This repository contains Jupyter notebooks for the course "LLMs from Prototypes to Production" by Sinan Ozdemir. Published by Pearson, the course covers effective best practices and industry case studies in using Large Language Models (LLMs).

Once you understand how large language models (LLMs) work, define your prompts, and train your models, it’s time to move your LLM prototypes to production and fine-tune your models for optimal performance. We will cover the best practices for integrating LLMs into various workflows, deployment options, and model evaluation. This course will empower you to confidently transition from LLM prototypes to fully realized applications and optimize their performance.

This course is the third in a three-part series by Sinan Ozdemir designed for machine learning engineers and software developers who want to expand their skillset and learn how to work with large language models (LLMs) like ChatGPT and FLAN-T5. The courses provide practical instruction on prompt engineering, language modeling, moving LLM prototypes to production, and fine-tuning GPT models. The three live courses in the series are:

LLMs, GPT, and Prompt Engineering for Developers Using Open- and Closed-Source LLMs in Real-World Applications LLMs from Prototypes to Production The book Quick Start Guide to LLMs by Sinan Ozdemir is recommended as companion material for for post-class reference.

1. LLMs, GPT, and Prompt Engineering for Developers

2. Using Open- and Closed-Source LLMs in Real World Applications

3. LLMs from Prototypes to Production

The book Quick Start Guide to LLMs by Sinan Ozdemir is recommended as companion material for post-class reference.

• How to effectively move LLM prototypes to production environments
• The various deployment options and considerations for LLMs
• Techniques for fine-tuning LLMs like GPT and FLAN-T5
• How to evaluate and improve LLM model performance

And you’ll be able to:

• Seamlessly transition LLM prototypes into production systems
• Integrate LLMs into various workflows and applications
• Fine-tune LLMs for optimal performance
• Evaluate and enhance LLM models based on specific use cases

1. Course Set-Up
2. Notebooks
3. Prerequisites
4. Schedule
5. Resources

• Jupyter notebooks can be run alongside the instructor, but you can also follow along without coding by viewing pre-run notebooks here.

• Experience with machine learning and proficiency in Python programming
• Familiarity with NLP is helpful but not required

• Attend the course "LLMs, GPT, and Prompt Engineering for Developers"
• Read the book "Quick Start Guide to Large Language Models"

• distillation.ipynb: Notebook covering distillation techniques

• rl_flan_t5_summaries.ipynb: Working with FLAN-T5 models using Reinforcement Learning

• openai_fine_tuned_classification.ipynb: Fine-tuning OpenAI models for classification

• BERT vs GPT for CLF.ipynb: Comparing Fine-tuned OpenAI to BERT

• Fine-tuning GPT2.ipynb: Seeing how small datasets can transform LLMs

• metrics.ipynb: Evaluation Metrics

• rlaif.ipynb: Investigating Principles of RLAIF and Constituional AI

• Quantization: loading models with less precision and testing for latency and performance

• Checkout out the Streamlit directory for Streamlit apps!

For a detailed schedule, refer to the Course Description.

• Course Website
• Other Useful Links