Project Submission , Team - TechDuo

.env

@@ -0,0 +1 @@
+GROQ_API_KEY = "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"

.gitignore

@@ -0,0 +1 @@
+myenv/

README.md

@@ -8,17 +8,17 @@ Submission Details
 
 Submit the following information via the provided Google Form for us to access your AI applications:
 
-Team Name: Your team's unique name.
+Team Name: TechDuo
 
-App Name: The name of your AI application.
+App Name: Study Sphere
 
-App Description: A brief overview of what your app does and its core features.
+App Description: STUDY-SPHERE is an AI-driven learning platform offering personalized domain-specific resources, interactive tools, and real-time support for students and professionals to enhance their knowledge and connect with like-minded communities.
 
 GitHub PR Link: The link to your Pull Request (PR) in the project repository.
 
 App Link: A link to the deployed version of your app.
 
-Loom Video: A demo video link showing your app in action.
+Loom Video: [A demo video link showing your app in action.](https://drive.google.com/drive/folders/1i3vVhHjqSb8PRuJB-pnu-UOewQNn6JN5?usp=sharing)
 
 
 For detailed submission guidelines, please refer to the last doubt session recording: https://www.buildfastwithai.com/hackathon.

a.py

@@ -0,0 +1,153 @@
+import streamlit as st
+from streamlit_chat import message
+from langchain_groq import ChatGroq
+from langchain.chains import ConversationalRetrievalChain
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.llms import CTransformers
+# from langchain.llms import Replicate
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.vectorstores import FAISS
+from langchain.memory import ConversationBufferMemory
+from langchain.document_loaders import PyPDFLoader
+from langchain.prompts import PromptTemplate
+
+from langchain.document_loaders import TextLoader
+from langchain.document_loaders import Docx2txtLoader
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
+import os
+from dotenv import load_dotenv
+import tempfile
+
+
+load_dotenv()
+
+# Create a prompt template
+template = """
+You are an assistant specialized in analyzing documents. The user will provide questions based on a document.
+Please answer the following question using the context provided:
+
+Context: {context}
+
+Question: {question}
+
+Answer in a concise and informative manner.
+Try to answer in bullet points
+"""
+
+prompt = PromptTemplate(input_variables=["context", "question"], template=template)
+
+def initialize_session_state():
+    if 'history' not in st.session_state:
+        st.session_state['history'] = []
+
+    if 'generated' not in st.session_state:
+        st.session_state['generated'] = ["Hello! Ask me anything about 🤗"]
+
+    if 'past' not in st.session_state:
+        st.session_state['past'] = ["Hey! 👋"]
+
+def conversation_chat(query, chain, history):
+    # Retrieve relevant context from the retriever first
+    context = chain.retriever.get_relevant_documents(query)
+
+    # Format the input with the context using the prompt template
+    formatted_prompt = prompt.format(context=context, question=query)
+
+    # Now pass the formatted prompt to the LLM within the chain
+    result = chain({"question": formatted_prompt, "chat_history": history})
+
+    # Append to the conversation history
+    history.append((query, result["answer"]))
+
+    return result["answer"]
+
+
+def display_chat_history(chain):
+    reply_container = st.container()
+    container = st.container()
+
+    with container:
+        with st.form(key='my_form', clear_on_submit=True):
+            user_input = st.text_input("Question:", placeholder="Ask about your Documents", key='input')
+            submit_button = st.form_submit_button(label='Send')
+
+        if submit_button and user_input:
+            with st.spinner('Generating response...'):
+                output = conversation_chat(user_input, chain, st.session_state['history'])
+
+            st.session_state['past'].append(user_input)
+            st.session_state['generated'].append(output)
+
+    if st.session_state['generated']:
+        with reply_container:
+            for i in range(len(st.session_state['generated'])):
+                message(st.session_state["past"][i], is_user=True, key=str(i) + '_user', avatar_style="thumbs")
+                message(st.session_state["generated"][i], key=str(i), avatar_style="fun-emoji")
+
+def create_conversational_chain(vector_store):
+    load_dotenv()
+    groq_api_key = os.getenv('GROQ_API_KEY')
+    model = 'mixtral-8x7b-32768'  # Or any other model you prefer
+
+    # Initialize Groq Langchain chat object
+    llm = ChatGroq(groq_api_key=groq_api_key, model_name=model)
+    memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
+
+    # Create the chain without passing the prompt directly
+    chain = ConversationalRetrievalChain.from_llm(
+        llm=llm, 
+        chain_type='stuff',
+        retriever=vector_store.as_retriever(search_kwargs={"k": 2}),
+        memory=memory
+    )
+    return chain
+
+
+def main():
+    load_dotenv()
+    # Initialize session state
+    initialize_session_state()
+    st.title("Multi-Docs ChatBot using llama2 :books:")
+    # Initialize Streamlit
+    st.sidebar.title("Document Processing")
+    uploaded_files = st.sidebar.file_uploader("Upload files", accept_multiple_files=True)
+
+
+    if uploaded_files:
+        text = []
+        for file in uploaded_files:
+            file_extension = os.path.splitext(file.name)[1]
+            with tempfile.NamedTemporaryFile(delete=False) as temp_file:
+                temp_file.write(file.read())
+                temp_file_path = temp_file.name
+
+            loader = None
+            if file_extension == ".pdf":
+                loader = PyPDFLoader(temp_file_path)
+            elif file_extension == ".docx" or file_extension == ".doc":
+                loader = Docx2txtLoader(temp_file_path)
+            elif file_extension == ".txt":
+                loader = TextLoader(temp_file_path)
+
+            if loader:
+                text.extend(loader.load())
+                os.remove(temp_file_path)
+
+        text_splitter = CharacterTextSplitter(separator="\n", chunk_size=1000, chunk_overlap=100, length_function=len)
+        text_chunks = text_splitter.split_documents(text)
+
+        # Create embeddings
+        embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2", 
+                                           model_kwargs={'device': 'cpu'})
+
+        # Create vector store
+        vector_store = FAISS.from_documents(text_chunks, embedding=embeddings)
+
+        # Create the chain object
+        chain = create_conversational_chain(vector_store)
+
+
+        display_chat_history(chain)
+
+if __name__ == "__main__":
+    main()

