README.md

metadata

language: mr
datasets:
  - wikiann
examples: null
widget:
  - text: राज्यसभा निवडणुकांसाठी उद्या मुंबईत मतदान होणार आहे.
    example_title: Sentence_1
  - text: विराट कोहली भारताकडून खेळतो.
    example_title: Sentence_2
  - text: नवी दिल्ली ही भारताची राजधानी आहे
    example_title: Sentence_3

Marathi Named Entity Recognition Model trained using transfer learning

Fine-tuning bert-base-multilingual-cased on Wikiann dataset for performing NER on Marathi language.

Label ID and its corresponding label name

Label list: (0), B-PER (1), I-PER (2), B-ORG (3), I-ORG (4), B-LOC (5), I-LOC (6)

Example

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("lakshaywadhwa1993/ner_marathi_bert")

model = AutoModelForTokenClassification.from_pretrained("lakshaywadhwa1993/ner_marathi_bert")

nlp = pipeline("ner", model=model, tokenizer=tokenizer)

example = ["राज्यसभा","निवडणुकांसाठी","मुंबईत","भाजपचे" ,"चिंचवडचे", "आमदार", "लक्ष्मण", "जगताप"]

results = nlp(example)

results