|
--- |
|
title: Aerospace Chatbot, Aerospace Mechanism Symposia |
|
emoji: 🚀 |
|
colorFrom: gray |
|
colorTo: blue |
|
sdk: docker |
|
pinned: false |
|
app_port: 8501 |
|
short_description: Aerospace Chatbot, AMS |
|
datasets: |
|
- ai-aerospace/ams_data_full_2000-2020 |
|
--- |
|
|
|
# Aerospace Chatbot |
|
Aerospace discipline-specific chatbots and AI tools. |
|
Help docs here: https://aerospace-chatbot.readthedocs.io/en/latest/ |
|
|
|
The github repository is located here: [aerospace chatbot](https://github.com/dan-s-mueller/aerospace_chatbot) |
|
|
|
## This Deployment |
|
This deployment is loaded with all Aerospace Mecahnism Symposia papers located in [/data/AMS/](https://github.com/dan-s-mueller/aerospace_chatbot/tree/documentation_update/data/AMS) on Hugging Face Spaces. |
|
|
|
There is persistent storage of vector databases when using RAGatouille or ChromaDB. There are a few of preloaded databases. * indicates the database config (OpenAI/Hugging Face, RAGatouille/ChromaDB). |
|
|
|
- *-ams-15merge-2000-2020: Database with each document representing 15 merged pages. Covers AMS years 2000-2020. |
|
- *-ams-nomerge-400-0-2000-2020: Database with each document representing chunks of 400 characters, no overlap, and no page merging. Covers AMS years 2000-2020. |
