diff options
Diffstat (limited to 'linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java')
-rw-r--r-- | linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java b/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java index ff7f4ae42bc..31964eac514 100644 --- a/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java +++ b/linguistics-components/src/main/java/com/yahoo/language/sentencepiece/SentencePieceEmbedder.java @@ -10,6 +10,7 @@ import com.yahoo.language.process.Segmenter; import com.yahoo.tensor.Tensor; import com.yahoo.tensor.TensorType; +import java.io.File; import java.nio.file.Path; import java.util.ArrayList; import java.util.Collections; @@ -136,13 +137,16 @@ public class SentencePieceEmbedder implements Segmenter, Embedder { return b.toString(); } - public static class Builder { + public static final class Builder { private final Map<Language, Path> models = new EnumMap<>(Language.class); private boolean collapseUnknowns = true; private Scoring scoring = Scoring.fewestSegments; - public Builder() { + public Builder() {} + + public Builder(String defaultModelFile) { + addDefaultModel(new File(defaultModelFile).toPath()); } private Builder(SentencePieceConfig config) { |