Update README.md
Browse files
README.md
CHANGED
@@ -11,8 +11,6 @@ short_description: AI assistant for University at Buffalo students.
|
|
11 |
# UB Virtual Student Assistant (UB-VSA)
|
12 |
**RAG-powered chatbot for UB admissions, visa & campus life questions**
|
13 |
|
14 |
-
[](https://huggingface.co/spaces/TeamSAS/UB_VSA)
|
15 |
-
[](Dockerfile)β[](#license)
|
16 |
|
17 |
UB-VSA is a **retrieval-augmented conversational agent** that gives accurate, source-grounded answers about the University at Buffalo, with special focus on the needs of international students (F-1/OPT, I-20, course selection, housing, etc.).
|
18 |
It combines:
|
@@ -26,4 +24,39 @@ It combines:
|
|
26 |
| **UI** | Flask (+ vanilla JS) single-page chat |
|
27 |
---
|
28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
11 |
# UB Virtual Student Assistant (UB-VSA)
|
12 |
**RAG-powered chatbot for UB admissions, visa & campus life questions**
|
13 |
|
|
|
|
|
14 |
|
15 |
UB-VSA is a **retrieval-augmented conversational agent** that gives accurate, source-grounded answers about the University at Buffalo, with special focus on the needs of international students (F-1/OPT, I-20, course selection, housing, etc.).
|
16 |
It combines:
|
|
|
24 |
| **UI** | Flask (+ vanilla JS) single-page chat |
|
25 |
---
|
26 |
|
27 |
+
## β¨ Features
|
28 |
+
* **Context-aware chat** β tracks the last few turns for coherent follow-ups.
|
29 |
+
* **Grounded citations** β every answer lists clickable sources.
|
30 |
+
* **Lightweight adapters** β fine-tuned with LoRA/QLoRA; base model stays frozen.
|
31 |
+
* **One-click deploy** β Dockerfile + `start.sh` spin up scraper, vector-store, API & UI.
|
32 |
+
* **Scalable vector DB** β FAISS index stored in `data/embeddings/`.
|
33 |
+
|
34 |
+
---
|
35 |
+
|
36 |
+
## π₯οΈ Online Demo
|
37 |
+
> π **Try it live:** https://huggingface.co/spaces/TeamSAS/UB_VSA
|
38 |
+
|
39 |
+
Weβre on the free HF hardware tier; first response may take ~30 s while the container wakes up.
|
40 |
+
|
41 |
+
---
|
42 |
+
|
43 |
+
## π Quick Start (Local)
|
44 |
+
|
45 |
+
```bash
|
46 |
+
git clone https://huggingface.co/spaces/TeamSAS/UB_VSA
|
47 |
+
cd UB_VSA
|
48 |
+
|
49 |
+
# 1. Python deps
|
50 |
+
python -m venv .venv && source .venv/bin/activate
|
51 |
+
pip install -r requirements.txt
|
52 |
+
|
53 |
+
# 2. Set your HF token (needed for Cerebras Inference API)
|
54 |
+
export HUGGINGFACEHUB_API_TOKEN=hf_your_token_here
|
55 |
+
|
56 |
+
# 3. Build embeddings (only first time, ~5 min)
|
57 |
+
python main.py --build
|
58 |
+
|
59 |
+
# 4. Launch backend + frontend
|
60 |
+
python main.py --run
|
61 |
+
|
62 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|