NLPclass/Named_entity_recognition_persian

1---2library_name: transformers3license: mit4language:5- fa6pipeline_tag: token-classification7---8Named entity recognition On Persian dataset9 10traindataset=20484 persian sentense 11 12valdataset=256113 14AutoTokenizer=HooshvareLab/bert-fa-base-uncased15 16ner_tags=17['O', 'B-pro',18'I-pro',19'B-pers', 20'I-pers', 21'B-org', 22'I-org',23'B-loc',24'I-loc', 25'B-fac',26'I-fac', 27'B-event',28'I-event']29 30training_args=31    learning_rate=2e-5,32    33    per_device_train_batch_size=16,34    35    per_device_eval_batch_size=16,36    37    num_train_epochs=4,38    39    weight_decay=0.0140    41 42Training Loss=0.00100043 44sample1:45  'entity': 'B-loc',46  'score': 0.9998902,47  'index': 2,48  'word': 'تهران',49 50sample2:51  'entity': 'B-pers',52  'score': 0.99988234,53  'index': 2,54  'word': 'عباس',55 56 57for use this model:58 59    from transformers import pipeline60    61    pipe = pipeline("token-classification", model="NLPclass/Named_entity_recognition_persian")62 63    sentence = ""64 65    predicted_ner = pipe(sentence)66    67    for entity in predicted_ner:68    69        print(f"Entity: {entity['word']}, Label: {entity['entity']}")