencePiece.Types.Type.Unknown)
63
_vocabReverse[i] = (modelProto.Pieces[i].Piece, modelProto.Pieces[i].Score,
ModelProto
.Types.SentencePiece.Types.Type.Unknown);
85
_vocabReverse[modelProto.TrainerSpec.BosId] = (modelProto.TrainerSpec.BosPiece, 0f,
ModelProto
.Types.SentencePiece.Types.Type.Control);
86
_vocabReverse[modelProto.TrainerSpec.EosId] = (modelProto.TrainerSpec.EosPiece, 0f,
ModelProto
.Types.SentencePiece.Types.Type.Control);
87
_vocabReverse[modelProto.TrainerSpec.UnkId] = (modelProto.TrainerSpec.UnkPiece, 0f,
ModelProto
.Types.SentencePiece.Types.Type.Unknown);
92
_vocabReverse[modelProto.TrainerSpec.PadId] = (modelProto.TrainerSpec.PadPiece, 0f,
ModelProto
.Types.SentencePiece.Types.Type.Control);
506
if (_vocabReverse[ret].Type ==
ModelProto
.Types.SentencePiece.Types.Type.Unused)
516
float score = _vocabReverse[ret].Type ==
ModelProto
.Types.SentencePiece.Types.Type.UserDefined ? length * _maxScore - 0.1f : _vocabReverse[ret].Score;
SentencepieceModel.cs (5)
178
internal static
ModelProto
Parse(byte[] data, int start, int end)
180
ModelProto
result = new();
288
internal
ModelProto
ParseFrom(Stream stream)
302
return
ModelProto
.Parse(segment.Array!, start, end);
318
return
ModelProto
.Parse(ms.GetBuffer(), 0, (int)ms.Length);