Spaces:
Running
Running
File size: 2,675 Bytes
3c23ea3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 | import os
import requests
from bs4 import BeautifulSoup
from flask import Flask, request, jsonify
from flask_cors import CORS
from transformers import pipeline
app = Flask(__name__)
CORS(app)
print("Initializing AI... Downloading robust RoBERTa model. Patience.")
try:
# Upgrading to a highly generalized, multi-source fine-tuned model
# to severely reduce formatting and keyword bias.
classifier = pipeline("text-classification", model="hamzab/roberta-fake-news-classification")
print("Model loaded successfully. Server is ready.")
except Exception as e:
print(f"Error loading model: {e}")
def extract_text_from_url(url):
"""Scrapes the main paragraph text from a given URL."""
try:
headers = {'User-Agent': 'Mozilla/5.0'}
response = requests.get(url, headers=headers, timeout=10)
response.raise_for_status()
soup = BeautifulSoup(response.content, 'html.parser')
paragraphs = soup.find_all('p')
text = ' '.join([p.get_text() for p in paragraphs])
return text.strip()
except Exception:
return None
@app.route('/health', methods=['GET'])
def health_check():
return jsonify({"status": "online"}), 200
@app.route('/analyze', methods=['POST'])
def analyze_content():
data = request.get_json()
if not data:
return jsonify({"error": "No payload provided."}), 400
content = ""
if 'url' in data:
content = extract_text_from_url(data['url'])
if not content:
return jsonify({"error": "Could not extract text from the provided URL."}), 400
elif 'text' in data:
content = data['text']
else:
return jsonify({"error": "Invalid payload format."}), 400
if len(content) < 15:
return jsonify({"error": "Content too short for accurate analysis."}), 400
truncated_content = content[:2500]
try:
# Run inference
prediction = classifier(truncated_content)[0]
label = prediction['label'].lower()
confidence = round(prediction['score'] * 100)
# For this model: LABEL_0 is Fake, LABEL_1 is Real
if "0" in label or "fake" in label:
verdict = "fake"
elif "1" in label or "real" in label or "true" in label:
verdict = "real"
else:
verdict = "inconclusive"
return jsonify({
"verdict": verdict,
"confidence": confidence,
"text_preview": content[:150] + "..." if len(content) > 150 else content
}), 200
except Exception as e:
return jsonify({"error": str(e)}), 500
if __name__ == '__main__':
app.run(host='0.0.0.0', port=7860) |