-
Notifications
You must be signed in to change notification settings - Fork 303
/
AzureSearchHelper.py
162 lines (154 loc) · 5.3 KB
/
AzureSearchHelper.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
from langchain.vectorstores.azuresearch import AzureSearch
from azure.search.documents.indexes.models import (
SearchableField,
SearchField,
SearchFieldDataType,
SimpleField,
)
from .LLMHelper import LLMHelper
from .EnvHelper import EnvHelper
class AzureSearchHelper:
_search_dimension: int | None = None
def __init__(self):
self.llm_helper = LLMHelper()
self.env_helper = EnvHelper()
@property
def search_dimensions(self) -> int:
if AzureSearchHelper._search_dimension is None:
AzureSearchHelper._search_dimension = len(
self.llm_helper.get_embedding_model().embed_query("Text")
)
return AzureSearchHelper._search_dimension
def get_vector_store(self):
fields = [
SimpleField(
name="id",
type=SearchFieldDataType.String,
key=True,
filterable=True,
),
SearchableField(
name="content",
type=SearchFieldDataType.String,
),
SearchField(
name="content_vector",
type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
searchable=True,
vector_search_dimensions=self.search_dimensions,
vector_search_profile_name="myHnswProfile",
),
SearchableField(
name="metadata",
type=SearchFieldDataType.String,
),
SearchableField(
name="title",
type=SearchFieldDataType.String,
facetable=True,
filterable=True,
),
SearchableField(
name="source",
type=SearchFieldDataType.String,
filterable=True,
),
SimpleField(
name="chunk",
type=SearchFieldDataType.Int32,
filterable=True,
),
SimpleField(
name="offset",
type=SearchFieldDataType.Int32,
filterable=True,
),
]
return AzureSearch(
azure_search_endpoint=self.env_helper.AZURE_SEARCH_SERVICE,
azure_search_key=(
self.env_helper.AZURE_SEARCH_KEY
if self.env_helper.AZURE_AUTH_TYPE == "keys"
else None
),
index_name=self.env_helper.AZURE_SEARCH_INDEX,
embedding_function=self.llm_helper.get_embedding_model().embed_query,
fields=fields,
search_type=(
"semantic_hybrid"
if self.env_helper.AZURE_SEARCH_USE_SEMANTIC_SEARCH
else "hybrid"
),
semantic_configuration_name=self.env_helper.AZURE_SEARCH_SEMANTIC_SEARCH_CONFIG,
user_agent="langchain chatwithyourdata-sa",
)
def get_conversation_logger(self):
fields = [
SimpleField(
name="id",
type=SearchFieldDataType.String,
key=True,
filterable=True,
),
SimpleField(
name="conversation_id",
type=SearchFieldDataType.String,
filterable=True,
facetable=True,
),
SearchableField(
name="content",
type=SearchFieldDataType.String,
),
SearchField(
name="content_vector",
type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
searchable=True,
vector_search_dimensions=self.search_dimensions,
vector_search_profile_name="myHnswProfile",
),
SearchableField(
name="metadata",
type=SearchFieldDataType.String,
),
SimpleField(
name="type",
type=SearchFieldDataType.String,
facetable=True,
filterable=True,
),
SimpleField(
name="user_id",
type=SearchFieldDataType.String,
filterable=True,
facetable=True,
),
SimpleField(
name="sources",
type=SearchFieldDataType.Collection(SearchFieldDataType.String),
filterable=True,
facetable=True,
),
SimpleField(
name="created_at",
type=SearchFieldDataType.DateTimeOffset,
filterable=True,
),
SimpleField(
name="updated_at",
type=SearchFieldDataType.DateTimeOffset,
filterable=True,
),
]
return AzureSearch(
azure_search_endpoint=self.env_helper.AZURE_SEARCH_SERVICE,
azure_search_key=(
self.env_helper.AZURE_SEARCH_KEY
if self.env_helper.AZURE_AUTH_TYPE == "keys"
else None
),
index_name=self.env_helper.AZURE_SEARCH_CONVERSATIONS_LOG_INDEX,
embedding_function=self.llm_helper.get_embedding_model().embed_query,
fields=fields,
user_agent="langchain chatwithyourdata-sa",
)