1 type derived from SentencePieceTokenizer
Microsoft.ML.Tokenizers (1)
Model\LlamaTokenizer.cs (1)
17
public sealed class LlamaTokenizer :
SentencePieceTokenizer
3 instantiations of SentencePieceTokenizer
Microsoft.ML.Tokenizers (3)
Model\SentencePieceTokenizer.cs (3)
464
return new
SentencePieceTokenizer
(modelProto, addBeginningOfSentence, addEndOfSentence, specialTokens);
523
return new
SentencePieceTokenizer
(model);
757
return new
SentencePieceTokenizer
(model);
151 references to SentencePieceTokenizer
Microsoft.ML.GenAI.Core (2)
Pipeline\CausalLMPipeline.cs (2)
275
SentencePieceTokenizer
bpeTokenizer => bpeTokenizer.Decode(tokenIds.Concat(tokenIds), considerSpecialTokens: true) ?? throw new InvalidOperationException("Failed to decode token ids"),
281
SentencePieceTokenizer
bpeTokenizer => bpeTokenizer.Decode(tokenIds, considerSpecialTokens: true) ?? throw new InvalidOperationException("Failed to decode token ids"),
Microsoft.ML.Tokenizers (7)
Model\SentencePieceTokenizer.cs (7)
451
public static
SentencePieceTokenizer
Create(
468
/// Creates a Unigram <see cref="
SentencePieceTokenizer
"/> from an in-memory vocabulary of (piece, score) pairs.
487
/// <returns>A new <see cref="
SentencePieceTokenizer
"/> instance.</returns>
498
public static
SentencePieceTokenizer
Create(
527
/// Creates a Unigram <see cref="
SentencePieceTokenizer
"/> by parsing a Hugging Face <c>tokenizer.json</c>
534
/// <returns>A new <see cref="
SentencePieceTokenizer
"/> instance.</returns>
562
public static
SentencePieceTokenizer
CreateFromTokenizerJson(
Microsoft.ML.Tokenizers.Tests (142)
LlamaTests.cs (5)
69
PropertyInfo? propertyInfo = typeof(
SentencePieceTokenizer
).GetProperty("TreatWhitespaceAsSuffix", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.Public);
511
SentencePieceTokenizer
sentencePieceBpe = (tokenizer as
SentencePieceTokenizer
)!;
565
SentencePieceTokenizer
sentencePieceBpe = (tokenizer as
SentencePieceTokenizer
)!;
SentencePieceTests.cs (57)
20
Assert.ThrowsAny<ArgumentException>(() =>
SentencePieceTokenizer
.Create(null!));
27
Assert.ThrowsAny<ArgumentException>(() =>
SentencePieceTokenizer
.Create(empty));
36
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
45
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(stream);
59
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(ms);
72
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(stream);
94
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(stream);
109
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(stream,
143
SentencePieceTokenizer
bpe =
SentencePieceTokenizer
.Create(bpeStream,
148
SentencePieceTokenizer
unigram =
SentencePieceTokenizer
.Create(unigramStream,
175
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
185
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
197
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
231
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
253
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
268
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
286
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
307
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
326
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
362
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
379
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
393
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
411
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
427
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
441
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
453
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
478
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
495
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(stream);
505
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
518
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
535
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
551
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
563
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
574
Assert.ThrowsAny<Exception>(() =>
SentencePieceTokenizer
.Create(ms));
587
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
602
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
617
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
639
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
656
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
671
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
684
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
707
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
730
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model.ToArray());
749
SentencePieceTokenizer
tokenizer = CreateFromSyntheticModel(model);
768
Assert.Throws<ArgumentException>(() =>
SentencePieceTokenizer
.Create(ms));
788
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(ms);
796
private static
SentencePieceTokenizer
CreateFromSyntheticModel(
804
return
SentencePieceTokenizer
.Create(ms, addBos, addEos);
TokenizerTests.cs (2)
155
if (tokenizer is
SentencePieceTokenizer
)
173
if (tokenizer is
SentencePieceTokenizer
)
UnigramTests.cs (78)
20
private static
SentencePieceTokenizer
_unigramTokenizer = CreateUnigramTokenizer();
21
private static
SentencePieceTokenizer
_unigramTokenizerWithSpecialTokens = CreateUnigramTokenizerWithSpecialTokens();
23
private static
SentencePieceTokenizer
CreateUnigramTokenizer()
27
return
SentencePieceTokenizer
.Create(remoteStream);
55
private static
SentencePieceTokenizer
CreateUnigramTokenizerWithSpecialTokens()
59
return
SentencePieceTokenizer
.Create(remoteStream, specialTokens:
283
SentencePieceTokenizer
tokenizer,
535
private static void DecodeWithTokenizerTest(
SentencePieceTokenizer
tokenizer, string decodedString, int[] ids)
581
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(
602
SentencePieceTokenizer
.Create((IEnumerable<(string Piece, float Score)>)null!, unkId: 0));
610
SentencePieceTokenizer
.Create(vocab, unkId: 5));
617
SentencePieceTokenizer
jsonTokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(
646
SentencePieceTokenizer
.CreateFromTokenizerJson(null!));
655
SentencePieceTokenizer
jsonTokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(
684
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(
703
SentencePieceTokenizer
.Create(vocab, unkId: 0, addBeginningOfSentence: true));
730
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false);
781
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
808
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
831
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
839
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1020
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1050
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1073
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));
1089
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1109
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1125
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1141
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));
1157
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));
1256
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));
1275
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(
1292
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.Create(
1319
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1339
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1358
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1377
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1397
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1423
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1443
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1463
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1483
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1512
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1541
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1561
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1581
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1601
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1620
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false);
1640
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1660
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1680
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false);
1701
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1715
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false));
1734
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(
1772
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1807
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1836
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));
1858
SentencePieceTokenizer
tokenizer =
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false);
1888
SentencePieceTokenizer
.CreateFromTokenizerJson(stream, addBeginningOfSentence: false, addEndOfSentence: false));