class M2M100Transformer extends AnnotatorModel[M2M100Transformer] with HasBatchedAnnotate[M2M100Transformer] with ParamsAndFeaturesWritable with WriteOnnxModel with WriteOpenvinoModel with HasGeneratorProperties with WriteSentencePieceModel with HasEngine
M2M100 : multilingual translation model
M2M100 is a multilingual encoder-decoder (seq-to-seq) model trained for Many-to-Many multilingual translation.
The model can directly translate between the 9,900 directions of 100 languages.
Pretrained models can be loaded with pretrained
of the companion object:
val m2m100 = M2M100Transformer.pretrained() .setInputCols("document") .setOutputCol("generation")
The default model is "m2m100_418M"
, if no name is provided. For available pretrained models
please see the Models Hub.
For extended examples of usage, see M2M100TestSpec.
References:
- Beyond English-Centric Multilingual Machine Translation
- https://github.com/pytorch/fairseq/tree/master/examples/m2m_100
Paper Abstract:
Existing work in translation demonstrated the potential of massively multilingual machine translation by training a single model able to translate between any pair of languages. However, much of this work is English-Centric by training only on data which was translated from or to English. While this is supported by large sources of training data, it does not reflect translation needs worldwide. In this work, we create a true Many-to-Many multilingual translation model that can translate directly between any pair of 100 languages. We build and open source a training dataset that covers thousands of language directions with supervised data, created through large-scale mining. Then, we explore how to effectively increase model capacity through a combination of dense scaling and language-specific sparse parameters to create high quality models. Our focus on non-English-Centric models brings gains of more than 10 BLEU when directly translating between non-English directions while performing competitively to the best single systems of WMT. We open-source our scripts so that others may reproduce the data, evaluation, and final M2M-100 model.
Languages Covered:
Afrikaans (af), Amharic (am), Arabic (ar), Asturian (ast), Azerbaijani (az), Bashkir (ba), Belarusian (be), Bulgarian (bg), Bengali (bn), Breton (br), Bosnian (bs), Catalan; Valencian (ca), Cebuano (ceb), Czech (cs), Welsh (cy), Danish (da), German (de), Greeek (el), English (en), Spanish (es), Estonian (et), Persian (fa), Fulah (ff), Finnish (fi), French (fr), Western Frisian (fy), Irish (ga), Gaelic; Scottish Gaelic (gd), Galician (gl), Gujarati (gu), Hausa (ha), Hebrew (he), Hindi (hi), Croatian (hr), Haitian; Haitian Creole (ht), Hungarian (hu), Armenian (hy), Indonesian (id), Igbo (ig), Iloko (ilo), Icelandic (is), Italian (it), Japanese (ja), Javanese (jv), Georgian (ka), Kazakh (kk), Central Khmer (km), Kannada (kn), Korean (ko), Luxembourgish; Letzeburgesch (lb), Ganda (lg), Lingala (ln), Lao (lo), Lithuanian (lt), Latvian (lv), Malagasy (mg), Macedonian (mk), Malayalam (ml), Mongolian (mn), Marathi (mr), Malay (ms), Burmese (my), Nepali (ne), Dutch; Flemish (nl), Norwegian (no), Northern Sotho (ns), Occitan (post 1500) (oc), Oriya (or), Panjabi; Punjabi (pa), Polish (pl), Pushto; Pashto (ps), Portuguese (pt), Romanian; Moldavian; Moldovan (ro), Russian (ru), Sindhi (sd), Sinhala; Sinhalese (si), Slovak (sk), Slovenian (sl), Somali (so), Albanian (sq), Serbian (sr), Swati (ss), Sundanese (su), Swedish (sv), Swahili (sw), Tamil (ta), Thai (th), Tagalog (tl), Tswana (tn), Turkish (tr), Ukrainian (uk), Urdu (ur), Uzbek (uz), Vietnamese (vi), Wolof (wo), Xhosa (xh), Yiddish (yi), Yoruba (yo), Chinese (zh), Zulu (zu)
Example