appAIML.py

@@ -0,0 +1,95 @@
+from flask import Flask, request, jsonify
+from flask_cors import CORS  # Import CORS
+from langchain.vectorstores import Qdrant
+from langchain.embeddings import HuggingFaceBgeEmbeddings
+from langchain_groq import ChatGroq
+from langchain.chains.conversation.memory import ConversationBufferWindowMemory
+from langchain.prompts import PromptTemplate
+from langchain.chains import ConversationChain
+from qdrant_client import QdrantClient
+from dotenv import load_dotenv
+import os
+
+# Load environment variables
+load_dotenv()
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Enable CORS for all routes
+CORS(app)  # This will allow all origins. You can customize it if needed.
+
+# Initialize Qdrant and embeddings
+model_name = "BAAI/bge-large-en"
+embeddings = HuggingFaceBgeEmbeddings(
+    model_name=model_name,
+    model_kwargs={'device': 'cpu'},
+    encode_kwargs={'normalize_embeddings': False}
+)
+url = "http://localhost:6333"
+client = QdrantClient(url=url, prefer_grpc=False)
+
+memory = ConversationBufferWindowMemory(k=5)
+collection_name = "AIML"
+db = Qdrant(client=client, embeddings=embeddings, collection_name=collection_name)
+
+# Initialize Groq LLM
+groq_llm = ChatGroq(
+    groq_api_key=os.getenv('GROQ_API_KEY'),
+    model='mixtral-8x7b-32768'
+)
+
+# Define the prompt template
+prompt_template = PromptTemplate(
+    input_variables=["query", "retrieved_documents"],
+    template=( 
+        "You are an expert in Artificial Intelligence and Machine Learning (AIML). "
+        "Your task is to assist users with their queries related to AIML concepts, applications, and recent advancements. "
+        "You have access to relevant documents that provide detailed information about various AIML topics, algorithms, and case studies.\n\n"
+        "User Query: {query}\n\n"
+        "{retrieved_documents}\n\n"
+        "Your Response:\n"
+        "- Provide a concise and informative answer based on the user's query and the retrieved documents.\n"
+        "- Include key definitions, explanations of relevant concepts, and any important points from the retrieved documents.\n"
+        "- Ensure that the response is accessible to users with varying levels of knowledge about AIML."
+    )
+)
+
+# API endpoint to handle query
+@app.route('/aiml_query', methods=['POST'])
+def handle_query():
+    # Get the query from the POST request
+    data = request.json
+    query = data.get('query', '')
+
+    if not query:
+        return jsonify({"error": "No query provided"}), 400
+
+    # Perform similarity search with context
+    docs = db.similarity_search_with_score(query=query, k=5)
+
+    # Prepare retrieved documents content
+    retrieved_content = ""
+    for i, (doc, score) in enumerate(docs):
+        retrieved_content += f"**Result {i + 1}:**\n"
+        retrieved_content += f"**Score:** {score}\n"
+        retrieved_content += f"**Content:** {doc.page_content}\n"
+        retrieved_content += f"**Metadata:** {doc.metadata}\n"
+        retrieved_content += "-" * 40 + "\n"
+
+    # Format the input using the prompt template
+    formatted_input = prompt_template.format(query=query, retrieved_documents=retrieved_content)
+
+    # Initialize Groq Langchain chat object and conversation
+    groq_chat = ChatGroq(groq_api_key=os.getenv('GROQ_API_KEY'), model_name='mixtral-8x7b-32768')
+    conversation = ConversationChain(llm=groq_chat, memory=memory)
+
+    # Generate response using the formatted input
+    response = conversation.run(formatted_input)
+
+    # Return the response as JSON
+    return jsonify({"response": response})
+
+# Run the Flask app
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=5000)

