Feeding the LLMs with Context from Internet Searches or Vector Searches.
Building the prompts programmatically with Python and Jinja Templates.
Example prompts that make the agents to do magical stuff.
Chapter Navigation:
0:00 Intro
0:25 Simple Agent Steps
1:05 What is Jinja
1:55 How Jinja Works
3:15 Feeding LLMs with Context
4:05 Jinja2 Code Walk Through
4:50 Reading Prompt Templates with Environment
6:20 What are Prompt Templates
7:55 Helper Functions
8:50 Loading Templates in Environment
9:45 Creating Input Context Variables
10:50 Rendering Templates
12:40 Prompt Template with Loops Explained
14:40 Rendering Templates with List Data
15:50 Retrieving Internet Search & Render Prompts
18:40 Render Code Extraction Prompt with Search Results
19:30 Working with System & User Prompts
20:00 Recap on Dismantling Prompts
21:33 Recovering from Agent Failures
22:30 Outro
The data and the code is located at
https://github.com/insightbuilder/pyt...
I believe you will like this video, and subscribe to the channel. Further uploads related to Big Data, Large Language models and Artificial Intelligence will be shared to your Youtube Dashboard Directly.
The supporting playlists are
The bard Project
• Google Bard LLM : New AI Model in the...
Practical Projects Playlist
• The Future is Here: Large Language Mo...
Huggingface Playlist
• Mastering NLP with Hugging Face: A Co...
Python Data Engineering Playlist
• Learn to Data Engineer and Problem So...
Python Ecosystem of Libraries
• Mastering the Python Ecosystem: Must-...
ChatGPT and AI Playlist
• Learn about AI Language Models and Re...
AWS and Python AWS Wrangler
• Building a Powerful Data Pipeline wit...
Exploring the Realm of Generative AI: Hardware and Software Discussions
• Exploring the Realm of Generative AI:...
PS: Got a question or have a feedback on my content. Get in touch
By leaving a Comment in the video
topmate.io/insightbuilder?utm_source=topmate&utm_medium=popup&utm_campaign=SocialProfile
@mail [email protected]
@twitter Handle is @KQrios
@medium / about
@github https://github.com/Kamalabot