Builder for creating auto-configured retrievers.
Provides factory method to create retrievers with dynamic configuration.
Source code in src/common/builders/retriever_builders.py
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 | class AutoRetrieverBuilder:
"""Builder for creating auto-configured retrievers.
Provides factory method to create retrievers with dynamic configuration.
"""
@staticmethod
def build(
vector_store: VectorStore,
embedding_model: BoundEmbeddingModel,
configuration: AutoRetrieverConfiguration,
llm: BoundAutoRetrieverLLM,
) -> CustomVectorIndexAutoRetriever:
"""Creates a configured auto-retriever instance.
Args:
vector_store: Vector storage backend.
embedding_model: Text embedding model.
configuration: Auto retriever parameters.
llm: Language model for metadata extraction.
Returns:
CustomVectorIndexAutoRetriever: Configured auto-retriever instance.
"""
index = VectorStoreIndex.from_vector_store(
vector_store=vector_store,
embed_model=embedding_model,
)
return CustomVectorIndexAutoRetriever(
index=index,
similarity_top_k=configuration.similarity_top_k,
llm=llm,
vector_store_info=VectorStoreInfo(
content_info="Knowledge base of FELD M company used for retrieval process in RAG system.",
metadata_info=[
MetadataInfo(
name="creation_date",
type="date",
description=(
"Date of creation of the chunk's document"
),
),
MetadataInfo(
name="last_update_date",
type="date",
description=(
"Date of the last update of the chunk's document."
),
),
],
),
)
|
build(vector_store, embedding_model, configuration, llm)
staticmethod
Creates a configured auto-retriever instance.
| Parameters: |
-
vector_store
(VectorStore)
–
-
embedding_model
(BoundEmbeddingModel)
–
-
configuration
(AutoRetrieverConfiguration)
–
Auto retriever parameters.
-
llm
(BoundAutoRetrieverLLM)
–
Language model for metadata extraction.
|
| Returns: |
-
CustomVectorIndexAutoRetriever( CustomVectorIndexAutoRetriever
) –
Configured auto-retriever instance.
|
Source code in src/common/builders/retriever_builders.py
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 | @staticmethod
def build(
vector_store: VectorStore,
embedding_model: BoundEmbeddingModel,
configuration: AutoRetrieverConfiguration,
llm: BoundAutoRetrieverLLM,
) -> CustomVectorIndexAutoRetriever:
"""Creates a configured auto-retriever instance.
Args:
vector_store: Vector storage backend.
embedding_model: Text embedding model.
configuration: Auto retriever parameters.
llm: Language model for metadata extraction.
Returns:
CustomVectorIndexAutoRetriever: Configured auto-retriever instance.
"""
index = VectorStoreIndex.from_vector_store(
vector_store=vector_store,
embed_model=embedding_model,
)
return CustomVectorIndexAutoRetriever(
index=index,
similarity_top_k=configuration.similarity_top_k,
llm=llm,
vector_store_info=VectorStoreInfo(
content_info="Knowledge base of FELD M company used for retrieval process in RAG system.",
metadata_info=[
MetadataInfo(
name="creation_date",
type="date",
description=(
"Date of creation of the chunk's document"
),
),
MetadataInfo(
name="last_update_date",
type="date",
description=(
"Date of the last update of the chunk's document."
),
),
],
),
)
|