app.py

# import streamlit as st
# from transformers import pipeline

# # Load the model from the Hugging Face Hub
# ner_pipeline = pipeline("ner", model="Beehzod/smart-finetuned-ner")

# # Example predictions
# text = st.text_area('enter text: ')
# results = ner_pipeline(text)

# for entity in results:
#     print(f"Entity: {entity['word']}, Label: {entity['entity']}, Score: {entity['score']:.4f}")
#     st.json(entity)


import streamlit as st
from transformers import pipeline

# Load the model from the Hugging Face Hub
ner_pipeline = pipeline("ner", model="Beehzod/smart-finetuned-ner")

# Helper function to combine subword tokens
def merge_entities(entities):
    merged_entities = []
    current_entity = None

    for token in entities:
        token_text = token['word'].replace("##", "")  # Remove subword prefix if any

        # If token starts a new entity
        if token['entity'].startswith('B-') or (current_entity and token['entity'] != current_entity['entity']):
            # Add the previous entity if it exists
            if current_entity:
                # Average the score for all subwords in the entity
                current_entity['score'] /= current_entity['count']
                del current_entity['count']
                merged_entities.append(current_entity)

            # Start a new entity
            current_entity = {
                "word": token_text,
                "entity": token['entity'],
                "score": token['score'],
                "start": token['start'],
                "end": token['end'],
                "count": 1  # Helper count for score averaging
            }
        else:
            # Continue with the current entity
            current_entity["word"] += token_text
            current_entity["end"] = token['end']
            current_entity["score"] += token['score']
            current_entity["count"] += 1

    # Add the last entity if it exists
    if current_entity:
        current_entity['score'] /= current_entity['count']
        del current_entity['count']
        merged_entities.append(current_entity)

    return merged_entities

# Streamlit app to display entities
st.title('Named Entity Recognition (NER) with Streamlit')

text = st.text_area('Enter text for NER analysis:')

# Run NER model and merge results
if text:
    results = ner_pipeline(text)
    merged_results = merge_entities(results)

    # Display results
    for entity in merged_results:
        st.write(f"Entity: {entity['word']}, Label: {entity['entity']}, Score: {entity['score']:.4f}")
        st.json(entity)