Create app.py

app.py

@@ -0,0 +1,111 @@
+import streamlit as st
+from PyPDF2 import PdfReader
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+import os
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+from langchain_community.vectorstores import Chroma
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain.chains.question_answering import load_qa_chain
+from langchain.prompts import PromptTemplate
+from langchain_community.document_loaders import PyPDFLoader
+from langchain_chroma import Chroma
+import tempfile
+from langchain_cohere import CohereEmbeddings
+
+#st.set_page_config(page_title="Document Genie", layout="wide")
+
+#st.markdown("""
+### PDFChat: Get instant insights from your PDF
+
+#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. **Upload Your Document**: The system accepts a PDF file at one time, analyzing the content to provide comprehensive insights.
+
+#2. **Ask a Question**: After processing the document, ask any question related to the content of your uploaded document for a precise answer.
+#""")
+
+#def get_pdf(pdf_docs):
+#   loader = PyPDFLoader(pdf_docs)
+#    docs = loader.load()
+#    return docs
+
+def get_pdf(uploaded_file):
+    if uploaded_file :
+        temp_file = "./temp.pdf"
+        # Delete the existing temp.pdf file if it exists
+        if os.path.exists(temp_file):
+            os.remove(temp_file)
+        with open(temp_file, "wb") as file:
+            file.write(uploaded_file.getvalue())
+            file_name = uploaded_file.name
+    loader = PyPDFLoader(temp_file)
+    docs = loader.load()
+    return docs
+
+def text_splitter(text):
+    text_splitter = RecursiveCharacterTextSplitter(
+    # Set a really small chunk size, just to show.
+    chunk_size=100000,
+    chunk_overlap=50000,
+    separators=["\n\n","\n"," ",".",","])
+    chunks=text_splitter.split_documents(text)
+    return chunks
+
+GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
+COHERE_API_KEY = os.getenv("COHERE_API_KEY")
+
+def get_conversational_chain():
+    prompt_template = """
+    Given the following extracted parts of a long document and a question, create a final answer.
+    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", and then ignore the context and add the answer from your knowledge like a simple llm prompt.
+    Try to give atleast the basic information.Donot return blank answer.\n\n
+    Make sure to understand the question and answer as per the question.
+    The answer should be a detailed one and try to incorporate examples for better understanding.
+    If the question involves terms like detailed or explained , give answer which involves complete detail about the question.\n\n
+    Context:\n {context}?\n
+    Question: \n{question}\n
+    Answer:
+    """
+    #model = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.3, google_api_key=GOOGLE_API_KEY)
+    model = ChatGoogleGenerativeAI(model="gemini-1.0-pro-latest", temperature=0.3, google_api_key=GOOGLE_API_KEY)
+    prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
+    chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
+    return chain
+
+def embedding(chunk,query):
+    #embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+    embeddings = CohereEmbeddings(model="embed-english-v3.0")
+    db = Chroma.from_documents(chunk,embeddings)
+    doc = db.similarity_search(query)
+    print(doc)
+    chain = get_conversational_chain()
+    response = chain({"input_documents": doc, "question": query}, return_only_outputs=True)
+    print(response)
+    return response["output_text"]
+    #st.write("Reply: ", response["output_text"])
+
+if 'messages' not in st.session_state:
+    st.session_state.messages = [{'role': 'assistant', "content": 'Hello! Upload a PDF and ask me anything about its content.'}]
+
+    
+st.header("Chat with your pdf💁")
+with st.sidebar:
+    st.title("PDF FILE UPLOAD:")
+    pdf_docs = st.file_uploader("Upload your PDF File and Click on the Submit & Process Button", accept_multiple_files=False, key="pdf_uploader")
+    
+query = st.chat_input("Ask a Question from the PDF File")    
+if query:
+    raw_text = get_pdf(pdf_docs)
+    text_chunks = text_splitter(raw_text)
+    st.session_state.messages.append({'role': 'user', "content": query})
+    response = embedding(text_chunks,query)
+    st.session_state.messages.append({'role': 'assistant', "content": response})
+
+for message in st.session_state.messages:
+    with st.chat_message(message['role']):
+        st.write(message['content'])