Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -104,11 +104,9 @@ for message in st.session_state.messages:
|
|
104 |
st.sidebar.markdown(
|
105 |
"""
|
106 |
|
107 |
-
|
108 |
|
109 |
-
This
|
110 |
-
|
111 |
-
Once entered a question, the system will search for relevant questions in the database and then produce an answer based on the context answers in the database with a distance score. The distance score represents how similar the user question is comparing with the question in the database. The smaller the distance score the more similar / relevant the database is at producing similar context. For hallucination cases, a viable strategy is to collect user preferences and update the database.
|
112 |
|
113 |
""")
|
114 |
|
|
|
104 |
st.sidebar.markdown(
|
105 |
"""
|
106 |
|
107 |
+
### Instructions:
|
108 |
|
109 |
+
This app guides you through YSA's website, utilizing a RAG-ready Q&A dataset [here](https://huggingface.co/datasets/eagle0504/youthless-homeless-shelter-web-scrape-dataset-qa-formatted) for chatbot assistance. 🤖 Enter a question, and it finds similar ones in the database, offering answers with a distance score to gauge relevance—the lower the score, the closer the match. 🎯 For better accuracy and to reduce errors, user feedback helps refine the database. ✨
|
|
|
|
|
110 |
|
111 |
""")
|
112 |
|