Update app.py

app.py

@@ -74,10 +74,11 @@ if uploaded_file is not None:
     pdf_vectors = embedding.encode(texts)
     vector_count = len(documents)
     example_data_generator = map(lambda i: (f'id-{i}', pdf_vectors[i], {"text": texts[i]}), range(vector_count))
-    if len(index.describe_index_stats()['namespaces'])!=0:
-        index.delete(delete_all=True)
+    #if len(index.describe_index_stats()['namespaces'])!=0:
+    #    index.delete(delete_all=True)
     for ids_vectors_chunk in chunks(example_data_generator, batch_size=100):
         index.upsert(vectors=ids_vectors_chunk)
+        index.upsert(vectors=ids_vectors_chunk,namespace='ns1')
 
 # Search query related context
 sample_query = st.text_input("Stellen Sie eine Frage zu dem PDF: (Ask a question related to the PDF:)")