Mastering LLM Applications with LangChain and Hugging Face : Practical insights into LLM deployment and use cases

Cover
Title Page
Copyright Page
Dedication Page
About the Authors
About the Reviewer
Acknowledgements
Preface
Table of Contents
1. Introduction to Python and Code Editors
   Introduction
   Structure
   Objectives
   Introduction to Python
   Introduction to code editors
   Conclusion
   References
   Further reading
2. Installation of Python, Required Packages, and Code Editors
   Introduction
   Structure
   Objectives
   General instructions
   Installation of Python on Windows
   Installation of Python on Linux
   Installation of Python on MacOS
      Using Docker for Python
      Installation of IDE
   Installation of PyCharm
   Installation of required packages
      Virtual environment
         virtualenv
         pipenv
      Folder structure
         Creating a virtual environment
      PEP 8 standards
         Following PEP 8 in PyCharm
   Object-Oriented Programming concepts in Python
      Classes in Python
      Functions in Python
         For loop in Python
         While loop in Python
         If-else in Python
   Conclusion
3. Ways to Run Python Scripts
   Introduction
   Structure
   Objectives
   Setting up the project
   Running Python scripts from PyCharm
   Running Python Scripts from Terminal
   Running Python scripts from Jupyter Lab and Notebook
   Running Python Scripts from Docker
   Conclusion
4. Introduction of NLP and its concepts
   Introduction
   Structure
   Objectives
   Natural Language Processing overview
      Key concepts
      Corpus
      N-grams
      Tokenization
         Difference in tokens and n-grams
      Stop words removal
      Stemming
      Lemmatization
      Lowercasing
      Part-of-speech tagging
      Named Entity Recognition
      Bag of words
      Word embeddings
      Topic modeling
      Sentiment analysis
   Large language models
      Transfer learning
   Text classification
   Prompt engineering
      Hallucination
      Syntactic relationship
      Semantic relationship
   Conclusion
5. Introduction to Large Language Models
   Introduction
   Structure
   Objectives
   History
   LLM use cases
   LLM terminologies
   Neural networks
   Transformers
   Pre-built transformers
      Bidirectional Encoder Representations from Transformers
      Generative Pre-trained Transformer
      Text-to-text transfer transformer
      DistilBERT
      XLNet
      RoBERTa
   Conclusion
   Further readings
   References
6. Introduction to LangChain, Usage and Importance
   Introduction
   Structure
   Objectives
   LangChain overview
   Installation and setup
   Usages
   Opensource LLM models usage
   Data loaders
   Opensource text embedding models usage
   Vector stores
   Model comparison
   Evaluation
      Types of evaluation
   Conclusion
   Points to remember
   References
7. Introduction to Hugging Face, its Usage and Importance
   Introduction
   Structure
   Objectives
   Exploring the Hugging Face platform
   Installation and setup
   Datasets
   Usage of opensource LLMs
   Generating vector embeddings
   Evaluation
   Transfer learning with Hugging Face API
   Real-world use cases of Hugging Face
   Conclusion
   References
8. Creating Chatbots using Custom Data with Langchain and Hugging Face Hub
   Introduction
   Structure
   Objectives
   Setup
   Overview
   Steps to create RAG based chatbot with custom data
   Dolly-V2-3B details
   Data loaders by LangChain
   Vector stores by LangChain
   Conclusion
   References
9. Hyperparameter Tuning and Fine Tuning Pre-Trained Models
   Introduction
   Structure
   Objectives
   Hyperparameters of an LLM
   Hyperparameters at inferencing or at text generation
   Fine-tuning of an LLM
   Data preparation for finetuning an LLM
   Performance improvement
   Conclusion
   References
10. Integrating LLMs into Real-World Applications: Case Studies
   Introduction
   Structure
   Objectives
   Case studies
   Use case with Telegram
      Setup
   Conclusion
   References
11. Deploying LLMs in Cloud Environments for Scalability
   Introduction
   Structure
   Objectives
   Amazon Web Services
      Step 1: Creating an Amazon SageMaker Notebook Instance
      Step 2: Create folders in SageMaker to store data
      Step 3: Create vector embeddings
      Step 4: Auto scaling
   Google Cloud Platform
   Conclusion
   References
12. Future Directions: Advances in LLMs and Beyond
   Introduction
   Structure
   Objectives
   Generative AI market growth
   Reasoning
   Emergence of multimodal models
   Small domain-specific models
      Multi agent framework
   Quantization and Parameter-Efficient Fine Tuning
   Vector databases
   Guardrails
   Model evaluation frameworks
   Ethical and bias mitigation
   Safety and security
   Conclusion
   References
Appendix A: Useful Tips for Efficient LLM Experimentation
   Structure
   Objectives
   Understanding the challenges of LLM experimentation
   Preparing data for LLM experimentation
   Optimizing model architecture and hyperparameters
   Efficient training strategies for LLMs
   Evaluating and interpreting experimental results
   Fine-tuning for specific applications
   Scaling up: Distributed training and parallel processing
   Deployment considerations for LLMs
   Conclusion
   References
Appendix B: Resources and References
   Introduction
   Books and articles
   Research papers
   LangChain resources
   Hugging Face resources
      Alternative resources to LangChain
   Community and support
   Other important resources
   Conclusion
Index