krzysztof/landing_page
1
0
Fork 0
mirror of https://github.com/qdrant/landing_page.git synced 2026-09-30 00:18:32 +02:00
Code Issues Packages Projects Releases Wiki Activity

add document_id payload index and tune upload_points

Browse Source 
Adds the missing keyword index on document_id so grouping works under
strict mode (Cloud default), and tunes the upload_points call to
batch_size=256, parallel=2 for faster ingestion. Mirrors the notebook
in qdrant/examples#103.
This commit is contained in:
Dylan Couzon
2026-05-11 22:05:24 -04:00
parent 5502b91064
commit 31a3c88379
1 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
qdrant-landing/content/documentation/tutorials-search-engineering/multi-representation-search.md
+7 -2
View File
@@ -99,7 +99,12 @@ client.create_collection(
},
)
# Index the 'tags' payload as keyword so we can filter on category at query time.
# Index 'document_id' so the Query API can group by it; index 'tags' so we can filter on category.
client.create_payload_index(
collection_name="arxiv_multi_repr",
field_name="document_id",
field_schema=models.PayloadSchemaType.KEYWORD,
)
client.create_payload_index(
collection_name="arxiv_multi_repr",
field_name="tags",
@@ -159,7 +164,7 @@ for paper in papers:
},
))
client.upload_points(collection_name="arxiv_multi_repr", points=points, batch_size=64)
client.upload_points(collection_name="arxiv_multi_repr", points=points, batch_size=256, parallel=2)
```
After the upload completes, opening any point in the Qdrant Cloud UI shows all four named vectors attached to one chunk. `dense_chunk` carries the chunk's own embedding, while `dense_title`, `dense_abstract`, and `sparse_title` are the same across every chunk of this paper.