import spark.implicits._ import com.johnsnowlabs.nlp.base.DocumentAssembler import com.johnsnowlabs.nlp.annotators.seq2seq.M2M100Transformer import org.apache.spark.ml.Pipeline val documentAssembler = new DocumentAssembler() .setInputCol("text") .setOutputCol("documents") val m2m100 = M2M100Transformer.pretrained("m2m100_418M") .setInputCols(Array("documents")) .setSrcLang("zh") .serTgtLang("en") .setMaxOutputLength(100) .setDoSample(false) .setOutputCol("generation") val pipeline = new Pipeline().setStages(Array(documentAssembler, m2m100)) val data = Seq( "生活就像一盒巧克力。" ).toDF("text") val result = pipeline.fit(data).transform(data) results.select("generation.result").show(truncate = false) +-------------------------------------------------------------------------------------------+ |result | +-------------------------------------------------------------------------------------------+ |[ Life is like a box of chocolate.] | +-------------------------------------------------------------------------------------------+
- Grouped
- Alphabetic
- By Inheritance
- M2M100Transformer
- HasEngine
- WriteSentencePieceModel
- HasGeneratorProperties
- WriteOpenvinoModel
- WriteOnnxModel
- HasBatchedAnnotate
- AnnotatorModel
- CanBeLazy
- RawAnnotator
- HasOutputAnnotationCol
- HasInputAnnotationCols
- HasOutputAnnotatorType
- ParamsAndFeaturesWritable
- HasFeatures
- DefaultParamsWritable
- MLWritable
- Model
- Transformer
- PipelineStage
- Logging
- Params
- Serializable
- Serializable
- Identifiable
- AnyRef
- Any
- Hide All
- Show All
- Public
- All
Parameters
A list of (hyper-)parameter keys this annotator can take. Users can set and get the parameter values through setters and getters, respectively.
-
val
batchSize: IntParam
Size of every batch (Default depends on model).
Size of every batch (Default depends on model).
- Definition Classes
- HasBatchedAnnotate
-
val
beamSize: IntParam
Beam size for the beam search algorithm (Default:
4
)Beam size for the beam search algorithm (Default:
4
)- Definition Classes
- HasGeneratorProperties
-
val
doSample: BooleanParam
Whether or not to use sampling, use greedy decoding otherwise (Default:
false
)Whether or not to use sampling, use greedy decoding otherwise (Default:
false
)- Definition Classes
- HasGeneratorProperties
-
val
engine: Param[String]
This param is set internally once via loadSavedModel.
This param is set internally once via loadSavedModel. That's why there is no setter
- Definition Classes
- HasEngine
-
var
ignoreTokenIds: IntArrayParam
A list of token ids which are ignored in the decoder's output (Default:
Array()
) -
val
inputAnnotatorTypes: Array[AnnotatorType]
Input annotator type : DOCUMENT
Input annotator type : DOCUMENT
- Definition Classes
- M2M100Transformer → HasInputAnnotationCols
-
val
maxInputLength: IntParam
max length of the input sequence (Default:
0
)max length of the input sequence (Default:
0
)- Definition Classes
- HasGeneratorProperties
-
val
maxOutputLength: IntParam
Maximum length of the sequence to be generated (Default:
20
)Maximum length of the sequence to be generated (Default:
20
)- Definition Classes
- HasGeneratorProperties
-
val
minOutputLength: IntParam
Minimum length of the sequence to be generated (Default:
0
)Minimum length of the sequence to be generated (Default:
0
)- Definition Classes
- HasGeneratorProperties
-
val
nReturnSequences: IntParam
The number of sequences to return from the beam search.
The number of sequences to return from the beam search.
- Definition Classes
- HasGeneratorProperties
-
val
noRepeatNgramSize: IntParam
If set to int >
0
, all ngrams of that size can only occur once (Default:0
)If set to int >
0
, all ngrams of that size can only occur once (Default:0
)- Definition Classes
- HasGeneratorProperties
-
val
outputAnnotatorType: String
Output annotator type : DOCUMENT
Output annotator type : DOCUMENT
- Definition Classes
- M2M100Transformer → HasOutputAnnotatorType
-
val
randomSeed: Option[Long]
Optional Random seed for the model.
