File size: 1,314 Bytes
48a66db 5d2efff 48a66db 1ef1493 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
import streamlit as st
import os
# Set up page
st.set_page_config(
page_title="Aerospace Chatbot: AMS",
)
st.title("Aerospace Chatbot Homepage")
st.markdown("Code base: https://github.com/dsmueller3760/aerospace_chatbot/tree/rag_study")
st.markdown('---')
st.title("Chatbots")
st.markdown("""
Chatbots for aerospace mechanisms symposia, using all available papers published since 2000
* Modular version meant to study retrieval methods
""")
st.subheader("AMS")
'''
It is recommended to run this streamlit app locally for improved performance. The hosted hugging face version is for proof of concept.
You must have poetry installed locally to manage depdenencies. To run locally, clone the repository and run the following commands.
```poetry config virtualenvs.in-project true
```poetry install
```source .venv/bin/activate
```cd ./scripts
```streamlit run Start.py
'''
st.subheader("AMS")
'''
This chatbot will look up from all Aerospace Mechanism Symposia in the following location: https://github.com/dsmueller3760/aerospace_chatbot/tree/main/data/AMS
* Available models: https://platform.openai.com/docs/models
* Model parameters: https://platform.openai.com/docs/api-reference/chat/create
* Pinecone: https://docs.pinecone.io/docs/projects#api-keys
* OpenAI API: https://platform.openai.com/api-keys
''' |