app.py

import streamlit as st
import os
from dotenv import load_dotenv
from transformers import (
    pipeline,
    AutoConfig,
    AutoTokenizer,
    AutoModelForCausalLM
)

# Load environment variables from .env (if you’re using one)
load_dotenv()

st.set_page_config(page_title="Educational Chatbot")
st.title("🎓 Educational Chatbot")

@st.cache_resource(show_spinner=False)
def load_model():
    # 1. Load the remote config (with trust_remote_code)
    config = AutoConfig.from_pretrained(
        "deepseek-ai/DeepSeek-R1",
        trust_remote_code=True
    )
    # 2. Remove unsupported fp8 quantization
    if hasattr(config, "quantization_config"):
        config.quantization_config = None

    # 3. Load tokenizer and model with patched config
    tokenizer = AutoTokenizer.from_pretrained(
        "deepseek-ai/DeepSeek-R1",
        trust_remote_code=True
    )
    model = AutoModelForCausalLM.from_pretrained(
        "deepseek-ai/DeepSeek-R1",
        trust_remote_code=True,
        config=config
    )

    # 4. Build the text-generation pipeline
    gen = pipeline(
        "text-generation",
        model=model,
        tokenizer=tokenizer,
        trust_remote_code=True,
        device_map="auto"        # or remove for CPU-only
    )
    return gen

# Load the model once
generator = load_model()

# Initialize chat history
if "history" not in st.session_state:
    st.session_state.history = []

# User input box
user_input = st.text_input("Ask me anything:")

# When user enters a question
if user_input:
    try:
        outputs = generator(user_input, return_full_text=False)
        reply = outputs[0]["generated_text"].strip()
        st.session_state.history.append(("You", user_input))
        st.session_state.history.append(("Bot", reply))
    except Exception as e:
        st.session_state.history.append(("Bot", f"⚠️ Error: {e}"))

# Display chat history
for sender, msg in reversed(st.session_state.history):
    st.markdown(f"**{sender}:** {msg}")