abdoltd/Ner_model

Somali NER Model

This repository contains a Named Entity Recognition (NER) model fine-tuned using Hugging Face's transformers library. The model is trained to detect named entities in Somali text such as people, locations, organizations, and more.

📦 Model Details

• Task: Named Entity Recognition (NER)
• Language: Somali
• Checkpoint: ./ner_model/checkpoint-1302

---

🔧 How to Use

1. Install Dependencies

Make sure you have the required packages installed:

pip install transformers

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import TokenClassificationPipeline

# Load model and tokenizer from Hugging Face Hub
model = AutoModelForTokenClassification.from_pretrained("abdoltd/Ner_model")
tokenizer = AutoTokenizer.from_pretrained("abdoltd/Ner_model")

# Create pipeline manually
ner_pipeline = TokenClassificationPipeline(model=model, tokenizer=tokenizer, aggregation_strategy="simple")

# Somali text input
text = "Kylian Mbappé oo fariin hambalyo ah u diray kooxdiisii hore ee PSG."

# Run prediction
outputs = ner_pipeline(text)

# Display results
for entity in outputs:
    print(f"{entity['word']} -> {entity['entity_group']} (score: {entity['score']:.2f})")