LLM Projects Repository

This repository contains a comprehensive collection of over 50 projects focusing on various aspects of Machine Learning (ML), Deep Learning (DL), and Large Language Models (LLMs). The projects span across multiple domains and approaches, including LangChain, fine-tuning, retrieval-augmented generation (RAG), and more.

Table of Contents

1. LangChain Projects
2. Fine-Tuning Projects
3. Retrieval-Augmented Generation (RAG) Projects
4. Advanced Topics
5. Agent and Integration Projects

LangChain Projects

1. Langchain_Basics with Google Generative AI.ipynb

This notebook provides an introduction to using LangChain with Google Generative AI. It covers the basics of setting up the environment, connecting to Google Generative AI, and implementing basic operations.

2. LangChain basics using LangChain-VertexAI.ipynb

Learn the fundamentals of LangChain with VertexAI, including setup, basic configurations, and running initial experiments.

3. Langchain_Basics_with_Native VertexAI.ipynb

This project demonstrates how to utilize LangChain with Native VertexAI, focusing on native integrations and basic functionalities.

4. Langchain_LCEL & Output Parsing.ipynb

Explore the capabilities of LangChain for Local Contextualized Embedding Language (LCEL) and output parsing techniques.

5. LangChain_Parsers.ipynb

An in-depth look at different parsers available in LangChain and how to utilize them effectively in your projects.

6. Langchain - Basics Projects.ipynb

A collection of basic projects to get you started with LangChain, covering various simple use cases and examples.

7. LangChain - Document Loaders.ipynb

Learn how to load and process documents using LangChain's document loader capabilities.

8. LangChain - Document Loaders with URL.ipynb

An extension of document loaders, this notebook focuses on loading documents from URLs and processing them.

44. LangChain CSV Agent - Talk to CSV & Excel Files.ipynb

Develop a LangChain agent that can interact with CSV and Excel files, providing data extraction and processing capabilities.

43. LangGraph - Self-RAG with Pinecone movies.ipynb

Develop a self-RAG system with LangGraph and Pinecone, focusing on movie data and retrieval methods.

38. LangGraph - Adaptative RAG.ipynb

Learn how to create adaptive RAG systems using LangGraph, showcasing dynamic adaptation capabilities.

42. LangGraph - Self-RAG.ipynb

Implement self-RAG systems using LangGraph, focusing on self-improving retrieval and generation methods.

40. LangGraph Agentic RAG.ipynb

Implement an agentic RAG system using LangGraph, focusing on agent-based retrieval and generation.

41. LangGraph - CRAG.ipynb

Explore CRAG techniques with LangGraph, focusing on contextual retrieval and generation.

9. Basic Multi-Modal app.ipynb

Create multi-modal AI apps with image and text support.

Fine-Tuning Projects

17. Fine Tune Falcon-7b.ipynb

Detailed instructions and examples on fine-tuning the Falcon-7b model for specific tasks.

18. Lora_YT_PEFT_Finetune_Bloom7B_tagger.ipynb

Learn how to fine-tune the Bloom7B model using LoRA and PEFT techniques for tagging tasks.

19. QLora_bnb_4bit_training_with_inference.ipynb

Implement QLora with 4-bit training and inference, focusing on efficient model training techniques.

20. Basic LLM Supervised Fine-Tuning with QLora.ipynb

Fine tune LLMs using QLora for better hardware adaptation.

13. Supervised Fine-Tuning LLM.ipynb

Understand the process of supervised fine-tuning for large language models, with practical examples and step-by-step guidance.

10. Basic Fine-Tuning LLMs with PEFT and LoRA.ipynb

Learn how to fine-tune large language models using Parameter Efficient Fine-Tuning (PEFT) and Low-Rank Adaptation (LoRA) techniques.

22. Fine-Tuning llama-2-7b-miniguanaco.ipynb

Fine-tune the Llama-2-7b model for specific tasks, using the miniguanco dataset as an example.

27. Fine-Tune Llama 2-7b.ipynb

A detailed guide on fine-tuning the Llama 2-7b model.

30. Fine_tune_Mistral_7b_with_SFT.ipynb

Fine-tune the Mistral-7b model using Supervised Fine-Tuning (SFT) techniques, with detailed examples and guidance.

Retrieval-Augmented Generation (RAG) Projects

12. Basic RAG app.ipynb

Create a basic Retrieval-Augmented Generation (RAG) application, showcasing the fundamentals of RAG models.

14. Basic conversational RAG app.ipynb

Develop a basic conversational RAG application, integrating conversational AI capabilities with RAG models.

16. Basic LlamaIndex RAG app.ipynb

Develop a basic RAG application using LlamaIndex, demonstrating the integration and usage of LlamaIndex in RAG systems.

21. Basic LLM-as-a-judge RAG app.ipynb

Create a basic RAG application where the LLM acts as a judge, providing evaluations and decisions based on input data.

24. Basic DSPy RAG app with Gemini.ipynb

Develop a basic DSPy RAG application using Gemini, focusing on integrating DSPy with RAG systems.

25. Advanced RAG - Self Querying Retrieval.ipynb

Explore advanced self-querying retrieval techniques in RAG models, with practical implementations and examples.

26. Advanced RAG - Parent Document Retriever.ipynb

Implement an advanced parent document retriever system in RAG, focusing on hierarchical document retrieval.

28. Advanced RAG - BM25 + Embeddings + Hybrid Search (Ensemble).ipynb

Create an advanced RAG system that combines BM25, embeddings, and hybrid search techniques in an ensemble model.

31. Advanced RAG - Contextual Compression + Filtering.ipynb

Develop an advanced RAG system that includes contextual compression and filtering techniques for improved retrieval.