appAPP.py

@@ -0,0 +1,83 @@
+from langchain.vectorstores import Qdrant
+from langchain.embeddings import HuggingFaceBgeEmbeddings
+from langchain_groq import ChatGroq
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.document_loaders import PyPDFLoader
+from langchain.chains.conversation.memory import ConversationBufferWindowMemory
+from langchain.prompts import PromptTemplate
+from langchain.chains import ConversationChain
+from qdrant_client import QdrantClient
+from dotenv import load_dotenv
+import os
+load_dotenv()
+
+# Initialize Qdrant and embeddings
+model_name = "BAAI/bge-large-en"
+embeddings = HuggingFaceBgeEmbeddings(
+    model_name=model_name,
+    model_kwargs={'device': 'cpu'},
+    encode_kwargs={'normalize_embeddings': False}
+)
+url = "http://localhost:6333"
+
+# Load the AIML vector database
+client = QdrantClient(
+    url=url, prefer_grpc=False
+)
+memory = ConversationBufferWindowMemory(k=5)
+collection_name = "APP"
+db = Qdrant(client=client , embeddings=embeddings,collection_name=collection_name)
+
+# Initialize Groq LLM
+groq_llm = ChatGroq(
+    groq_api_key = os.environ['GROQ_API_KEY'],
+    model = 'mixtral-8x7b-32768' 
+)
+
+# User input for query
+query = input("Enter your query: ")
+
+if query:
+    # Perform similarity search with context
+    docs = db.similarity_search_with_score(query=query, k=5)
+
+    # Display retrieved documents
+    print("Retrieved Documents:")
+    retrieved_content = ""
+    for i, (doc, score) in enumerate(docs):
+        retrieved_content += f"**Result {i + 1}:**\n"
+        retrieved_content += f"**Score:** {score}\n"
+        retrieved_content += f"**Content:** {doc.page_content}\n"
+        retrieved_content += f"**Metadata:** {doc.metadata}\n"
+        retrieved_content += "-" * 40 + "\n"
+
+    # Define the prompt template
+    prompt_template = PromptTemplate(
+        input_variables=["query", "retrieved_documents"],
+        template=(
+            "You are an expert in APP DEVELOPMENT. "
+            "Your task is to assist users with their queries related to APP DEVELOPMENT concepts, applications, and recent advancements. "
+            "You have access to relevant documents that provide detailed information about various APP DEVELOPMENT topics, algorithms, and case studies.\n\n"
+            "User Query: {query}\n\n"
+            "{retrieved_documents}\n\n"
+            "Your Response:\n"
+            "- Provide a concise and informative answer based on the user's query and the retrieved documents.\n"
+            "- Include key definitions, explanations of relevant concepts, and any important points from the retrieved documents.\n"
+            "- Ensure that the response is accessible to users with varying levels of knowledge about APP DEVELOPMENT."
+        )
+    )
+
+    # Format the input using the prompt template
+    formatted_input = prompt_template.format(query=query, retrieved_documents=retrieved_content)
+
+    # Set up generation model
+    groq_api_key = os.getenv('GROQ_API_KEY')
+    model = 'mixtral-8x7b-32768'  # Or any other model you prefer
+
+    # Initialize Groq Langchain chat object and conversation
+    groq_chat = ChatGroq(groq_api_key=groq_api_key, model_name=model)
+    conversation = ConversationChain(llm=groq_chat, memory=memory)
+
+    # Generate response using the formatted input
+    response = conversation.run(formatted_input)
+    print(response)