app.py

import streamlit as st
import os
import cv2
import torch
import torchaudio
import torchvision
import tensorflow as tf
from transformers import pipeline
from groq import Groq
from openai import OpenAI

# Set up the Groq client
client = Groq(api_key=os.environ.get("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"))

# Load a fake news detection model from Hugging Face
fake_news_pipeline = pipeline("text-classification", model="mrm8488/bert-tiny-finetuned-fake-news-detection")

# Streamlit UI
st.set_page_config(page_title="Fake News Detector", layout="centered")
st.title("\U0001F4F0 Fake News Detector")

# User input
news_text = st.text_area("Enter the news content to check:", height=200)

if st.button("Analyze News"):
    if not news_text.strip():
        st.warning("Please enter some text.")
    else:
        with st.spinner("Analyzing..."):
            # Check using Groq API
            chat_completion = client.chat.completions.create(
                messages=[{"role": "user", "content": f"Classify this news as Real or Fake: {news_text}"}],
                model="llama-3.3-70b-versatile",
                stream=False,
            )
            groq_result = chat_completion.choices[0].message.content.strip().lower()
            
            # Check using Hugging Face model
            hf_result = fake_news_pipeline(news_text)[0]['label'].lower()
            
            # Determine the final result
            if "fake" in groq_result or hf_result == "fake":
                st.error("\u274C This news is likely **Fake**!", icon="⚠️")
                st.markdown('<style>div.stAlert {background-color: #ffdddd;}</style>', unsafe_allow_html=True)
            elif "real" in groq_result or hf_result == "real":
                st.success("✅ This news is likely **Real**!", icon="✅")
                st.markdown('<style>div.stAlert {background-color: #ddffdd;}</style>', unsafe_allow_html=True)
            else:
                st.info("🤔 The result is uncertain. Please verify from trusted sources.")