mirror of
https://github.com/qdrant/qdrant.git
synced 2026-07-27 05:01:08 -05:00
7caf415958
* [edge]dbsf fusion example along with hybrid search * Reduce query limits in hybrid search example
75 lines
1.9 KiB
Python
75 lines
1.9 KiB
Python
#!/usr/bin/env python3
|
|
"""Hybrid dense + BM25 sparse search with DBSF fusion."""
|
|
|
|
import os
|
|
import shutil
|
|
from pathlib import Path
|
|
|
|
from qdrant_edge import (
|
|
Bm25, Bm25Config, Distance,
|
|
EdgeConfig, EdgeShard, EdgeSparseVectorParams,
|
|
EdgeVectorParams, Fusion, Modifier, Point,
|
|
Prefetch, Query, QueryRequest, UpdateOperation,
|
|
)
|
|
|
|
|
|
DATA_DIR = Path(__file__).parent.parent.parent / "data"
|
|
TMP_DIR = DATA_DIR / "tmp"
|
|
path = TMP_DIR / "qdrant_edge_hybrid_search_dbsf"
|
|
shutil.rmtree(path, ignore_errors=True)
|
|
os.makedirs(path)
|
|
|
|
config = EdgeConfig(
|
|
vectors={"dense": EdgeVectorParams(size=4, distance=Distance.Cosine)},
|
|
sparse_vectors={"sparse": EdgeSparseVectorParams(modifier=Modifier.Idf)},
|
|
)
|
|
shard = EdgeShard.create(path, config)
|
|
|
|
bm25 = Bm25(Bm25Config(language="english"))
|
|
|
|
documents = [
|
|
(1, "red apple fresh fruit", [0.90, 0.10, 0.10, 0.05]),
|
|
(2, "green apple tart fruit", [0.85, 0.15, 0.12, 0.10]),
|
|
(3, "fast red sports car", [0.10, 0.90, 0.15, 0.05]),
|
|
(4, "electric vehicle charging", [0.12, 0.80, 0.20, 0.20]),
|
|
(5, "fresh fruit market", [0.88, 0.12, 0.18, 0.05]),
|
|
]
|
|
|
|
|
|
|
|
shard.update(UpdateOperation.upsert_points([
|
|
Point(
|
|
point_id,
|
|
{
|
|
"dense": dense_vector,
|
|
"sparse": bm25.embed_document(text),
|
|
},
|
|
{"text": text},
|
|
)
|
|
for point_id, text, dense_vector in documents
|
|
]))
|
|
shard.optimize()
|
|
|
|
dense_query = [0.90, 0.10, 0.10, 0.05]
|
|
sparse_query = bm25.embed_query("fresh apple fruit")
|
|
|
|
results = shard.query(QueryRequest(
|
|
prefetches=[
|
|
Prefetch(
|
|
query=Query.Nearest(dense_query, using="dense"),
|
|
limit=3,
|
|
),
|
|
Prefetch(
|
|
query=Query.Nearest(sparse_query, using="sparse"),
|
|
limit=3,
|
|
),
|
|
],
|
|
query=Fusion.Dbsf(),
|
|
limit=3,
|
|
with_payload=True,
|
|
))
|
|
|
|
print("=== Hybrid Search with DBSF ===")
|
|
for point in results:
|
|
print(f"id={point.id} score={point.score:.4f} text={point.payload['text']}")
|