app.py

from flask import Flask, request, jsonify, render_template
from transformers import AutoModelForTokenClassification, AutoTokenizer, pipeline
from flask_cors import CORS
import logging

logging.basicConfig(level=logging.DEBUG)

# Log untuk mendeteksi worker Gunicorn
if __name__ != '__main__':
    logging.info("Gunicorn worker started")

app = Flask(__name__)
CORS(app)

model_path = "./model"
logging.info("Loading model and tokenizer...")
model = AutoModelForTokenClassification.from_pretrained(model_path)
tokenizer = AutoTokenizer.from_pretrained(model_path)
nlp = pipeline("token-classification", model=model, tokenizer=tokenizer)
logging.info("Model and tokenizer loaded successfully.")

@app.route('/')
def home():
    return render_template('index.html')

@app.route('/pos_tag', methods=['POST'])
def pos_tag():
    data = request.json
    text = data.get('text') if data else None

    if not text:
        return jsonify({"error": "Please provide input text"}), 400

    # Handling UTF-8 encoding issues
    text = text.encode('utf-8').decode('utf-8')

    try:
        results = nlp(text)
        tagged_tokens = [
            {"word": res["word"].replace('Ġ', ''), 
             "tag": res["entity"].replace('B-', '').replace('I-', '')}
            for res in results
        ]
        return jsonify({"tagged_tokens": tagged_tokens})
    except Exception as e:
        logging.error(f"Error processing text: {str(e)}")
        return jsonify({"error": "Error processing text"}), 500

if __name__ == '__main__':
    app.run(port=5007, debug=True)