Upload 2 files
Browse files
buffalo_rag/api/README.md
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# API Directory
|
| 2 |
+
|
| 3 |
+
This directory contains the FastAPI backend for the UB-VSA project. It includes routes, schemas, and static assets for serving the API endpoints.
|
buffalo_rag/api/background_tasks.py
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from buffalo_rag.scraper.scraper import BuffaloScraper
|
| 2 |
+
from buffalo_rag.embeddings.chunker import DocumentChunker
|
| 3 |
+
from buffalo_rag.vector_store.db import VectorStore
|
| 4 |
+
from buffalo_rag.model.rag import BuffaloRAG
|
| 5 |
+
|
| 6 |
+
def run_scraper(seed_url: str, max_pages: int):
|
| 7 |
+
"""Run the web scraper in the background."""
|
| 8 |
+
scraper = BuffaloScraper(seed_url=seed_url)
|
| 9 |
+
scraper.scrape(max_pages=max_pages)
|
| 10 |
+
|
| 11 |
+
# After scraping, update the embeddings and index
|
| 12 |
+
chunker = DocumentChunker()
|
| 13 |
+
chunks = chunker.create_chunks()
|
| 14 |
+
chunker.create_embeddings(chunks)
|
| 15 |
+
|
| 16 |
+
# Reload the vector store
|
| 17 |
+
global vector_store
|
| 18 |
+
vector_store = VectorStore()
|
| 19 |
+
|
| 20 |
+
# Update the RAG model
|
| 21 |
+
global rag
|
| 22 |
+
rag = BuffaloRAG(vector_store=vector_store)
|
| 23 |
+
|
| 24 |
+
def refresh_index():
|
| 25 |
+
"""Refresh the vector index in the background."""
|
| 26 |
+
chunker = DocumentChunker()
|
| 27 |
+
chunks = chunker.create_chunks()
|
| 28 |
+
chunker.create_embeddings(chunks)
|
| 29 |
+
|
| 30 |
+
# Reload the vector store
|
| 31 |
+
global vector_store
|
| 32 |
+
vector_store = VectorStore()
|
| 33 |
+
|
| 34 |
+
# Update the RAG model
|
| 35 |
+
global rag
|
| 36 |
+
rag = BuffaloRAG(vector_store=vector_store)
|