Update app.py
Browse files
app.py
CHANGED
|
@@ -65,21 +65,15 @@ def get_url_content(url):
|
|
| 65 |
return None
|
| 66 |
|
| 67 |
|
| 68 |
-
|
| 69 |
@st.cache_resource
|
| 70 |
def get_retriever(urls):
|
| 71 |
-
documents = DocumentArray([
|
| 72 |
|
| 73 |
-
# Load the model and encode document texts directly
|
| 74 |
model = SentenceTransformer('all-MiniLM-L6-v2')
|
| 75 |
-
documents.embeddings = model.encode(documents
|
| 76 |
-
|
| 77 |
-
# Applying HNSW for efficient similarity search
|
| 78 |
-
documents.embeddings.faiss.ann['HNSW'] = {'nlinks': 16}
|
| 79 |
|
| 80 |
return documents
|
| 81 |
|
| 82 |
-
|
| 83 |
@st.cache_resource
|
| 84 |
def create_chain(_retriever):
|
| 85 |
# A stream handler to direct streaming output on the chat screen.
|
|
|
|
| 65 |
return None
|
| 66 |
|
| 67 |
|
|
|
|
| 68 |
@st.cache_resource
|
| 69 |
def get_retriever(urls):
|
| 70 |
+
documents = DocumentArray([doc for url in urls if (doc := get_url_content(url)) is not None])
|
| 71 |
|
|
|
|
| 72 |
model = SentenceTransformer('all-MiniLM-L6-v2')
|
| 73 |
+
documents.embeddings = model.encode([doc.text for doc in documents], show_progress_bar=True)
|
|
|
|
|
|
|
|
|
|
| 74 |
|
| 75 |
return documents
|
| 76 |
|
|
|
|
| 77 |
@st.cache_resource
|
| 78 |
def create_chain(_retriever):
|
| 79 |
# A stream handler to direct streaming output on the chat screen.
|