https://github.com/trilogy-group/bedrock-workshop
https://github.com/trilogy-group/bedrock-workshop
Last synced: 9 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/trilogy-group/bedrock-workshop
- Owner: trilogy-group
- Created: 2023-10-13T15:39:16.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-11-02T19:20:13.000Z (about 2 years ago)
- Last Synced: 2025-01-29T12:18:06.311Z (11 months ago)
- Language: Python
- Size: 5.86 KB
- Stars: 0
- Watchers: 3
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AWS Bedrock and LangChain Workshop
This repository contains code to create a question answering bot that can answer questions about AWS based on the [AWS Made Easy Livestream](https://www.youtube.com/channel/UC-iqkyAw-jFd0RtdRdqzvRQ) livestream. The bot is built using [AWS Bedrock](https://aws.amazon.com/bedrock/) for llms and [LangChain](https://www.langchain.com/).
## Workshop Agenda
In this workshop we will:
- Set up an AWS Bedrock environment
- Scrape transcripts from the AWS Made Easy podcast
- Preprocess and format the text data
- Store the transcripts in vector store usin Titan Embeddings
- Implement RAG to retrieve relevant documents when asked a question
## Prerequisites
- An AWS account
- Python 3.7+
- Git and a GitHub account
## Getting Started
1. Clone this repository
2. Download `ame-data-oct5.csv` to get all podcast transcripts
3. Run `create_scripts.py` to get txt transcrits
4. Run `streamlit run app.py` to start the streamlit server
## Resources
- [AWS Bedrock](https://aws.amazon.com/bedrock/)
- [LangChain](https://www.langchain.com/)
- [AWS Made Easy Livestream](https://www.youtube.com/channel/UC-iqkyAw-jFd0RtdRdqzvRQ)
## Credits
This workshop was created by [Harsh Arya](https://github.com/harsharyacnu).