32. Advanced RAG - HyDE.ipynb

Explore the HyDE technique in advanced RAG systems, focusing on dynamic and adaptive retrieval methodologies.

34. RagFusion_MakerSuite_PaLM_2_Langchain_with_Chroma_&_BGE.ipynb

Implement RagFusion with MakerSuite PaLM 2, integrating LangChain with Chroma and BGE for advanced RAG capabilities.

35. Advanced RAG - RAPTOR Retrieving.ipynb

Develop an advanced RAG system with RAPTOR retrieving, focusing on efficient retrieval techniques.

37. Advanced RAG - Dynamic RAPTOR.ipynb

Develop an advanced RAG system with Dynamic RAPTOR, showcasing adaptive and dynamic retrieval methods.

39. Email response automation with CrewAI and LangGraph

Automate email responses using CrewAI and LangGraph, showcasing practical applications of automated email handling.

45. Advanced RAG with Merger Retriever (LOTR) and Re-Ranking retriever (For long context reorder).ipynb

Implement an advanced RAG system with Merger Retriever and Re-Ranking retriever, focusing on long-context reordering techniques.

Advanced Topics

9. Two Stage Retrieval with Cross Encoder (BERT).ipynb

Implement a two-stage retrieval system using Cross Encoder with BERT, showcasing advanced retrieval techniques.

15. Basic Q&A system with SQL Data.ipynb

Create a basic Question and Answer system that interacts with SQL databases to fetch and process data.

33. Introduction to Weight Quantization.ipynb

A primer on weight quantization techniques in neural networks, demonstrating the basics and importance of quantization.

36. 4_bit_LLM_Quantization_with_GPTQ.ipynb

Learn about 4-bit quantization techniques for LLMs using GPTQ, focusing on efficient model quantization.

48. SQL Agent with OpenAI GPT 3.5.ipynb

Create an SQL agent using OpenAI GPT 3.5, enabling SQL query processing and data retrieval capabilities.

49. Quantized Llama 3 with Gradio UI.ipynb

Develop a quantized Llama 3 model with a Gradio UI, showcasing efficient model deployment and interaction.

62. Quantized_Llama_3_1_with_Gradio_UI.ipynb

Develop a quantized Llama 3.1 model with a Gradio UI, showcasing efficient model deployment and interaction.

64. Quantized Mistral 7B with Gradio UI.ipynb

Develop a quantized Mistral 7B model with a Gradio UI, showcasing efficient model deployment and interaction.

65. Quantized Phi-3-mini-4K with Gradio UI.ipynb

Develop a quantized Phi-3-mini-4K with a Gradio UI, showcasing efficient model deployment and interaction.

66. Quantized Llama 3.1 Instruct with Gradio UI.ipynb

Develop a quantized Llama 3.1 Instruct with a Gradio UI, showcasing efficient model deployment and interaction.

Agent and Integration Projects

23. PAL_chain_with_LangChain.ipynb

Implement the PAL chain technique using LangChain, showcasing advanced chain methodologies.

29. Gemma 2 9B Model - Google.ipynb

Explore the Gemma 2 9B model from Google, with detailed examples and use cases.

46. BabyAGI - LangChain with Tools.ipynb

Implement BabyAGI using LangChain and various tools, focusing on autonomous agent capabilities.

47. BabyAGI - LangChain without Tools.ipynb

Develop BabyAGI using LangChain without external tools, showcasing minimalist agent development.

50. Reflection with OpenAI and LangGraph.ipynb

Explore reflection techniques with OpenAI and LangGraph, focusing on self-improving models.

51. GPT 4o mini - Testing.ipynb

A testing notebook for the GPT 4o mini model, demonstrating various testing scenarios and results.

52. Agent Executor From Scratch using Google GenAI.ipynb

Learn how to build an agent executor from scratch using Google Generative AI, with detailed implementation steps.

53. Agent Executor with force-calling-a-tool-first.ipynb

Implement an agent that can run web-searching tasks with a first compulsary tool calling process and enhancing flexibility for further tasks.

54. Agents - Human_in_the_loop.ipynb

Implements agents with Human intervention for accomplishing certain defined tasks in LangGraph

55. Agents - Managing Agent Steps.ipynb

Integrates a mechanism for limiting the number of agent steps in iterative processes for reducing the computational and token cost.

56. Chat Agent Executor with GPT-4o-mini.ipynb

Implement a Chat Agent Executor with ToolNodes (TavilySearch) for searching for real-time info. if needed with GPT-4o-mini.

57. Chat_Agent_Executor_(base)_with_GPT_4o_mini.ipynb

Implement a Chat Agent Executor with ToolNodes (TavilySearch) for searching for real-time info. if needed with GPT-4o-mini and the base agent model.

58. Chat_Agent_Dynamically_Returning_Directly

Implement a Chat Agent and adding the dynamically returning directly when the output of the tool call is appropriate for answering the user's query.

59. Chat_Agent_Force_Calling_A_Tool_First.ipynb

Develop a Chat Agent that first mades a call to a tool and then offers flexibility for further tool calling activities.

60. ReAct_Agent_High_Level_Tools.ipynb

Implement a ReAct Agent with a WebSearch Tool (Tavily API) for agile development and deployment.

61. Chat_Agent_Human_In_The_Loop.ipynb

Develop a Chat Agent that has a "human-in-the-loop" intervention in the AI workflow.

63. ChatAgent - Managing Agent Steps.ipynb

Implement a Chat Agent limiting the number of messages in the chat history for optimal performance.

67. ChatAgent - Prebuilt Tool Node.ipynb

Develop a Chat Agent with a pre built Tool Node with Tavily for searching info. in the web.

68. ChatAgent - Respond in format.ipynb

Develop a Chat Agent with deterministic responses (With appropriate formatting).