app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
from wordcloud import WordCloud
from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification

# ✅ MUST be first Streamlit command
st.set_page_config(page_title="📰 News Classifier & Q&A App", layout="wide")

# ----------------- Model Loader -----------------
@st.cache_resource
def load_text_classifier():
    model_name = "MihanTilk/News_Classifier"
    tokenizer = AutoTokenizer.from_pretrained(model_name)
    model = AutoModelForSequenceClassification.from_pretrained(
        model_name
    )
    return pipeline("text-classification", model=model, tokenizer=tokenizer)

# Load Classifier & QA pipeline
classifier = load_text_classifier()
qa_pipeline = pipeline("question-answering", model="deepset/roberta-base-squad2")

# ----------------- CSS Styling -----------------
st.markdown(
    """
    <style>
        .main { background-color: #f4f4f4; }
        .stTextInput, .stFileUploader { border: 2px solid #ff4b4b; border-radius: 10px; }
        .stButton>button { background-color: #ff4b4b; color: white; border-radius: 10px; }
        .stDownloadButton>button { background-color: #4CAF50; color: white; border-radius: 10px; }
        h1, h2, h3, h4, h5, h6, p { color: #333333; }
    </style>
    """,
    unsafe_allow_html=True
)

# ----------------- App Title -----------------
st.title("📰 News Classification & Q&A App")
st.markdown("<h4 style='color:#ff4b4b;'>Upload a CSV to classify news headlines and ask questions!</h4>", unsafe_allow_html=True)

# ----------------- Upload CSV -----------------
st.subheader("📂 Upload a CSV File")
uploaded_file = st.file_uploader("Choose a CSV file...", type=["csv"])

if uploaded_file:
    # Read and preprocess
    df = pd.read_csv(uploaded_file)
    if "content" not in df.columns:
        st.error("❌ The uploaded CSV must contain a 'content' column.")
        st.stop()

    # Preprocess text
    df['cleaned_text'] = df['content'].astype(str).str.lower().str.strip()
    st.write("📊 Preview of Uploaded Data:", df.head())

    # ----------------- Classification -----------------
    with st.spinner("🔍 Classifying news articles..."):
        df['class'] = df['cleaned_text'].apply(lambda text: classifier(text)[0]['label'])

    st.success("✅ Classification Complete!")
    st.write("🔎 Classified Results:", df[['content', 'class']].head())

    # ----------------- Download -----------------
    st.subheader("📥 Download Results")
    csv_output = df.to_csv(index=False).encode('utf-8')
    st.download_button("Download Output CSV", data=csv_output, file_name="output.csv", mime="text/csv")

    # ----------------- Q&A Section -----------------
    st.subheader("💬 Ask a Question")
    question = st.text_input("🔍 What do you want to know about the content?")

    if st.button("Get Answer"):
        context = " ".join(df['cleaned_text'].tolist())
        with st.spinner("Answering..."):
            result = qa_pipeline(question=question, context=context)
        st.success(f"📝 **Answer:** {result['answer']}")

    # ----------------- Word Cloud -----------------
    st.subheader("☁️ Word Cloud of News Text")
    text = " ".join(df['cleaned_text'].tolist())
    wordcloud = WordCloud(width=800, height=400, background_color="white").generate(text)

    fig, ax = plt.subplots()
    ax.imshow(wordcloud, interpolation="bilinear")
    ax.axis("off")
    st.pyplot(fig)

# ----------------- Footer -----------------
st.markdown("---")
st.markdown("<p style='text-align:center; color:#666;'>🚀 Built with using Streamlit & Hugging Face</p>", unsafe_allow_html=True)