Python version, langchain libraries and document upgrade

README.md

@@ -8,7 +8,30 @@ Transform unstructured data (PDFs, DOCs, TXT, YouTube videos, web pages, etc.) i
 
 This application allows you to upload files from various sources (local machine, GCS, S3 bucket, or web sources), choose your preferred LLM model, and generate a Knowledge Graph. 
 
----
+## Getting Started
+
+### **Prerequisites**
+- **Python 3.12 or higher** (for local/separate backend deployment)
+- Neo4j Database **5.23 or later** with APOC installed.
+  - **Neo4j Aura** databases (including the free tier) are supported.
+  - If using **Neo4j Desktop**, you will need to deploy the backend and frontend separately (docker-compose is not supported).
+
+#### **Backend Setup**
+1. Create the `.env` file in the `backend` folder by copying `backend/example.env`.
+2. Preconfigure user credentials in the `.env` file to bypass the login dialog:
+   ```bash
+   NEO4J_URI=<your-neo4j-uri>
+   NEO4J_USERNAME=<your-username>
+   NEO4J_PASSWORD=<your-password>
+   NEO4J_DATABASE=<your-database-name>
+   ```
+3. Run:
+   ```bash
+   cd backend
+   python3.12 -m venv venv
+   source venv/bin/activate  # On Windows: venv\Scripts\activate
+   pip install -r requirements.txt -c constraints.txt
+   uvicorn score:app --reload
 
 ## Key Features
 

backend/Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.10-slim
+FROM python:3.12-slim
 WORKDIR /code
 ENV PORT 8000
 EXPOSE 8000

backend/README.md

@@ -1,6 +1,11 @@
 # Project Overview
 Welcome to our project! This project is built using FastAPI framework to create a fast and modern API with Python.
 
+## Prerequisites
+
+- Python 3.12 or higher
+- pip (Python package manager)
+
 ## Feature
 API Endpoint : This project provides various API endpoint to perform specific tasks.
 Data Validation : Utilize FastAPI data validation and serialization feature.
@@ -16,9 +21,14 @@ Follow these steps to set up and run the project locally:
 
 > cd llm-graph-builder
 
-2. Install Dependency :
+2. Create a virtual environment (recommended):
+
+> python3.12 -m venv venv
+> source venv/bin/activate  # On Windows: venv\Scripts\activate
+
+3. Install Dependency :
 
-> pip install -t requirements.txt
+> pip install -r requirements.txt -c constraints.txt
 
 ## Run backend project using unicorn
 Run the server:

backend/requirements.txt

@@ -1,65 +1,64 @@
-accelerate==1.7.0
-asyncio==3.4.3
-boto3==1.38.36
-botocore==1.38.36
-certifi==2025.6.15
-fastapi==0.115.12
+accelerate==1.12.0
+asyncio==4.0.0
+boto3==1.40.23
+botocore==1.40.23
+certifi==2025.8.3
+fastapi==0.116.1
 fastapi-health==0.4.0
 fireworks-ai==0.15.12
 google-api-core==2.25.1
 google-auth==2.40.3
 google_auth_oauthlib==1.2.2
 google-cloud-core==2.4.3
-json-repair==0.39.1
+json-repair==0.44.1
 pip-install==1.3.5
-langchain==0.3.25
-langchain-aws==0.2.25
-langchain-anthropic==0.3.15
-langchain-fireworks==0.3.0
-langchain-community==0.3.25
-langchain-core==0.3.65
-langchain-experimental==0.3.4
-langchain-google-vertexai==2.0.25
-langchain-groq==0.3.2
-langchain-openai==0.3.23
-langchain-text-splitters==0.3.8
-langchain-huggingface==0.3.0
+langchain==1.1.2
+langchain-aws==1.1.0
+langchain-anthropic==1.2.0
+langchain-fireworks==1.1.0
+langchain-community==0.4.1
+langchain-core==1.1.1
+langchain-experimental==0.4.0
+langchain-google-vertexai==3.1.1
+langchain-groq==1.1.0
+langchain-openai==1.1.0
+langchain-text-splitters==1.0.0
+langchain-huggingface==1.1.0
+langchain-classic==1.0.0
 langdetect==1.0.9
-langsmith==0.3.45
-langserve==0.3.1
-neo4j-rust-ext==5.28.1.0
+langsmith==0.4.55
+langserve==0.3.3
+neo4j-rust-ext==5.28.2.1
 nltk==3.9.1
-openai==1.86.0
-opencv-python==4.11.0.86
+openai==2.9.0
 psutil==7.0.0
-pydantic==2.11.7
-python-dotenv==1.1.0
+pydantic==2.12.5
+python-dotenv==1.1.1
 python-magic==0.4.27
 PyPDF2==3.0.1
-PyMuPDF==1.26.1
-starlette==0.46.2
-sse-starlette==2.3.6
+PyMuPDF==1.26.4
+starlette==0.47.3
+sse-starlette==3.0.2
 starlette-session==0.4.3
 tqdm==4.67.1
-unstructured[all-docs]
-unstructured==0.17.2
-unstructured-client==0.36.0
+unstructured[all-docs]==0.18.14
+unstructured-client==0.42.3
 unstructured-inference==1.0.5
-urllib3==2.4.0
-uvicorn==0.34.3
+urllib3==2.5.0
+uvicorn==0.35.0
 gunicorn==23.0.0
 wikipedia==1.4.0
-wrapt==1.17.2
+wrapt==1.17.3
 yarl==1.20.1
-youtube-transcript-api==1.1.0
+youtube-transcript-api==1.2.2
 zipp==3.23.0
-sentence-transformers==5.0.0
+sentence-transformers==5.1.0
 google-cloud-logging==3.12.1
 pypandoc==1.15
-graphdatascience==1.15.1
-Secweb==1.18.1
-ragas==0.3.1
+graphdatascience==1.18a1
+Secweb==1.25.2
+ragas==0.3.2
 rouge_score==0.1.2
-langchain-neo4j==0.4.0
+langchain-neo4j==0.6.0
 pypandoc-binary==1.15
 chardet==5.2.0

backend/src/QA_integration.py

