基于原始文本相似度检索

import tcvectordb
from tcvectordb.model.enum import FieldType, IndexType, MetricType, EmbeddingModel, 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-emb')

# search by text
# embeddingItems 指定了检索的文本
# 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

# 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.searchByText(
                 embeddingItems=['天下大势，分久必合，合久必分'],
                 filter=Filter(Filter.In("bookName",["三国演义", "西游记"])),
                 params=SearchParams(ef=200),
                 limit=3,
                 retrieve_vector=False,
                 output_fields=['bookName','author']
             )           
# printf             
for i, docs in enumerate(doc_lists.get("documents")):
                print(i)
                for doc in docs:
                        print(doc)
              


0
{'id': '0001', 'score': 0.9792741537094116, 'author': '罗贯中', 'bookName': '三国演义'}
{'id': '0002', 'score': 0.7909858226776123, 'bookName': '西游记', 'author': '吴承恩'}