Optional Random seed for the model. Needs to be of type
Int
.- Definition Classes
- HasGeneratorProperties
-
val
repetitionPenalty: DoubleParam
The parameter for repetition penalty (Default:
1.0
).The parameter for repetition penalty (Default:
1.0
).1.0
means no penalty. See this paper for more details.- Definition Classes
- HasGeneratorProperties
-
var
srcLang: Param[String]
Source Language (Default:
en
) -
val
stopTokenIds: IntArrayParam
Stop tokens to terminate the generation
Stop tokens to terminate the generation
- Definition Classes
- HasGeneratorProperties
-
val
task: Param[String]
Set transformer task, e.g.
Set transformer task, e.g.
"summarize:"
(Default:""
).- Definition Classes
- HasGeneratorProperties
-
val
temperature: DoubleParam
The value used to module the next token probabilities (Default:
1.0
)The value used to module the next token probabilities (Default:
1.0
)- Definition Classes
- HasGeneratorProperties
-
var
tgtLang: Param[String]
Target Language (Default:
fr
) -
val
topK: IntParam
The number of highest probability vocabulary tokens to keep for top-k-filtering (Default:
50
)The number of highest probability vocabulary tokens to keep for top-k-filtering (Default:
50
)- Definition Classes
- HasGeneratorProperties
-
val
topP: DoubleParam
If set to float <
1.0
, only the most probable tokens with probabilities that add up totopP
or higher are kept for generation (Default:1.0
)If set to float <
1.0
, only the most probable tokens with probabilities that add up totopP
or higher are kept for generation (Default:1.0
)- Definition Classes
- HasGeneratorProperties
-
val
vocabulary: MapFeature[String, Int]
Vocabulary used to encode the words to ids with bpeTokenizer.encode
Members
-
type
AnnotatorType = String
- Definition Classes
- HasOutputAnnotatorType
-
def
batchAnnotate(batchedAnnotations: Seq[Array[Annotation]]): Seq[Seq[Annotation]]
takes a document and annotations and produces new annotations of this annotator's annotation type
takes a document and annotations and produces new annotations of this annotator's annotation type
- batchedAnnotations
Annotations that correspond to inputAnnotationCols generated by previous annotators if any
- returns
any number of annotations processed for every input annotation. Not necessary one to one relationship
- Definition Classes
- M2M100Transformer → HasBatchedAnnotate
-
def
batchProcess(rows: Iterator[_]): Iterator[Row]
- Definition Classes
- HasBatchedAnnotate
-
final
def
clear(param: Param[_]): M2M100Transformer.this.type
- Definition Classes
- Params
-
def
copy(extra: ParamMap): M2M100Transformer
requirement for annotators copies
requirement for annotators copies
- Definition Classes
- RawAnnotator → Model → Transformer → PipelineStage → Params
-
def
explainParam(param: Param[_]): String
- Definition Classes
- Params
-
def
explainParams(): String
- Definition Classes
- Params
-
final
def
extractParamMap(): ParamMap
- Definition Classes
- Params
-
final
def
extractParamMap(extra: ParamMap): ParamMap
- Definition Classes
- Params
-
val
features: ArrayBuffer[Feature[_, _, _]]
- Definition Classes
- HasFeatures
- val generationConfig: StructFeature[GenerationConfig]
-
final
def
get[T](param: Param[T]): Option[T]
- Definition Classes
- Params
-
final
def
getDefault[T](param: Param[T]): Option[T]
- Definition Classes
- Params
- def getGenerationConfig: GenerationConfig
-
def
getInputCols: Array[String]
- returns
input annotations columns currently used
- Definition Classes
- HasInputAnnotationCols
-
def
getLazyAnnotator: Boolean
- Definition Classes
- CanBeLazy
-
final
def
getOrDefault[T](param: Param[T]): T
