Delete app.py

app.py

@@ -1,119 +0,0 @@
-import gradio as gr
-from langchain.document_loaders import PyPDFLoader
-from langchain.text_splitter import CharacterTextSplitter
-from langchain.embeddings import SentenceTransformerEmbeddings
-from langchain.vectorstores import FAISS
-from langchain.memory import ConversationBufferMemory
-from groq import Groq
-import requests
-from bs4 import BeautifulSoup
-
-client = Groq(api_key="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")
-embedding_model = SentenceTransformerEmbeddings(model_name="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2")
-
-memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
-
-def process_pdf_with_langchain(pdf_path):
-    """Process the PDF file using LangChain for RAG."""
-    loader = PyPDFLoader(pdf_path)
-    documents = loader.load()
-    text_splitter = CharacterTextSplitter(chunk_size=500, chunk_overlap=50)
-    split_documents = text_splitter.split_documents(documents)
-
-    vectorstore = FAISS.from_documents(split_documents, embedding_model)
-    retriever = vectorstore.as_retriever(search_kwargs={"k": 3})
-    return retriever
-
-def scrape_google_search(query, num_results=3):
-    """Search Google and return the top results."""
-    headers = {"User-Agent": "Mozilla/5.0"}
-    search_url = f"https://www.google.com/search?q={query}"
-    response = requests.get(search_url, headers=headers)
-    soup = BeautifulSoup(response.text, "html.parser")
-
-    results = []
-    for g in soup.find_all('div', class_='tF2Cxc')[:num_results]:
-        title = g.find('h3').text
-        link = g.find('a')['href']
-        results.append(f"{title}: {link}")
-    return "\n".join(results)
-
-def generate_response(query, retriever=None, use_web_search=False):
-    """Generate a response using LangChain with optional retriever and web search."""
-    knowledge = ""
-
-    if retriever:
-        relevant_docs = retriever.get_relevant_documents(query)
-        knowledge += "\n".join([doc.page_content for doc in relevant_docs])
-
-    if use_web_search:
-        web_results = scrape_google_search(query)
-        knowledge += f"\n\nWeb Search Results:\n{web_results}"
-
-    chat_history = memory.load_memory_variables({}).get("chat_history", "")
-    context = (
-        f"This is a conversation with ParvizGPT, an AI model designed by Amir Mahdi Parviz, "
-        f"to help with tasks like answering questions in Persian, providing recommendations, and decision-making."
-    )
-    if knowledge:
-        context += f"\n\nRelevant Knowledge:\n{knowledge}"
-    if chat_history:
-        context += f"\n\nChat History:\n{chat_history}"
-
-    context += f"\n\nYou: {query}\nParvizGPT:"
-
-    chat_completion = client.chat.completions.create(
-        messages=[{"role": "user", "content": context}],
-        model="llama-3.3-70b-versatile",
-    )
-    response = chat_completion.choices[0].message.content.strip()
-
-    memory.save_context({"input": query}, {"output": response})
-    return response
-
-def gradio_interface(user_message, pdf_file=None, enable_web_search=False):
-    global retriever
-    if pdf_file is not None:
-        try:
-            retriever = process_pdf_with_langchain(pdf_file.name)
-        except Exception as e:
-            return f"Error processing PDF: {e}"
-
-    response = generate_response(user_message, retriever=retriever, use_web_search=enable_web_search)
-    return response
-
-def clear_memory():
-    memory.clear()
-    return "Memory cleared!"
-
-retriever = None  
-
-with gr.Blocks() as interface:
-    gr.Markdown("## ParvizGPT with Memory and Web Search Toggle")
-    with gr.Row():
-        user_message = gr.Textbox(label="Your Question", placeholder="Type your question here...", lines=1)
-        submit_btn = gr.Button("Submit")
-    with gr.Row():
-        pdf_file = gr.File(label="Upload PDF for Context (Optional)", type="filepath")
-        enable_web_search = gr.Checkbox(label="Enable Web Search", value=False)
-    with gr.Row():
-        clear_memory_btn = gr.Button("Clear Memory")
-    response_output = gr.Textbox(label="Response", placeholder="ParvizGPT's response will appear here.")
-
-    submit_btn.click(gradio_interface, inputs=[user_message, pdf_file, enable_web_search], outputs=response_output)
-    clear_memory_btn.click(clear_memory, inputs=[], outputs=response_output)
-
-    gr.HTML(
-        """
-        <script>
-        document.addEventListener("keydown", function(event) {
-            if (event.key === "Enter" && !event.shiftKey) {
-                event.preventDefault();
-                document.querySelector('button[title="Submit"]').click();
-            }
-        });
-        </script>
-        """
-    )
-
-interface.launch()