@@ -11,12 +11,11 @@
 from langchain_neo4j import Neo4jVector
 from langchain_neo4j import Neo4jChatMessageHistory
 from langchain_neo4j import GraphCypherQAChain
-from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.runnables import RunnableBranch
-from langchain.retrievers import ContextualCompressionRetriever
-from langchain_community.document_transformers import EmbeddingsRedundantFilter
-from langchain.retrievers.document_compressors import EmbeddingsFilter, DocumentCompressorPipeline
+from langchain_classic.retrievers import ContextualCompressionRetriever
+from langchain_classic.retrievers.document_compressors import EmbeddingsFilter, DocumentCompressorPipeline
 from langchain_text_splitters import TokenTextSplitter
 from langchain_core.messages import HumanMessage, AIMessage
 from langchain_community.chat_message_histories import ChatMessageHistory 

backend/src/create_chunks.py

@@ -1,5 +1,5 @@
 from langchain_text_splitters import TokenTextSplitter
-from langchain.docstore.document import Document
+from langchain_core.documents import Document
 from langchain_neo4j import Neo4jGraph
 import logging
 from src.document_sources.youtube import get_chunks_with_timestamps, get_calculated_timestamps

backend/src/document_sources/youtube.py

@@ -1,4 +1,4 @@
-from langchain.docstore.document import Document
+from langchain_core.documents import Document
 from src.shared.llm_graph_builder_exception import LLMGraphBuilderException
 from youtube_transcript_api import YouTubeTranscriptApi 
 from youtube_transcript_api.proxies import GenericProxyConfig

backend/src/llm.py

@@ -1,5 +1,5 @@
 import logging
-from langchain.docstore.document import Document
+from langchain_core.documents import Document
 import os
 from langchain_openai import ChatOpenAI, AzureChatOpenAI
 from langchain_google_vertexai import ChatVertexAI

backend/src/make_relationships.py

@@ -1,6 +1,5 @@
 from langchain_neo4j import Neo4jGraph
-from langchain.docstore.document import Document
-from src.shared.common_fn import load_embedding_model,execute_graph_query
+from langchain_core.documents import Document
 from src.shared.common_fn import load_embedding_model,execute_graph_query
 import logging
 from typing import List
@@ -34,7 +33,6 @@ def merge_relationship_between_chunk_and_entites(graph: Neo4jGraph, graph_docume
                     MERGE (c)-[:HAS_ENTITY]->(n)
                 """
         execute_graph_query(graph,unwind_query, params={"batch_data": batch_data})
-        execute_graph_query(graph,unwind_query, params={"batch_data": batch_data})
 
 
 def create_chunk_embeddings(graph, chunkId_chunkDoc_list, file_name):
@@ -61,7 +59,6 @@ def create_chunk_embeddings(graph, chunkId_chunkDoc_list, file_name):
         MERGE (c)-[:PART_OF]->(d)
     """       
     execute_graph_query(graph,query_to_create_embedding, params={"fileName":file_name, "data":data_for_query})
-    execute_graph_query(graph,query_to_create_embedding, params={"fileName":file_name, "data":data_for_query})
 
 def create_relation_between_chunks(graph, file_name, chunks: List[Document])->list:
     logging.info("creating FIRST_CHUNK and NEXT_CHUNK relationships between chunks")
@@ -130,7 +127,6 @@ def create_relation_between_chunks(graph, file_name, chunks: List[Document])->li
         MERGE (c)-[:PART_OF]->(d)
     """
     execute_graph_query(graph,query_to_create_chunk_and_PART_OF_relation, params={"batch_data": batch_data})
-    execute_graph_query(graph,query_to_create_chunk_and_PART_OF_relation, params={"batch_data": batch_data})
 
     query_to_create_FIRST_relation = """ 
         UNWIND $relationships AS relationship
@@ -140,7 +136,6 @@ def create_relation_between_chunks(graph, file_name, chunks: List[Document])->li
                 MERGE (d)-[:FIRST_CHUNK]->(c))
         """
     execute_graph_query(graph,query_to_create_FIRST_relation, params={"f_name": file_name, "relationships": relationships})
-    execute_graph_query(graph,query_to_create_FIRST_relation, params={"f_name": file_name, "relationships": relationships})
 
     query_to_create_NEXT_CHUNK_relation = """ 
         UNWIND $relationships AS relationship

docs/project_docs.adoc

@@ -21,6 +21,11 @@ This document provides comprehensive documentation for the Neo4j llm-graph-build
 
 == Local Setup and Execution
 
+Prerequisites:
+- Python 3.12 or higher
+- Node.js 20 or higher
+- Docker (optional, for containerized deployment)
+
 Run Docker Compose to build and start all components:
 ....
 docker-compose up --build
@@ -38,8 +43,8 @@ yarn run dev
 ** For backend
 ....
 cd backend
-python -m venv envName
-source envName/bin/activate 
+python3.12 -m venv venv
+source venv/bin/activate 
 pip install -r requirements.txt
 uvicorn score:app --reload
 ....