Hindi Lemmatizer


This model uses context and language knowledge to assign all forms and inflections of a word to a single root. This enables the pipeline to treat the past and present tense of a verb, for example, as the same word instead of two completely different words. The lemmatizer takes into consideration the context surrounding a word to determine which root is correct when the word form alone is ambiguous.

Open in Colab Download

How to use

lemmatizer = LemmatizerModel.pretrained("lemma", "hi") \
        .setInputCols(["token"]) \
nlp_pipeline = Pipeline(stages=[document_assembler, tokenizer, lemmatizer])
light_pipeline = LightPipeline(nlp_pipeline.fit(spark.createDataFrame([['']]).toDF("text")))
results = light_pipeline.fullAnnotate("उत्तर के राजा होने के अलावा, जॉन स्नो एक अंग्रेजी चिकित्सक और संज्ञाहरण और चिकित्सा स्वच्छता के विकास में अग्रणी है।")
val lemmatizer = LemmatizerModel.pretrained("lemma", "hi")
val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, lemmatizer))
val result = pipeline.fit(Seq.empty["उततर     अल,   एक  ििसक और हरण और िि वचछत  ि ें अगरण ै।"].toDS.toDF("text")).transform(data)
import nlu

text = ["""उत्तर के राजा होने के अलावा, जॉन स्नो एक अंग्रेजी चिकित्सक और संज्ञाहरण और चिकित्सा स्वच्छता के विकास में अग्रणी है।"""]
lemma_df = nlu.load('hi.lemma').predict(text, output_level='document')


[Row(annotatorType='token', begin=0, end=4, result='उत्तर', metadata={'sentence': '0'}, embeddings=[]),
Row(annotatorType='token', begin=6, end=7, result='विपिन', metadata={'sentence': '0'}, embeddings=[]),
Row(annotatorType='token', begin=9, end=12, result='राजा', metadata={'sentence': '0'}, embeddings=[]),
Row(annotatorType='token', begin=14, end=17, result='हो', metadata={'sentence': '0'}, embeddings=[]),
Row(annotatorType='token', begin=19, end=20, result='विपिन', metadata={'sentence': '0'}, embeddings=[]),

Model Information

Model Name: lemma
Type: lemmatizer
Compatibility: Spark NLP 2.5.5+
Edition: Official
Input labels: [token]
Output labels: [lemma]
Language: hi
Case sensitive: false
License: Open Source

Data Source

The model is imported from https://universaldependencies.org