Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit58f82be

Browse files
authored
Updated docs (#1542)
1 parent638cf7a commit58f82be

File tree

3 files changed

+250
-207
lines changed

3 files changed

+250
-207
lines changed

‎pgml-cms/docs/api/client-sdk/search.md

Lines changed: 116 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -641,3 +641,119 @@ char **results = pgml_collectionc_vector_search(collection, "{\
641641
{% endtabs %}
642642

643643
The above query would filter out all documents that do not have a key`special` with a value`True` or (have a key`user_id` equal to 1 and a key`user_score` less than 100).
644+
645+
##**Re-ranking**
646+
647+
Vector search results can be reranked in the same query they are retrieved in. To enable this, provide the`rerank` key.
648+
649+
{% tabs %}
650+
{% tab title="JavaScript" %}
651+
```javascript
652+
constresults=awaitcollection.vector_search(
653+
{
654+
query: {
655+
fields: {
656+
body: {
657+
query:"What is the best database?", parameters: {
658+
prompt:
659+
"Represent this sentence for searching relevant passages:",
660+
}
661+
},
662+
},
663+
},
664+
rerank: {
665+
model:"mixedbread-ai/mxbai-rerank-base-v1",
666+
query:"What is the best database?",
667+
num_documents_to_rerank:100,
668+
},
669+
limit:5,
670+
},
671+
pipeline,
672+
);
673+
```
674+
{% endtab %}
675+
676+
{% tab title="Python" %}
677+
```python
678+
results=await collection.vector_search(
679+
{
680+
"query": {
681+
"fields": {
682+
"body": {
683+
"query":"What is the best database?",
684+
"parameters": {
685+
"prompt":"Represent this sentence for searching relevant passages:",
686+
},
687+
},
688+
},
689+
},
690+
"rerank": {
691+
"model":"mixedbread-ai/mxbai-rerank-base-v1",
692+
"query":"What is the best database",
693+
"num_documents_to_rerank":100,
694+
},
695+
"limit":5,
696+
},
697+
pipeline,
698+
)
699+
```
700+
{% endtab %}
701+
702+
{% tab title="Rust" %}
703+
```rust
704+
letresults=collection
705+
.vector_search(
706+
serde_json::json!({
707+
"query": {
708+
"fields": {
709+
"body": {
710+
"query":"What is the best database?",
711+
"parameters": {
712+
"prompt":"Represent this sentence for searching relevant passages:",
713+
},
714+
},
715+
},
716+
},
717+
"rerank": {
718+
"model":"mixedbread-ai/mxbai-rerank-base-v1",
719+
"query":"What is the best database",
720+
"num_documents_to_rerank":100,
721+
},
722+
"limit":5,
723+
})
724+
.into(),
725+
&mutpipeline,
726+
)
727+
.await?;
728+
```
729+
{% endtab %}
730+
731+
{% tab title="C" %}
732+
```cpp
733+
r_size =0;
734+
char **results = pgml_collectionc_vector_search(collection,"{\
735+
\"query\": {\
736+
\"fields\": {\
737+
\"body\": {\
738+
\"query\":\"What is the best database?\",\
739+
\"parameters\": {\
740+
\"prompt\":\"Represent this sentence for searching relevant passages:\"\
741+
}\
742+
}\
743+
}\
744+
},\
745+
\"rerank\": {\
746+
\"model\":\"mixedbread-ai/mxbai-rerank-base-v1\",\
747+
\"query\":\"What is the best database\",\
748+
\"num_documents_to_rerank\": 100\
749+
},\
750+
\"limit\": 5\
751+
}",
752+
pipeline, &r_size);
753+
```
754+
{% endtab %}
755+
{% endtabs %}
756+
757+
This query will first get the top 100 documents from the initial vector search and then rerank them using the`mixedbread-ai/mxbai-rerank-base-v1` cross-encoder.
758+
759+
You can specify the number of documents to rerank with the`num_documents_to_rerank` parameter. The query returns the top`limit` results after re-ranking.

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp