forked from lancedb/lancedb
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathopenai.py
More file actions
140 lines (121 loc) · 5.29 KB
/
openai.py
File metadata and controls
140 lines (121 loc) · 5.29 KB
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
# Copyright (c) 2023. LanceDB Developers
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import json
import os
from functools import cached_property
from typing import Optional
import pyarrow as pa
from ..util import attempt_import_or_raise
from .base import Reranker
class OpenaiReranker(Reranker):
"""
Reranks the results using the OpenAI API.
WARNING: This is a prompt based reranker that uses chat model that is
not a dedicated reranker API. This should be treated as experimental.
Parameters
----------
model_name : str, default "gpt-4-turbo-preview"
The name of the cross encoder model to use.
column : str, default "text"
The name of the column to use as input to the cross encoder model.
return_score : str, default "relevance"
options are "relevance" or "all". Only "relevance" is supported for now.
api_key : str, default None
The API key to use. If None, will use the OPENAI_API_KEY environment variable.
"""
def __init__(
self,
model_name: str = "gpt-4-turbo-preview",
column: str = "text",
return_score="relevance",
api_key: Optional[str] = None,
):
super().__init__(return_score)
self.model_name = model_name
self.column = column
self.api_key = api_key
def _rerank(self, result_set: pa.Table, query: str):
docs = result_set[self.column].to_pylist()
response = self._client.chat.completions.create(
model=self.model_name,
response_format={"type": "json_object"},
temperature=0,
messages=[
{
"role": "system",
"content": "You are an expert relevance ranker. Given a list of\
documents and a query, your job is to determine the relevance\
each document is for answering the query. Your output is JSON,\
which is a list of documents. Each document has two fields,\
content and relevance_score. relevance_score is from 0.0 to\
1.0 indicating the relevance of the text to the given query.\
Make sure to include all documents in the response.",
},
{"role": "user", "content": f"Query: {query} Docs: {docs}"},
],
)
results = json.loads(response.choices[0].message.content)["documents"]
docs, scores = list(
zip(*[(result["content"], result["relevance_score"]) for result in results])
) # tuples
# replace the self.column column with the docs
result_set = result_set.drop(self.column)
result_set = result_set.append_column(
self.column, pa.array(docs, type=pa.string())
)
# add the scores
result_set = result_set.append_column(
"_relevance_score", pa.array(scores, type=pa.float32())
)
return result_set
def rerank_hybrid(
self,
query: str,
vector_results: pa.Table,
fts_results: pa.Table,
):
combined_results = self.merge_results(vector_results, fts_results)
combined_results = self._rerank(combined_results, query)
if self.score == "relevance":
combined_results = self._keep_relevance_score(combined_results)
elif self.score == "all":
raise NotImplementedError(
"OpenAI Reranker does not support score='all' yet"
)
combined_results = combined_results.sort_by(
[("_relevance_score", "descending")]
)
return combined_results
def rerank_vector(self, query: str, vector_results: pa.Table):
vector_results = self._rerank(vector_results, query)
if self.score == "relevance":
vector_results = vector_results.drop_columns(["_distance"])
vector_results = vector_results.sort_by([("_relevance_score", "descending")])
return vector_results
def rerank_fts(self, query: str, fts_results: pa.Table):
fts_results = self._rerank(fts_results, query)
if self.score == "relevance":
fts_results = fts_results.drop_columns(["_score"])
fts_results = fts_results.sort_by([("_relevance_score", "descending")])
return fts_results
@cached_property
def _client(self):
openai = attempt_import_or_raise(
"openai"
) # TODO: force version or handle versions < 1.0
if os.environ.get("OPENAI_API_KEY") is None and self.api_key is None:
raise ValueError(
"OPENAI_API_KEY not set. Either set it in your environment or \
pass it as `api_key` argument to the CohereReranker."
)
return openai.OpenAI(api_key=os.environ.get("OPENAI_API_KEY") or self.api_key)