https://github.com/ankush-chander/cricket-ama
RAG application on cricket using haystack and mistral
https://github.com/ankush-chander/cricket-ama
haystack llm mistral question-answering rag
Last synced: 12 months ago
JSON representation
RAG application on cricket using haystack and mistral
- Host: GitHub
- URL: https://github.com/ankush-chander/cricket-ama
- Owner: Ankush-Chander
- Created: 2023-10-07T11:03:50.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-10-07T18:05:55.000Z (over 2 years ago)
- Last Synced: 2025-02-14T21:57:43.821Z (12 months ago)
- Topics: haystack, llm, mistral, question-answering, rag
- Language: Jupyter Notebook
- Homepage:
- Size: 6.84 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Cricket-ama
Ask me anything about cricket. I will try to answer it to the best of my knowledge.
[RAG(Retrieval Augmented Generation)](https://www.pinecone.io/learn/retrieval-augmented-generation/) is a technique in which we use powers of LLM to generate accurate information by grounding it with a known data source. This way we can leverage an LLM interface while avoiding hallucination .
This notebook is an attempt to build a cricket ama interface using RAG.
# Data source
1. Data has been generated using [pages-articles-multistream wiki dump](https://dumps.wikimedia.org/enwiki/latest/). Alternatively it can be downloaded from [here](https://huggingface.co/datasets/mrsearchwolf/cricket-wiki)
# Tech Stack
1. [Haystack](https://docs.haystack.deepset.ai/docs)
2. [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1)
# Acknowlegement
This notebook is inspired from [mistral-haystack](https://github.com/anakin87/mistral-haystack) by [Stefano](https://github.com/anakin87).