diff options
Diffstat (limited to 'linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java')
-rw-r--r-- | linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java | 49 |
1 files changed, 49 insertions, 0 deletions
diff --git a/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java b/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java new file mode 100644 index 00000000000..1ba7c9b472d --- /dev/null +++ b/linguistics-components/src/test/java/com/yahoo/language/sentencepiece/SentencePieceTester.java @@ -0,0 +1,49 @@ +// Copyright Yahoo. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root. +// + +package com.yahoo.language.sentencepiece; + +import com.yahoo.language.Language; +import com.yahoo.tensor.Tensor; +import com.yahoo.tensor.TensorType; + +import java.nio.file.Path; + +import static org.junit.Assert.assertArrayEquals; +import static org.junit.Assert.assertEquals; + +class SentencePieceTester { + + private final SentencePieceEncoder encoder; + + public SentencePieceTester(Path model) { + this(new SentencePieceEncoder.Builder().addDefaultModel(model)); + } + + public SentencePieceTester(SentencePieceEncoder.Builder builder) { + this(builder.build()); + } + + public SentencePieceTester(SentencePieceEncoder encoder) { + this.encoder = encoder; + } + + public void assertEncoded(String input, Integer... expectedCodes) { + assertArrayEquals(expectedCodes, encoder.encode(input, Language.UNKNOWN).toArray()); + } + + public void assertEncoded(String input, String tensorType, String tensor) { + TensorType type = TensorType.fromSpec(tensorType); + Tensor expected = Tensor.from(type, tensor); + assertEquals(expected, encoder.encode(input, Language.UNKNOWN, type)); + } + + public void assertSegmented(String input, String... expectedSegments) { + assertSegmented(Language.UNKNOWN, input, expectedSegments); + } + + public void assertSegmented(Language language, String input, String... expectedSegments) { + assertArrayEquals(expectedSegments, encoder.segment(input, language).toArray()); + } + +} |