app.py

import gradio as gr
from huggingface_hub import InferenceClient
from langchain.vectorstores import FAISS
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.text_splitter import CharacterTextSplitter
from langchain.document_loaders import TextLoader

client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

# Funktion zum Laden und Indexieren eines Dokuments
def load_and_index_document(file_path: str):
    loader = TextLoader(file_path)
    documents = loader.load()
    text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
    chunks = text_splitter.split_documents(documents)
    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
    vector_store = FAISS.from_documents(chunks, embeddings)
    return vector_store

# Antwortfunktion für den RAG-Chatbot
def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, top_p, file):
    # Dateipfad des hochgeladenen Dokuments
    file_path = file.name
    
    # Dokument laden und indexieren
    vector_store = load_and_index_document(file_path)
    
    # Historie und Systemnachricht aufbereiten
    messages = [{"role": "system", "content": system_message}]
    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})
    
    # Abruf relevanter Abschnitte aus dem Dokument
    docs = vector_store.similarity_search(message, k=3)  # Abrufen von 3 relevanten Dokumentabschnitten
    context = "\n".join([doc.page_content for doc in docs])

    # Nachricht an das Modell
    full_message = f"{context}\n\nUser: {message}\nAssistant:"
    
    response = ""
    try:
        # Generierung der Antwort
        for message in client.chat_completion(
            [{"role": "system", "content": system_message}, {"role": "user", "content": full_message}],
            max_tokens=max_tokens,
            stream=True,
            temperature=temperature,
            top_p=top_p,
        ):
            token = message.choices[0].delta.content
            response += token
            yield response
    except Exception as e:
        yield f"An error occurred: {str(e)}"

# Gradio-UI erstellen
def create_gradio_ui():
    demo = gr.Interface(
        fn=respond,
        inputs=[
            gr.Textbox(value="You are a helpful assistant.", label="System message"),
            gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
            gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
            gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
            gr.File(label="Upload Document")  # Datei-Upload
        ],
        live=True
    )
    return demo

if __name__ == "__main__":
    ui = create_gradio_ui()
    ui.launch()