diff options
author | Jon Bratseth <bratseth@gmail.com> | 2021-12-17 12:41:17 +0100 |
---|---|---|
committer | Jon Bratseth <bratseth@gmail.com> | 2021-12-17 12:41:17 +0100 |
commit | 601b117281b74a578126a0f3effead55bc79c680 (patch) | |
tree | 29619184a8459763cc024b23e74960e6c9ec7f81 /linguistics-components/abi-spec.json | |
parent | 767cb63af0f530605180f5438767406e1db27520 (diff) |
BERT -> WordPiece, make subword prefix configurable
Diffstat (limited to 'linguistics-components/abi-spec.json')
-rw-r--r-- | linguistics-components/abi-spec.json | 236 |
1 files changed, 122 insertions, 114 deletions
diff --git a/linguistics-components/abi-spec.json b/linguistics-components/abi-spec.json index 6dba8b602bd..39666fd93a3 100644 --- a/linguistics-components/abi-spec.json +++ b/linguistics-components/abi-spec.json @@ -1,5 +1,22 @@ { - "com.yahoo.language.bert.BertConfig$Builder": { + "com.yahoo.language.sentencepiece.Scoring": { + "superClass": "java.lang.Enum", + "interfaces": [], + "attributes": [ + "public", + "final", + "enum" + ], + "methods": [ + "public static com.yahoo.language.sentencepiece.Scoring[] values()", + "public static com.yahoo.language.sentencepiece.Scoring valueOf(java.lang.String)" + ], + "fields": [ + "public static final enum com.yahoo.language.sentencepiece.Scoring highestScore", + "public static final enum com.yahoo.language.sentencepiece.Scoring fewestSegments" + ] + }, + "com.yahoo.language.sentencepiece.SentencePieceConfig$Builder": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigInstance$Builder" @@ -9,23 +26,25 @@ ], "methods": [ "public void <init>()", - "public void <init>(com.yahoo.language.bert.BertConfig)", - "public com.yahoo.language.bert.BertConfig$Builder model(com.yahoo.language.bert.BertConfig$Model$Builder)", - "public com.yahoo.language.bert.BertConfig$Builder model(java.util.function.Consumer)", - "public com.yahoo.language.bert.BertConfig$Builder model(java.util.List)", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder collapseUnknowns(boolean)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder scoring(com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(java.util.function.Consumer)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(java.util.List)", "public final boolean dispatchGetConfig(com.yahoo.config.ConfigInstance$Producer)", "public final java.lang.String getDefMd5()", "public final java.lang.String getDefName()", "public final java.lang.String getDefNamespace()", "public final boolean getApplyOnRestart()", "public final void setApplyOnRestart(boolean)", - "public com.yahoo.language.bert.BertConfig build()" + "public com.yahoo.language.sentencepiece.SentencePieceConfig build()" ], "fields": [ "public java.util.List model" ] }, - "com.yahoo.language.bert.BertConfig$Model$Builder": { + "com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigBuilder" @@ -35,14 +54,14 @@ ], "methods": [ "public void <init>()", - "public void <init>(com.yahoo.language.bert.BertConfig$Model)", - "public com.yahoo.language.bert.BertConfig$Model$Builder language(java.lang.String)", - "public com.yahoo.language.bert.BertConfig$Model$Builder path(com.yahoo.config.FileReference)", - "public com.yahoo.language.bert.BertConfig$Model build()" + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Model)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder language(java.lang.String)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder path(com.yahoo.config.FileReference)", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model build()" ], "fields": [] }, - "com.yahoo.language.bert.BertConfig$Model": { + "com.yahoo.language.sentencepiece.SentencePieceConfig$Model": { "superClass": "com.yahoo.config.InnerNode", "interfaces": [], "attributes": [ @@ -50,13 +69,13 @@ "final" ], "methods": [ - "public void <init>(com.yahoo.language.bert.BertConfig$Model$Builder)", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder)", "public java.lang.String language()", "public java.nio.file.Path path()" ], "fields": [] }, - "com.yahoo.language.bert.BertConfig$Producer": { + "com.yahoo.language.sentencepiece.SentencePieceConfig$Producer": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigInstance$Producer" @@ -67,11 +86,44 @@ "abstract" ], "methods": [ - "public abstract void getConfig(com.yahoo.language.bert.BertConfig$Builder)" + "public abstract void getConfig(com.yahoo.language.sentencepiece.SentencePieceConfig$Builder)" ], "fields": [] }, - "com.yahoo.language.bert.BertConfig": { + "com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum": { + "superClass": "java.lang.Enum", + "interfaces": [], + "attributes": [ + "public", + "final", + "enum" + ], + "methods": [ + "public static com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum[] values()", + "public static com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum valueOf(java.lang.String)" + ], + "fields": [ + "public static final enum com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum highestScore", + "public static final enum com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum fewestSegments" + ] + }, + "com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring": { + "superClass": "com.yahoo.config.EnumNode", + "interfaces": [], + "attributes": [ + "public", + "final" + ], + "methods": [ + "public void <init>()", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum)" + ], + "fields": [ + "public static final com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum highestScore", + "public static final com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum fewestSegments" + ] + }, + "com.yahoo.language.sentencepiece.SentencePieceConfig": { "superClass": "com.yahoo.config.ConfigInstance", "interfaces": [], "attributes": [ @@ -83,9 +135,11 @@ "public static java.lang.String getDefName()", "public static java.lang.String getDefNamespace()", "public static java.lang.String getDefVersion()", - "public void <init>(com.yahoo.language.bert.BertConfig$Builder)", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Builder)", + "public boolean collapseUnknowns()", + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum scoring()", "public java.util.List model()", - "public com.yahoo.language.bert.BertConfig$Model model(int)" + "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model model(int)" ], "fields": [ "public static final java.lang.String CONFIG_DEF_MD5", @@ -95,56 +149,47 @@ "public static final java.lang.String[] CONFIG_DEF_SCHEMA" ] }, - "com.yahoo.language.bert.BertEmbedder$Builder": { + "com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder": { "superClass": "java.lang.Object", "interfaces": [], "attributes": [ - "public" + "public", + "final" ], "methods": [ "public void <init>()", + "public void <init>(java.lang.String)", "public void addModel(com.yahoo.language.Language, java.nio.file.Path)", - "public com.yahoo.language.bert.BertEmbedder$Builder addDefaultModel(java.nio.file.Path)", + "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder addDefaultModel(java.nio.file.Path)", "public java.util.Map getModels()", - "public com.yahoo.language.bert.BertEmbedder build()" + "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder setCollapseUnknowns(boolean)", + "public boolean getCollapseUnknowns()", + "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder setScoring(com.yahoo.language.sentencepiece.Scoring)", + "public com.yahoo.language.sentencepiece.Scoring getScoring()", + "public com.yahoo.language.sentencepiece.SentencePieceEmbedder build()" ], "fields": [] }, - "com.yahoo.language.bert.BertEmbedder": { + "com.yahoo.language.sentencepiece.SentencePieceEmbedder": { "superClass": "java.lang.Object", "interfaces": [ - "com.yahoo.language.process.Embedder", - "com.yahoo.language.process.Segmenter" + "com.yahoo.language.process.Segmenter", + "com.yahoo.language.process.Embedder" ], "attributes": [ "public" ], "methods": [ - "public void <init>(com.yahoo.language.bert.BertConfig)", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig)", + "public void <init>(com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder)", "public java.util.List segment(java.lang.String, com.yahoo.language.Language)", "public java.util.List embed(java.lang.String, com.yahoo.language.process.Embedder$Context)", - "public com.yahoo.tensor.Tensor embed(java.lang.String, com.yahoo.language.process.Embedder$Context, com.yahoo.tensor.TensorType)" + "public com.yahoo.tensor.Tensor embed(java.lang.String, com.yahoo.language.process.Embedder$Context, com.yahoo.tensor.TensorType)", + "public java.lang.String normalize(java.lang.String)" ], "fields": [] }, - "com.yahoo.language.sentencepiece.Scoring": { - "superClass": "java.lang.Enum", - "interfaces": [], - "attributes": [ - "public", - "final", - "enum" - ], - "methods": [ - "public static com.yahoo.language.sentencepiece.Scoring[] values()", - "public static com.yahoo.language.sentencepiece.Scoring valueOf(java.lang.String)" - ], - "fields": [ - "public static final enum com.yahoo.language.sentencepiece.Scoring highestScore", - "public static final enum com.yahoo.language.sentencepiece.Scoring fewestSegments" - ] - }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Builder": { + "com.yahoo.language.wordpiece.WordPieceConfig$Builder": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigInstance$Builder" @@ -154,25 +199,24 @@ ], "methods": [ "public void <init>()", - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder collapseUnknowns(boolean)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder scoring(com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(java.util.function.Consumer)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Builder model(java.util.List)", + "public void <init>(com.yahoo.language.wordpiece.WordPieceConfig)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Builder subwordPrefix(java.lang.String)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Builder model(com.yahoo.language.wordpiece.WordPieceConfig$Model$Builder)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Builder model(java.util.function.Consumer)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Builder model(java.util.List)", "public final boolean dispatchGetConfig(com.yahoo.config.ConfigInstance$Producer)", "public final java.lang.String getDefMd5()", "public final java.lang.String getDefName()", "public final java.lang.String getDefNamespace()", "public final boolean getApplyOnRestart()", "public final void setApplyOnRestart(boolean)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig build()" + "public com.yahoo.language.wordpiece.WordPieceConfig build()" ], "fields": [ "public java.util.List model" ] }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder": { + "com.yahoo.language.wordpiece.WordPieceConfig$Model$Builder": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigBuilder" @@ -182,14 +226,14 @@ ], "methods": [ "public void <init>()", - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Model)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder language(java.lang.String)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder path(com.yahoo.config.FileReference)", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model build()" + "public void <init>(com.yahoo.language.wordpiece.WordPieceConfig$Model)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Model$Builder