Learn how to train GPT-3 on a corpus of Supreme Court opinions using ChatGPT and knowledge graphs in this comprehensive video tutorial. Explore techniques for creating cross-linked webs of precedent, breaking down files into manageable chunks, and using English as a primary programming language. Master troubleshooting skills for encoding issues, clean up duplicated characters using Spacey and NLTK, and optimize file sizes with Python. Discover how to write scripts using the OpenAI module, create self-training machines, and debug using pseudocode. Gain insights into UTF-8 file functions, Windows error troubleshooting, and data merging and visualization techniques.
Train GPT-3 on Any Corpus of Data with ChatGPT and Knowledge Graphs - SCOTUS Opinions Part 2
David Shapiro ~ AI via YouTube
Overview
Syllabus
- Creating a Cross-Linked Web of Precedent
- Breaking down files into chunks of four pages
- English as the Primary Programming Language
- Troubleshooting Encoding Issues
- Talking to Hal 9000
- Cleaning up Duplicated Characters with Spacey
- Debugging a Python Program
- Troubleshooting Spacey Installation
- De-duplicating Characters with NLTK
- Modifying the Script to Use Regex
- Optimizing File Sizes with Python
- Writing a Python Script with Open AI Module
- Creating a Machine that can Train Itself
- Debugging a Script with Pseudocode
- Writing a File Function in UTF-8
- Troubleshooting a Windows Error
- Merging and Visualizing
Taught by
David Shapiro ~ AI
Related Courses
-
Train GPT-3 on Any Corpus of Data with ChatGPT and Knowledge Graphs - SCOTUS Opinions Part 1
-
Answer Complex Questions From an Arbitrarily Large Set of Documents With Vector Search and GPT-3
-
Finetune GPT-3 to Write a Coherent Novel - Success! With Minor Bugs
-
Creating a GPT-3 Webapp with OpenAI API, FastAPI, HTML, JavaScript, and Python
-
How to Fine-Tune a GPT-3 Model - Step by Step
-
Fine-Tune GPT-3 to Write an Entire Coherent Novel - Part 2
