app.py

import streamlit as st
from transformers import (
    AutoTokenizer,
    AutoModelForCausalLM,
    TextIteratorStreamer,
    AutoConfig
)
from huggingface_hub import login
from threading import Thread
import PyPDF2
import pandas as pd
import torch
import time
import os

# Check if 'peft' is installed
try:
    from peft import PeftModel, PeftConfig
except ImportError:
    raise ImportError(
        "The 'peft' library is required but not installed. "
        "Please install it using: `pip install peft`"
    )

# 🔐 Hugging Face Token via Environment Variable
HF_TOKEN = os.environ.get("HF_TOKEN")
if not HF_TOKEN:
    raise ValueError("Missing Hugging Face Token. Please set the HF_TOKEN environment variable.")

# 🎛 Model base and adapters
BASE_MODEL_NAME = "unicamp-dl/ptt5-base-portuguese-vocab" #"neuralmind/bert-base-portuguese-cased" #"pierreguillou/gpt2-small-portuguese" # #"mistralai/Mistral-7B-Instruct-v0.2"
MODEL_OPTIONS = {
    "Full Fine-Tuned": "amiguel/mistral-angolan-laborlaw-ptt5" #"amiguel/mistral-angolan-laborlaw-bert-base-pt", #"amiguel/mistral-angolan-laborlaw-gpt2",#, #"amiguel/mistral-angolan-laborlaw",
    "LoRA Adapter": "amiguel/SmolLM2-360M-concise-reasoning-lora",
    "QLoRA Adapter": "amiguel/SmolLM2-360M-concise-reasoning-qlora"
}


# 🖼 UI Setup
st.set_page_config(page_title="Assistente LGT | Angola", page_icon="🚀", layout="centered")
st.title("🚀 Assistente LGT | Angola 🚀")

USER_AVATAR = "https://raw.githubusercontent.com/achilela/vila_fofoka_analysis/9904d9a0d445ab0488cf7395cb863cce7621d897/USER_AVATAR.png"
BOT_AVATAR = "https://raw.githubusercontent.com/achilela/vila_fofoka_analysis/991f4c6e4e1dc7a8e24876ca5aae5228bcdb4dba/Ataliba_Avatar.jpg"

# Sidebar
with st.sidebar:
    st.header("Model Selection 🤖")
    model_type = st.selectbox("Choose Model Type", list(MODEL_OPTIONS.keys()), index=0)
    selected_model = MODEL_OPTIONS[model_type]

    st.header("Upload Documents 📂")
    uploaded_file = st.file_uploader("Choose a PDF or XLSX file", type=["pdf", "xlsx"], label_visibility="collapsed")

# Chat memory
if "messages" not in st.session_state:
    st.session_state.messages = []

# 🔍 File processing
@st.cache_data
def process_file(uploaded_file):
    if uploaded_file is None:
        return ""
    try:
        if uploaded_file.type == "application/pdf":
            reader = PyPDF2.PdfReader(uploaded_file)
            return "\n".join(page.extract_text() or "" for page in reader.pages)
        elif uploaded_file.type == "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet":
            df = pd.read_excel(uploaded_file)
            return df.to_markdown()
    except Exception as e:
        st.error(f"📄 Error processing file: {str(e)}")
        return ""

# 🧠 Load model and tokenizer
@st.cache_resource
def load_model(model_type, selected_model):
    try:
        login(token=HF_TOKEN)
        device = "cuda" if torch.cuda.is_available() else "cpu"
        dtype = torch.bfloat16 if torch.cuda.is_bf16_supported() else torch.float32

        tokenizer = AutoTokenizer.from_pretrained(selected_model, token=HF_TOKEN)

        if model_type == "Full Fine-Tuned":
            model = AutoModelForCausalLM.from_pretrained(
                selected_model,
                device_map="auto",
                torch_dtype=dtype,
                token=HF_TOKEN
            )
        else:
            base_model = AutoModelForCausalLM.from_pretrained(
                BASE_MODEL_NAME,
                device_map="auto",
                torch_dtype=dtype,
                token=HF_TOKEN
            )
            model = PeftModel.from_pretrained(
                base_model,
                selected_model,
                is_trainable=False,
                torch_dtype=dtype,
                token=HF_TOKEN
            )
        return model, tokenizer
    except Exception as e:
        st.error(f"🤖 Model loading failed: {str(e)}")
        return None, None

# 🚀 Generate response
def generate_with_streaming(prompt, file_context, model, tokenizer):
    full_prompt = f"Analisa este contexto:\n{file_context}\n\nPergunta: {prompt}\nResposta:"

    inputs = tokenizer(full_prompt, return_tensors="pt")
    inputs = {k: v.to(model.device) for k, v in inputs.items()}

    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
    gen_kwargs = {
        "input_ids": inputs["input_ids"],
        "attention_mask": inputs["attention_mask"],
        "max_new_tokens": 1024,
        "temperature": 0.7,
        "top_p": 0.9,
        "repetition_penalty": 1.1,
        "do_sample": True,
        "use_cache": True,
        "streamer": streamer
    }

    Thread(target=model.generate, kwargs=gen_kwargs).start()
    return streamer

# 🧾 Display chat history
for msg in st.session_state.messages:
    avatar = USER_AVATAR if msg["role"] == "user" else BOT_AVATAR
    with st.chat_message(msg["role"], avatar=avatar):
        st.markdown(msg["content"])

# 🔎 Main interaction loop
if prompt := st.chat_input("Pergunta sobre a LGT?"):
    # Display user message
    with st.chat_message("user", avatar=USER_AVATAR):
        st.markdown(prompt)
    st.session_state.messages.append({"role": "user", "content": prompt})

    # Load model if needed
    if "model" not in st.session_state or st.session_state.get("model_type") != model_type:
        with st.spinner("🔄 A carregar modelo..."):
            model, tokenizer = load_model(model_type, selected_model)
            if not model:
                st.stop()
            st.session_state.model = model
            st.session_state.tokenizer = tokenizer
            st.session_state.model_type = model_type
    else:
        model = st.session_state.model
        tokenizer = st.session_state.tokenizer

    # Prepare context
    file_context = process_file(uploaded_file) or "Sem contexto adicional disponível."

    # Generate assistant response
    with st.chat_message("assistant", avatar=BOT_AVATAR):
        response_box = st.empty()
        full_response = ""
        try:
            start_time = time.time()
            streamer = generate_with_streaming(prompt, file_context, model, tokenizer)

            for chunk in streamer:
                full_response += chunk.strip() + " "
                response_box.markdown(full_response + "▌", unsafe_allow_html=True)

            # Token and speed metrics
            end_time = time.time()
            input_tokens = len(tokenizer(prompt)["input_ids"])
            output_tokens = len(tokenizer(full_response)["input_ids"])
            speed = output_tokens / (end_time - start_time)
            cost_usd = ((input_tokens / 1e6) * 5) + ((output_tokens / 1e6) * 15)
            cost_aoa = cost_usd * 1160

            st.caption(
                f"🔑 Input Tokens: {input_tokens} | Output Tokens: {output_tokens} | "
                f"🕒 Speed: {speed:.1f}t/s | 💰 USD: ${cost_usd:.4f} | 🇦🇴 AOA: {cost_aoa:.2f}"
            )

            response_box.markdown(full_response.strip())
            st.session_state.messages.append({"role": "assistant", "content": full_response.strip()})

        except Exception as e:
            st.error(f"⚡ Erro ao gerar resposta: {str(e)}")