language(java.lang.String)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Model$Builder path(com.yahoo.config.FileReference)", + "public com.yahoo.language.wordpiece.WordPieceConfig$Model build()" ], "fields": [] }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Model": { + "com.yahoo.language.wordpiece.WordPieceConfig$Model": { "superClass": "com.yahoo.config.InnerNode", "interfaces": [], "attributes": [ @@ -197,13 +241,13 @@ "final" ], "methods": [ - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Model$Builder)", + "public void <init>(com.yahoo.language.wordpiece.WordPieceConfig$Model$Builder)", "public java.lang.String language()", "public java.nio.file.Path path()" ], "fields": [] }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Producer": { + "com.yahoo.language.wordpiece.WordPieceConfig$Producer": { "superClass": "java.lang.Object", "interfaces": [ "com.yahoo.config.ConfigInstance$Producer" @@ -214,44 +258,11 @@ "abstract" ], "methods": [ - "public abstract void getConfig(com.yahoo.language.sentencepiece.SentencePieceConfig$Builder)" + "public abstract void getConfig(com.yahoo.language.wordpiece.WordPieceConfig$Builder)" ], "fields": [] }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum": { - "superClass": "java.lang.Enum", - "interfaces": [], - "attributes": [ - "public", - "final", - "enum" - ], - "methods": [ - "public static com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum[] values()", - "public static com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum valueOf(java.lang.String)" - ], - "fields": [ - "public static final enum com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum highestScore", - "public static final enum com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum fewestSegments" - ] - }, - "com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring": { - "superClass": "com.yahoo.config.EnumNode", - "interfaces": [], - "attributes": [ - "public", - "final" - ], - "methods": [ - "public void <init>()", - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum)" - ], - "fields": [ - "public static final com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum highestScore", - "public static final com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum fewestSegments" - ] - }, - "com.yahoo.language.sentencepiece.SentencePieceConfig": { + "com.yahoo.language.wordpiece.WordPieceConfig": { "superClass": "com.yahoo.config.ConfigInstance", "interfaces": [], "attributes": [ @@ -263,11 +274,10 @@ "public static java.lang.String getDefName()", "public static java.lang.String getDefNamespace()", "public static java.lang.String getDefVersion()", - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig$Builder)", - "public boolean collapseUnknowns()", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Scoring$Enum scoring()", + "public void <init>(com.yahoo.language.wordpiece.WordPieceConfig$Builder)", + "public java.lang.String subwordPrefix()", "public java.util.List model()", - "public com.yahoo.language.sentencepiece.SentencePieceConfig$Model model(int)" + "public com.yahoo.language.wordpiece.WordPieceConfig$Model model(int)" ], "fields": [ "public static final java.lang.String CONFIG_DEF_MD5", @@ -277,41 +287,39 @@ "public static final java.lang.String[] CONFIG_DEF_SCHEMA" ] }, - "com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder": { + "com.yahoo.language.wordpiece.WordPieceEmbedder$Builder": { "superClass": "java.lang.Object", "interfaces": [], "attributes": [ - "public" + "public", + "final" ], "methods": [ "public void <init>()", + "public void <init>(java.lang.String)", + "public com.yahoo.language.wordpiece.WordPieceEmbedder$Builder setSubwordPrefix(java.lang.String)", + "public java.lang.String getSubwordPrefix()", "public void addModel(com.yahoo.language.Language, java.nio.file.Path)", - "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder addDefaultModel(java.nio.file.Path)", + "public com.yahoo.language.wordpiece.WordPieceEmbedder$Builder addDefaultModel(java.nio.file.Path)", "public java.util.Map getModels()", - "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder setCollapseUnknowns(boolean)", - "public boolean getCollapseUnknowns()", - "public com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder setScoring(com.yahoo.language.sentencepiece.Scoring)", - "public com.yahoo.language.sentencepiece.Scoring getScoring()", - "public com.yahoo.language.sentencepiece.SentencePieceEmbedder build()" + "public com.yahoo.language.wordpiece.WordPieceEmbedder build()" ], "fields": [] }, - "com.yahoo.language.sentencepiece.SentencePieceEmbedder": { + "com.yahoo.language.wordpiece.WordPieceEmbedder": { "superClass": "java.lang.Object", "interfaces": [ - "com.yahoo.language.process.Segmenter", - "com.yahoo.language.process.Embedder" + "com.yahoo.language.process.Embedder", + "com.yahoo.language.process.Segmenter" ], "attributes": [ "public" ], "methods": [ - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceConfig)", - "public void <init>(com.yahoo.language.sentencepiece.SentencePieceEmbedder$Builder)", + "public void <init>(com.yahoo.language.wordpiece.WordPieceConfig)", "public java.util.List segment(java.lang.String, com.yahoo.language.Language)", "public java.util.List embed(java.lang.String, com.yahoo.language.process.Embedder$Context)", - "public com.yahoo.tensor.Tensor embed(java.lang.String, com.yahoo.language.process.Embedder$Context, com.yahoo.tensor.TensorType)", - "public java.lang.String normalize(java.lang.String)" + "public com.yahoo.tensor.Tensor embed(java.lang.String, com.yahoo.language.process.Embedder$Context, com.yahoo.tensor.TensorType)" ], "fields": [] } |