Part of Speech for Yoruba

Description

This model annotates the part of speech of tokens in a text. The parts of speech annotated include PRON (pronoun), CCONJ (coordinating conjunction), and 15 others. The part of speech model is useful for extracting the grammatical structure of a piece of text automatically.

Open in Colab Download Copy S3 URI

How to use

...
pos = PerceptronModel.pretrained("pos_ud_ytb", "yo") \
.setInputCols(["document", "token"]) \
.setOutputCol("pos")
nlp_pipeline = Pipeline(stages=[document_assembler, sentence_detector, tokenizer, pos])
light_pipeline = LightPipeline(nlp_pipeline.fit(spark.createDataFrame([['']]).toDF("text")))
results = light_pipeline.fullAnnotate("Yato si jijẹ ọba ariwa, John Snow jẹ oṣoogun ara Gẹẹsi kan ati adari ninu idagbasoke anaesthesia ati imototo ilera.")
...
val pos = PerceptronModel.pretrained("pos_ud_ytb", "yo")
.setInputCols(Array("document", "token"))
.setOutputCol("pos")
val pipeline = new Pipeline().setStages(Array(document_assembler, sentence_detector, tokenizer, pos))
val data = Seq("Yato si jijẹ ọba ariwa, John Snow jẹ oṣoogun ara Gẹẹsi kan ati adari ninu idagbasoke anaesthesia ati imototo ilera.").toDF("text")
val result = pipeline.fit(data).transform(data)
import nlu

text = ["""Yato si jijẹ ọba ariwa, John Snow jẹ oṣoogun ara Gẹẹsi kan ati adari ninu idagbasoke anaesthesia ati imototo ilera."""]
pos_df = nlu.load('yo.pos').predict(text)
pos_df

Results

[Row(annotatorType='pos', begin=0, end=3, result='NOUN', metadata={'word': 'Yato'}),
Row(annotatorType='pos', begin=5, end=6, result='VERB', metadata={'word': 'si'}),
Row(annotatorType='pos', begin=8, end=11, result='VERB', metadata={'word': 'jijẹ'}),
Row(annotatorType='pos', begin=13, end=15, result='NOUN', metadata={'word': 'ọba'}),
Row(annotatorType='pos', begin=17, end=21, result='NOUN', metadata={'word': 'ariwa'}),
...]

Model Information

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

Data Source

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