Update app.py

app.py

@@ -1,94 +1,94 @@
-import streamlit as st
-from PyPDF2 import PdfReader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain_google_genai import GoogleGenerativeAIEmbeddings
-import google.generativeai as genai
-from langchain.vectorstores import FAISS
-from langchain_google_genai import ChatGoogleGenerativeAI
-from langchain.chains.question_answering import load_qa_chain
-from langchain.prompts import PromptTemplate
-import os
-
-st.set_page_config(page_title="Document Genie", layout="wide")
-
-st.markdown("""
-## Document Genie: Get instant insights from your Documents
-
-This chatbot is built using the Retrieval-Augmented Generation (RAG) framework, leveraging Google's Generative AI model Gemini-PRO. It processes uploaded PDF documents by breaking them down into manageable chunks, creates a searchable vector store, and generates accurate answers to user queries. This advanced approach ensures high-quality, contextually relevant responses for an efficient and effective user experience.
-
-### How It Works
-
-Follow these simple steps to interact with the chatbot:
-
-1. **Enter Your API Key**: You'll need a Google API key for the chatbot to access Google's Generative AI models. Obtain your API key https://makersuite.google.com/app/apikey.
-
-2. **Upload Your Documents**: The system accepts multiple PDF files at once, analyzing the content to provide comprehensive insights.
-
-3. **Ask a Question**: After processing the documents, ask any question related to the content of your uploaded documents for a precise answer.
-""")
-
-
-
-# This is the first API key input; no need to repeat it in the main function.
-api_key = st.text_input("Enter your Google API Key:", type="password", key="api_key_input")
-
-def get_pdf_text(pdf_docs):
-    text = ""
-    for pdf in pdf_docs:
-        pdf_reader = PdfReader(pdf)
-        for page in pdf_reader.pages:
-            text += page.extract_text()
-    return text
-
-def get_text_chunks(text):
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=10000, chunk_overlap=1000)
-    chunks = text_splitter.split_text(text)
-    return chunks
-
-def get_vector_store(text_chunks, api_key):
-    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
-    vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
-    vector_store.save_local("faiss_index")
-
-def get_conversational_chain():
-    prompt_template = """
-    Answer the question as detailed as possible from the provided context, make sure to provide all the details, if the answer is not in
-    provided context just say, "answer is not available in the context", don't provide the wrong answer\n\n
-    Context:\n {context}?\n
-    Question: \n{question}\n
-
-    Answer:
-    """
-    model = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.3, google_api_key=api_key)
-    prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
-    chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
-    return chain
-
-def user_input(user_question, api_key):
-    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
-    new_db = FAISS.load_local("faiss_index", embeddings)
-    docs = new_db.similarity_search(user_question)
-    chain = get_conversational_chain()
-    response = chain({"input_documents": docs, "question": user_question}, return_only_outputs=True)
-    st.write("Reply: ", response["output_text"])
-
-def main():
-    st.header("AI clone chatbot💁")
-
-    user_question = st.text_input("Ask a Question from the PDF Files", key="user_question")
-
-    if user_question and api_key:  # Ensure API key and user question are provided
-        user_input(user_question, api_key)
-
-    with st.sidebar:
-        st.title("Menu:")
-        pdf_docs = st.file_uploader("Upload your PDF Files and Click on the Submit & Process Button", accept_multiple_files=True, key="pdf_uploader")
-        if st.button("Submit & Process", key="process_button") and api_key:  # Check if API key is provided before processing
-            with st.spinner("Processing..."):
-                raw_text = get_pdf_text(pdf_docs)
-                text_chunks = get_text_chunks(raw_text)
-                get_vector_store(text_chunks, api_key)
-                st.success("Done")
-
-if __name__ == "__main__":
-    main()
+import streamlit as st
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+from PyPDF2 import PdfReader
+import os
+from sentence_transformers import SentenceTransformer
+from sklearn.metrics.pairwise import cosine_similarity
+import numpy as np
+
+# Load environment variables
+HUGGINGFACE_API_KEY = os.environ.get('HUGGINGFACE_API_KEY')
+
+# Initialize Sentence Transformer for semantic search
+sentence_model = SentenceTransformer('all-MiniLM-L6-v2')
+
+# Load the Mistral model and tokenizer
+model_name = "mistralai/Mistral-7B-Instruct-v0.2"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
+
+# Function to extract text from PDF
+def extract_text_from_pdf(pdf_path):
+    with open(pdf_path, 'rb') as file:
+        pdf_reader = PdfReader(file)
+        text = ""
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+    return text
+
+# Function to split text into chunks
+def split_text(text, chunk_size=1000):
+    return [text[i:i+chunk_size] for i in range(0, len(text), chunk_size)]
+
+# Load and process the Levo.ai PDF
+pdf_path = "levo_ai_documentation.pdf"  # Replace with the actual path to your PDF
+levo_text = extract_text_from_pdf(pdf_path)
+levo_chunks = split_text(levo_text)
+
+# Encode chunks for semantic search
+chunk_embeddings = sentence_model.encode(levo_chunks)
+
+# Function to find most relevant chunks
+def find_relevant_chunks(query, top_k=3):
+    query_embedding = sentence_model.encode([query])
+    similarities = cosine_similarity(query_embedding, chunk_embeddings)[0]
+    top_indices = np.argsort(similarities)[-top_k:][::-1]
+    return [levo_chunks[i] for i in top_indices]
+
+# Function to generate response
+def generate_response(prompt, context, max_length=1000):
+    full_prompt = f"""You are an advanced AI assistant for Levo.ai, a cutting-edge API security product. Your role is to provide accurate, helpful, and friendly support to users based on the following context from Levo.ai's documentation. Always maintain a professional and supportive tone.
+
+Context:
+{context}
+
+User Query: {prompt}
+Response:"""
+    inputs = tokenizer.encode(full_prompt, return_tensors="pt").to(model.device)
+    
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs, 
+            max_length=max_length, 
+            num_return_sequences=1, 
+            temperature=0.7,
+            top_p=0.95,
+            do_sample=True
+        )
+    
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response
+
+# Streamlit UI
+st.title("Levo.ai Support Chatbot")
+
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+
+for message in st.session_state.messages:
+    with st.chat_message(message["role"]):
+        st.markdown(message["content"])
+
+if prompt := st.chat_input("How can I assist you with Levo.ai today?"):
+    st.session_state.messages.append({"role": "user", "content": prompt})
+    with st.chat_message("user"):
+        st.markdown(prompt)
+
+    relevant_chunks = find_relevant_chunks(prompt)
+    context = "\n".join(relevant_chunks)
+
+    response = generate_response(prompt, context)
+    st.session_state.messages.append({"role": "assistant", "content": response})
+    with st.chat_message("assistant"):
+        st.markdown(response)