Use Naive RAG Chatbot at HF-Space

AutoRAG HF Space Tutorial #1. Naive RAG Chatbot

[embed]Naive RAG Chatbot - a Hugging Face Space by AutoRAG Quickest way to test naive RAG run with AutoRAG.huggingface.co

Quick Start

Set OPENAI API KEY → Select Your own data → Use Chatbot

0. Basic

AutoRAG optimizes for Advanced RAG, but this space is for those who want to try Naive RAGs quickly and easily.

Literally, you can try Naive RAG, which are the basics of RAG.

Comparison between the three paradigms of RAG (Gao et al. 2024)

🛠️ Default Setting

Embedding Model: OPENAI text-embedding-ada-002

LLM: gpt-4o-mini

prompt:

You are the helpful assistant to answer the question. I will give you a context to read. The context can be unrelated to the question.
If the context is related, you must answer the question base on the context.
If there is no context that relates to the question, you must say that you don't know about the answer.
DO NOT MAKE UP THE ANSWER.
If you can solve the question with your own knowledge, you can answer the question. But please do not lie or make up the answer without relevant information.
Question: {query} \\n Context: {retrieved_contents} \\n Answer :

VectorDB: Chroma

You’ll need 2 pieces.

“OPENAI API KEY”, “Your own Data”

1. Set API Key

API Key setting complete

📌 Note

❗AutoRAG do not store your API Key❗

2. Ingest Data

File uploaded complete

Click to upload the data.

And Click “Submit file” This will ingest the data using the embedding model

3. Use Chatbot

Example