app.py

import gradio as gr
from gradio_pdf import PDF
from qdrant_client import models, QdrantClient
from sentence_transformers import SentenceTransformer
from PyPDF2 import PdfReader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.callbacks.manager import CallbackManager
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from ctransformers import AutoModelForCausalLM

# Load the embedding model
encoder = SentenceTransformer('jinaai/jina-embedding-b-en-v1')
print("Embedding model loaded...")

# Load the LLM
callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
llm = AutoModelForCausalLM.from_pretrained(
    "TheBloke/Llama-2-7B-Chat-GGUF",
    model_file="llama-2-7b-chat.Q3_K_S.gguf",
    model_type="llama",
    temperature=0.2,
    repetition_penalty=1.5,
    max_new_tokens=300,
)
print("LLM loaded...")

def get_chunks(text):
    text_splitter = RecursiveCharacterTextSplitter(
        chunk_size=250,
        chunk_overlap=50,
        length_function=len,
    )
    return text_splitter.split_text(text)

def setup_database(files):
    all_chunks = []
    for file in files:
        reader = PdfReader(file)
        text = "".join(page.extract_text() for page in reader.pages)
        chunks = get_chunks(text)
        all_chunks.extend(chunks)

    client = QdrantClient(path="./db")
    client.recreate_collection(
        collection_name="my_facts",
        vectors_config=models.VectorParams(
            size=encoder.get_sentence_embedding_dimension(),
            distance=models.Distance.COSINE,
        ),
    )

    records = [
        models.Record(
            id=idx,
            vector=encoder.encode(chunk).tolist(),
            payload={f"chunk_{idx}": chunk}
        ) for idx, chunk in enumerate(all_chunks)
    ]
    
    client.upload_records(
        collection_name="my_facts",
        records=records,
    )

def answer_question(question):
    client = QdrantClient(path="./db")
    hits = client.search(
        collection_name="my_facts",
        query_vector=encoder.encode(question).tolist(),
        limit=3
    )
    
    context = " ".join(hit.payload[f"chunk_{hit.id}"] for hit in hits)

    system_prompt = """You are a helpful co-worker, you will use the provided context to answer user questions.
    Read the given context before answering questions and think step by step. If you cannot answer a user question based on 
    the provided context, inform the user. Do not use any other information for answering user. Provide a detailed answer to the question."""

    B_INST, E_INST = "[INST]", "[/INST]"
    B_SYS, E_SYS = "<<SYS>>\n", "\n<</SYS>>\n\n"

    instruction = f"Context: {context}\nUser: {question}"
    prompt_template = f"{B_INST}{B_SYS}{system_prompt}{E_SYS}{instruction}{E_INST}"
    
    response = llm(prompt_template)
    return response

def chat(messages, files):
    if files:
        setup_database(files)
    if messages:
        question = messages[-1]["text"]
        answer = answer_question(question)
        messages.append({"text": answer, "is_user": False})
    return messages

interface = gr.Interface(
    fn=chat,
    inputs=[
        gr.Chatbot(label="Chat"),
        gr.File(label="Upload PDFs", file_count="multiple")
    ],
    outputs=gr.Chatbot(label="Chat"),
    title="Q&A with PDFs 👩🏻‍💻📓✍🏻💡",
    description="This app facilitates a conversation with PDFs uploaded💡",
    theme="soft",
    live=True,
)

interface.launch()