Spaces:
Sleeping
Sleeping
Update src/vector_db.py
Browse files- src/vector_db.py +7 -0
src/vector_db.py
CHANGED
@@ -6,6 +6,11 @@ import pyarrow as pa
|
|
6 |
import pandas as pd
|
7 |
import numpy as np
|
8 |
import tqdm
|
|
|
|
|
|
|
|
|
|
|
9 |
|
10 |
class VectorDB:
|
11 |
|
@@ -90,8 +95,10 @@ class VectorDB:
|
|
90 |
def retrieve_prefiltered_hits(self, query, k):
|
91 |
|
92 |
query_vec = self.retriever.encode(query)
|
|
|
93 |
documents = self.table.search(query_vec, vector_column_name=self.vector_column).limit(k).to_list()
|
94 |
names = [doc[self.name_column] for doc in documents]
|
95 |
descriptions = [doc[self.description_column] for doc in documents]
|
|
|
96 |
|
97 |
return names, descriptions
|
|
|
6 |
import pandas as pd
|
7 |
import numpy as np
|
8 |
import tqdm
|
9 |
+
import logging
|
10 |
+
|
11 |
+
logger = logging.getLogger(__name__)
|
12 |
+
logger.setLevel(logging.DEBUG)
|
13 |
+
|
14 |
|
15 |
class VectorDB:
|
16 |
|
|
|
95 |
def retrieve_prefiltered_hits(self, query, k):
|
96 |
|
97 |
query_vec = self.retriever.encode(query)
|
98 |
+
logger.info('encoded')
|
99 |
documents = self.table.search(query_vec, vector_column_name=self.vector_column).limit(k).to_list()
|
100 |
names = [doc[self.name_column] for doc in documents]
|
101 |
descriptions = [doc[self.description_column] for doc in documents]
|
102 |
+
logger.info('done topK lookup')
|
103 |
|
104 |
return names, descriptions
|