app.py

import json
import requests
import os
import streamlit as st
import pandas as pd
import openpyxl
import torch
import faiss
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from sentence_transformers import SentenceTransformer

# ✅ Streamlit UI Setup
st.set_page_config(page_title="AI-Powered Timetable", layout="wide")
st.markdown("<h1 style='text-align: center; color: #4CAF50;'>📅 AI-Powered Timetable with RAG</h1>", unsafe_allow_html=True)

# ✅ API Key Input
st.sidebar.markdown("## 🔑 Enter Hugging Face API Key")
hf_api_key = st.sidebar.text_input("API Key", type="password")

# ✅ File Upload Section
st.sidebar.markdown("## 📂 Upload Your Timetable Files")
uploaded_master = st.sidebar.file_uploader("Upload Master Timetable", type=["xlsx", "pdf"])
uploaded_lab = st.sidebar.file_uploader("Upload Lab Timetable", type=["xlsx", "pdf"])
uploaded_classroom = st.sidebar.file_uploader("Upload Classroom Timetable", type=["xlsx", "pdf"])
uploaded_individual = st.sidebar.file_uploader("Upload Individual Timetable", type=["xlsx", "pdf"])

uploaded_files = {
    "Master Timetable": uploaded_master,
    "Lab Timetable": uploaded_lab,
    "Classroom Timetable": uploaded_classroom,
    "Individual Timetable": uploaded_individual,
}

# ✅ Load Timetable Data
def load_timetable(file):
    if not file:
        return None
    if file.name.endswith('.xlsx'):
        wb = openpyxl.load_workbook(file)
        sheet = wb.active
        return [row for row in sheet.iter_rows(values_only=True)]
    elif file.name.endswith('.pdf'):
        import PyPDF2
        pdf_reader = PyPDF2.PdfReader(file)
        text = ""
        for page in pdf_reader.pages:
            text += page.extract_text() + "\n"
        return text

# ✅ Extract and Store Data
rag_data = {}
for name, file in uploaded_files.items():
    if file:
        rag_data[name] = load_timetable(file)

# ✅ Encode Timetable Data with Sentence Embeddings
embedder = SentenceTransformer("all-MiniLM-L6-v2")
data_texts = ["\n".join(map(str, data)) for data in rag_data.values() if data]
data_embeddings = embedder.encode(data_texts, convert_to_tensor=True)

# ✅ Create FAISS Vector Store
dimension = data_embeddings.shape[1]
index = faiss.IndexFlatL2(dimension)
index.add(data_embeddings.cpu().numpy())

# ✅ Retrieve Relevant Data using RAG
def retrieve_data(query):
    query_embedding = embedder.encode([query], convert_to_tensor=True)
    D, I = index.search(query_embedding.cpu().numpy(), k=1)
    return data_texts[I[0][0]] if I[0][0] < len(data_texts) else "No relevant data found."

# ✅ Ask LLaMA-3-8B with RAG Context
def ask_llama_api(query):
    if not hf_api_key:
        return "Error: Please enter your API key."
    
    context = retrieve_data(query)  # Get relevant timetable data
    final_query = f"Based on the timetable data:\n{context}\nAnswer this query: {query}"

    url = "https://api-inference.huggingface.co/v1/chat/completions"
    headers = {
        "Authorization": f"Bearer {hf_api_key}",
        "Content-Type": "application/json"
    }
    payload = {
        "model": "meta-llama/Meta-Llama-3-8B",
        "messages": [{"role": "user", "content": final_query}],
        "max_tokens": 500
    }

    response = requests.post(url, headers=headers, json=payload)
    if response.status_code == 200:
        return response.json()["choices"][0]["message"]["content"]
    else:
        return f"API Error: {response.status_code} - {response.text}"

# ✅ AI Query Section
st.markdown("## 🤖 Ask LLaMA-3 AI About Your Timetable")
user_query = st.text_input("Type your question here (e.g., 'Who is free at 10 AM on Monday?')")

if st.button("Ask AI via RAG"):
    ai_response = ask_llama_api(user_query)
    st.write("🧠 **LLaMA-3 AI Suggests:**", ai_response)

# ✅ Auto-Schedule Feature
st.markdown("## 📅 Auto-Schedule Missing Timetable Slots")
selected_file = st.selectbox("Choose a timetable file to auto-fill missing slots:", list(uploaded_files.keys()))

if st.button("Auto-Schedule"):
    result = retrieve_data(selected_file)
    st.write("✅", result)

# ✅ Display Uploaded Timetables
st.markdown("## 📜 View Uploaded Timetables")

for name, file in uploaded_files.items():
    if file:
        df = pd.read_excel(file) if file.name.endswith(".xlsx") else None
        st.markdown(f"### {name}")
        if df is not None:
            st.dataframe(df)
        else:
            st.text(rag_data[name])

# ✅ PDF Export Feature
st.sidebar.markdown("## 📄 Export AI Responses to PDF")
if st.sidebar.button("Export as PDF"):
    c = canvas.Canvas("Timetable_Responses.pdf", pagesize=letter)
    c.drawString(100, 750, "AI-Powered Timetable Responses")
    y = 720
    for query, response in [("Example Query", "Example Response")]:  # Replace with actual
        c.drawString(50, y, f"Q: {query}")
        y -= 20
        c.drawString(70, y, f"A: {response}")
        y -= 30
    c.save()
    st.sidebar.success("PDF Exported: Timetable_Responses.pdf")