src/streamlit_app.py

import os 
import streamlit as st
from langchain.prompts import ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate
from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace

os.environ['HUGGINGFACEHUB_API_TOKEN'] = os.getenv("keys")
os.environ['HF_TOKEN'] = os.getenv("keys")

st.set_page_config(page_title="AI-Powered Mentor Assistant", page_icon="🧠")

st.title("🧠AI-Powered Mentor Assistant")




col1, col2, col3, col4, col5, col6 = st.columns(6)

import streamlit as st

st.markdown("### 🎓 **Select Your Mentor:**")

col1, col2, col3, col4, col5, col6 = st.columns(6)

with col1:
    st.markdown("#### 🐍<br><strong>Python</strong>", unsafe_allow_html=True)  # Snake for Python

with col2:
    st.markdown("#### 🤖<br><strong>Machine Learning</strong>", unsafe_allow_html=True)  # Robot for ML

with col3:
    st.markdown("#### 🧠<br><strong>Deep Learning</strong>", unsafe_allow_html=True)  # Brain for DL

with col4:
    st.markdown("#### 📊<br><strong>Statistics</strong>", unsafe_allow_html=True)

with col5:
    st.markdown("#### 📈<br><strong>Data Analysis</strong>", unsafe_allow_html=True)

with col6:
    st.markdown("#### 🗄️<br><strong>SQL & Power BI</strong>", unsafe_allow_html=True)

st.markdown("""
    <h4 style='color:#4CAF50; font-family:sans-serif;'>
        🚀 <strong>Select Your Mentor Domain:</strong>
    </h4>
""", unsafe_allow_html=True)

# Dropdown
mentor_type = st.selectbox(
    "", 
    ["", "🐍 Python", "🤖 Machine Learning", "🧠 Deep Learning", "📊 Statistics", "📈 Data Analysis", "🗄️ SQL & Power BI"]
)
if mentor_type:
    st.subheader(f"🧠 {mentor_type.upper()} Mentor Chat")
    experience = st.slider("Your experience (in years):", 0, 20, 1)
    user_input = st.text_input("Ask your question:")


    output_container = st.empty()

    if mentor_type == "python":
        model = HuggingFaceEndpoint(repo_id="meta-llama/Llama-3.1-8B-Instruct", provider="nebius", temperature=0.5, max_new_tokens=150, task="conversational")
    elif mentor_type == "machine_learning":
        model = HuggingFaceEndpoint(repo_id="deepseek-ai/DeepSeek-R1", provider="nebius", temperature=0.5, max_new_tokens=150, task="conversational")
    elif mentor_type == "deep_learning":
        model = HuggingFaceEndpoint(repo_id="deepseek-ai/DeepSeek-R1", provider="nebius", temperature=0.5, max_new_tokens=150, task="conversational")
    elif mentor_type == "stats":
        model = HuggingFaceEndpoint(repo_id="meta-llama/Llama-3.2-1B-Instruct", provider="nebius", temperature=0.5, max_new_tokens=150, task="conversational")
    elif mentor_type == "data_anaylasis":
        model = HuggingFaceEndpoint(repo_id="meta-llama/Llama-3.3-70B-Instruct", provider="nebius", temperature=0.5, max_new_tokens=150, task="conversational")
    elif mentor_type == "sql and powerbi":
        model = HuggingFaceEndpoint(repo_id="meta-llama/Meta-Llama-3-70B-Instruct", provider="hyperbolic", temperature=0.5, max_new_tokens=150, task="conversational")
    model = HuggingFaceEndpoint(repo_id="meta-llama/Meta-Llama-3-70B-Instruct", provider="hyperbolic", temperature=0.5, max_new_tokens=150, task="conversational")
    chat_model = ChatHuggingFace(
        llm=model,
        repo_id=model.repo_id,
        provider="nebius",
        temperature=0.5,
        max_new_tokens=150,
        task="conversational"
    )


    if st.button("Ask") and user_input:
        prompt = ChatPromptTemplate.from_messages([
            SystemMessagePromptTemplate.from_template(
                f"You are a helpful and experienced {mentor_type.upper()} mentor assisting a learner with {experience} years of experience."
            ),
            HumanMessagePromptTemplate.from_template("{question}")
        ])
        formatted_prompt = prompt.format_messages(question=user_input)

        with st.spinner("Mentor is thinking..."):
            response = chat_model.invoke(formatted_prompt)

        output_container.markdown(f"👤 You: {user_input}")
        output_container.markdown(f"🧠 Mentor: {response.content}")

    if st.button("Clear Output"):
        output_container.empty()