基于 Doc ID 相似度检索

import tcvectordb
from tcvectordb.model.enum import FieldType, IndexType, MetricType, ReadConsistency
from tcvectordb.model.index import Index, VectorIndex, FilterIndex, HNSWParams
from tcvectordb.model.document import Document, Filter, SearchParams

#create a database client object
client = tcvectordb.VectorDBClient(url='http://10.0.X.X', username='root', key='eC4bLRy2va******************************', read_consistency=ReadConsistency.EVENTUAL_CONSISTENCY, timeout=30)

db = client.database('db-test')
coll = db.collection('book-vector')
# search by document id
# document_ids 指定了检索文档 id
# filter 指定了过滤条件
# params 指定索引类型对应的查询参数，HNSW 类型需要设置 ef，指定查询的遍历范围;IVF 系列需要设置 nprobe,指定查询的单位数量
# retrieve_vector 指定是否输出向量字段
# limit 指定返回最相似的 Top K 条结果。如果插入的数据不足 K 条，则返回实际插入的 Document 数量。
# output_fields 指定输出字段   
doc_lists = coll.searchById(
                 document_ids=['0001','0002'],
                 filter=Filter(Filter.In("bookName",["三国演义", "西游记"])),
                 params=SearchParams(ef=200),
                 limit=3,
                 retrieve_vector=True,
                 output_fields=['bookName','author']
             )

for i, docs in enumerate(doc_lists):
                print(i)
                for doc in docs:
                        print(doc)

0
{'id': '0001', 'vector': [0.21230000257492065, 0.23000000417232513, 0.21299999952316284], 'score': 1.0000001192092896, 'author': '罗贯中', 'bookName': '三国演义'}
{'id': '0002', 'vector': [0.21230000257492065, 0.2199999988079071, 0.21299999952316284], 'score': 0.9997729659080505, 'author': '吴承恩', 'bookName': '西游记'}
1
{'id': '0002', 'vector': [0.21230000257492065, 0.2199999988079071, 0.21299999952316284], 'score': 0.9997580051422119, 'author': '吴承恩', 'bookName': '西游记'}
{'id': '0001', 'vector': [0.21230000257492065, 0.23000000417232513, 0.21299999952316284], 'score': 0.9990617632865906, 'bookName': '三国演义', 'author': '罗贯中'}