- Definition Classes
- Params
-
final
def
getOutputCol: String
Gets annotation column name going to generate
Gets annotation column name going to generate
- Definition Classes
- HasOutputAnnotationCol
-
def
getParam(paramName: String): Param[Any]
- Definition Classes
- Params
- def getSrcLangToken: Int
- def getTgtLangToken: Int
-
final
def
hasDefault[T](param: Param[T]): Boolean
- Definition Classes
- Params
-
def
hasParam(paramName: String): Boolean
- Definition Classes
- Params
-
def
hasParent: Boolean
- Definition Classes
- Model
-
final
def
isDefined(param: Param[_]): Boolean
- Definition Classes
- Params
-
final
def
isSet(param: Param[_]): Boolean
- Definition Classes
- Params
-
val
lazyAnnotator: BooleanParam
- Definition Classes
- CanBeLazy
-
def
onWrite(path: String, spark: SparkSession): Unit
- Definition Classes
- M2M100Transformer → ParamsAndFeaturesWritable
-
val
optionalInputAnnotatorTypes: Array[String]
- Definition Classes
- HasInputAnnotationCols
-
lazy val
params: Array[Param[_]]
- Definition Classes
- Params
-
var
parent: Estimator[M2M100Transformer]
- Definition Classes
- Model
-
def
save(path: String): Unit
- Definition Classes
- MLWritable
- Annotations
- @Since( "1.6.0" ) @throws( ... )
-
final
def
set[T](param: Param[T], value: T): M2M100Transformer.this.type
- Definition Classes
- Params
- def setGenerationConfig(value: GenerationConfig): M2M100Transformer.this.type
-
final
def
setInputCols(value: String*): M2M100Transformer.this.type
- Definition Classes
- HasInputAnnotationCols
-
def
setInputCols(value: Array[String]): M2M100Transformer.this.type
Overrides required annotators column if different than default
Overrides required annotators column if different than default
- Definition Classes
- HasInputAnnotationCols
-
def
setLazyAnnotator(value: Boolean): M2M100Transformer.this.type
- Definition Classes
- CanBeLazy
-
def
setMaxInputLength(value: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
final
def
setOutputCol(value: String): M2M100Transformer.this.type
Overrides annotation column name when transforming
Overrides annotation column name when transforming
- Definition Classes
- HasOutputAnnotationCol
-
def
setParent(parent: Estimator[M2M100Transformer]): M2M100Transformer
- Definition Classes
- Model
- def setSrcLang(value: String): M2M100Transformer.this.type
- def setTgtLang(value: String): M2M100Transformer.this.type
-
def
toString(): String
- Definition Classes
- Identifiable → AnyRef → Any
-
final
def
transform(dataset: Dataset[_]): DataFrame
Given requirements are met, this applies ML transformation within a Pipeline or stand-alone Output annotation will be generated as a new column, previous annotations are still available separately metadata is built at schema level to record annotations structural information outside its content
Given requirements are met, this applies ML transformation within a Pipeline or stand-alone Output annotation will be generated as a new column, previous annotations are still available separately metadata is built at schema level to record annotations structural information outside its content
- dataset
Dataset[Row]
- Definition Classes
- AnnotatorModel → Transformer
-
def
transform(dataset: Dataset[_], paramMap: ParamMap): DataFrame
- Definition Classes
- Transformer
- Annotations
- @Since( "2.0.0" )
-
def
transform(dataset: Dataset[_], firstParamPair: ParamPair[_], otherParamPairs: ParamPair[_]*): DataFrame
- Definition Classes
- Transformer
- Annotations
- @Since( "2.0.0" ) @varargs()
-
final
def
transformSchema(schema: StructType): StructType
requirement for pipeline transformation validation.
requirement for pipeline transformation validation. It is called on fit()
- Definition Classes
- RawAnnotator → PipelineStage
-
val
uid: String
- Definition Classes
- M2M100Transformer → Identifiable
-
def
write: MLWriter
- Definition Classes
- ParamsAndFeaturesWritable → DefaultParamsWritable → MLWritable
-
def
writeOnnxModel(path: String, spark: SparkSession, onnxWrapper: OnnxWrapper, suffix: String, fileName: String): Unit
- Definition Classes
- WriteOnnxModel
-
def
writeOnnxModels(path: String, spark: SparkSession, onnxWrappersWithNames: Seq[(OnnxWrapper, String)], suffix: String): Unit
- Definition Classes
- WriteOnnxModel
-
def
writeOpenvinoModel(path: String, spark: SparkSession, openvinoWrapper: OpenvinoWrapper, suffix: String, fileName: String): Unit
- Definition Classes
- WriteOpenvinoModel
-
def
writeOpenvinoModels(path: String, spark: SparkSession, ovWrappersWithNames: Seq[(OpenvinoWrapper, String)], suffix: String): Unit
- Definition Classes
- WriteOpenvinoModel
-
def
writeSentencePieceModel(path: String, spark: SparkSession, spp: SentencePieceWrapper, suffix: String, filename: String): Unit
- Definition Classes
- WriteSentencePieceModel
Parameter setters
-
def
setBatchSize(size: Int): M2M100Transformer.this.type
Size of every batch.
Size of every batch.
- Definition Classes
- HasBatchedAnnotate
-
def
setBeamSize(beamNum: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setDoSample(value: Boolean): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
- def setIgnoreTokenIds(tokenIds: Array[Int]): M2M100Transformer.this.type
-
def
setMaxOutputLength(value: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setMinOutputLength(value: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
- def setModelIfNotSet(spark: SparkSession, onnxWrappers: Option[EncoderDecoderWithoutPastWrappers], openvinoWrapper: Option[EncoderDecoderWithoutPastWrappers], spp: SentencePieceWrapper): M2M100Transformer.this.type
-
def
setNReturnSequences(beamNum: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setNoRepeatNgramSize(value: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
- def setRandomSeed(value: Int): M2M100Transformer.this.type
-
def
setRandomSeed(value: Long): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setRepetitionPenalty(value: Double): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setStopTokenIds(value: Array[Int]): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setTask(value: String): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setTemperature(value: Double): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setTopK(value: Int): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
-
def
setTopP(value: Double): M2M100Transformer.this.type
- Definition Classes
- HasGeneratorProperties
- def setVocabulary(value: Map[String, Int]): M2M100Transformer.this.type
Parameter getters
-
def
getBatchSize: Int
Size of every batch.
Size of every batch.
- Definition Classes
- HasBatchedAnnotate
-
def
getBeamSize: Int
- Definition Classes
- HasGeneratorProperties
-
def
getDoSample: Boolean
- Definition Classes
- HasGeneratorProperties
-
def
getEngine: String
- Definition Classes
- HasEngine
- def getIgnoreTokenIds: Array[Int]
-
def
getMaxOutputLength: Int
- Definition Classes
- HasGeneratorProperties
-
def
getMinOutputLength: Int
- Definition Classes
- HasGeneratorProperties
- def getModelIfNotSet: M2M100
-
def
getNReturnSequences: Int
- Definition Classes
- HasGeneratorProperties
-
def
getNoRepeatNgramSize: Int
- Definition Classes
- HasGeneratorProperties
-
def
getRandomSeed: Option[Long]
- Definition Classes
- HasGeneratorProperties
-
def
getRepetitionPenalty: Double
- Definition Classes
- HasGeneratorProperties
-
def
getStopTokenIds: Array[Int]
- Definition Classes
- HasGeneratorProperties
-
def
getTask: Option[String]
- Definition Classes
- HasGeneratorProperties
-
def
getTemperature: Double
- Definition Classes
- HasGeneratorProperties
-
def
getTopK: Int
- Definition Classes
- HasGeneratorProperties
-
def
getTopP: Double
- Definition Classes
- HasGeneratorProperties