Skip to content

Bagel

BagelVectorStore #

Bases: VectorStore

Vector store for Bagel.

Examples:

pip install llama-index-vector-stores-bagel

from llama_index.core import VectorStoreIndex, StorageContext
from llama_index.vector_stores.bagel import BagelVectorStore

import bagel
from bagel import Settings

server_settings = Settings(
    bagel_api_impl="rest", bagel_server_host="api.bageldb.ai"
)

client = bagel.Client(server_settings)

collection = client.get_or_create_cluster("testing_embeddings")
vector_store = BagelVectorStore(collection=collection)
Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-bagel/llama_index/vector_stores/bagel/base.py 
 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
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
class BagelVectorStore(VectorStore):
    """Vector store for Bagel.

    Examples:
        `pip install llama-index-vector-stores-bagel`

        ```python
        from llama_index.core import VectorStoreIndex, StorageContext
        from llama_index.vector_stores.bagel import BagelVectorStore

        import bagel
        from bagel import Settings

        server_settings = Settings(
            bagel_api_impl="rest", bagel_server_host="api.bageldb.ai"
        )

        client = bagel.Client(server_settings)

        collection = client.get_or_create_cluster("testing_embeddings")
        vector_store = BagelVectorStore(collection=collection)
        ```
    """

    # support for Bagel specific parameters
    stores_text: bool = True
    flat_metadata: bool = True

    def __init__(self, collection: Any, **kwargs: Any) -> None:
        """
        Initialize BagelVectorStore.

        Args:
            collection: Bagel collection.
            **kwargs: Additional arguments.
        """
        try:
            from bagel.api.Cluster import Cluster
        except ImportError:
            raise ImportError("Bagel is not installed. Please install bagel.")

        if not isinstance(collection, Cluster):
            raise ValueError("Collection must be a bagel Cluster.")

        self._collection = collection

    def add(self, nodes: List[BaseNode], **add_kwargs: Any) -> List[str]:
        """
        Add a list of nodes with embeddings to the vector store.

        Args:
            nodes: List of nodes with embeddings.
            kwargs: Additional arguments.

        Returns:
            List of document ids.
        """
        if not self._collection:
            raise ValueError("collection not set")

        ids = []
        embeddings = []
        metadatas = []
        documents = []

        for node in nodes:
            ids.append(node.node_id)
            embeddings.append(node.get_embedding())
            metadatas.append(
                node_to_metadata_dict(
                    node,
                    remove_text=True,
                    flat_metadata=self.flat_metadata,
                )
            )
            documents.append(node.get_content(metadata_mode=MetadataMode.NONE) or "")

        self._collection.add(
            ids=ids, embeddings=embeddings, metadatas=metadatas, documents=documents
        )

        return ids

    def delete(self, ref_doc_id: str, **kwargs: Any) -> None:
        """
        Delete a document from the vector store.

        Args:
            ref_doc_id: Reference document id.
            kwargs: Additional arguments.
        """
        if not self._collection:
            raise ValueError("collection not set")

        results = self._collection.get(where={"doc_id": ref_doc_id})
        if results and "ids" in results:
            self._collection.delete(ids=results["ids"])

    @property
    def client(self) -> Any:
        """
        Get the Bagel cluster.
        """
        return self._collection

    def query(self, query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult:
        """
        Query the vector store.

        Args:
            query: Query to run.
            kwargs: Additional arguments.

        Returns:
            Query result.
        """
        if not self._collection:
            raise ValueError("collection not set")

        if query.filters is not None:
            if "where" in kwargs:
                raise ValueError("Cannot specify both filters and where")
            where = _to_bagel_filter(query.filters)
        else:
            where = kwargs.get("where", {})

        results = self._collection.find(
            query_embeddings=query.query_embedding,
            where=where,
            n_results=query.similarity_top_k,
            **kwargs,
        )

        logger.debug(f"query results: {results}")

        nodes = []
        similarities = []
        ids = []

        for node_id, text, metadata, distance in zip(
            results["ids"][0],
            results["documents"][0],
            results["metadatas"][0],
            results["distances"][0],
        ):
            try:
                node = metadata_dict_to_node(metadata)
                node.set_content(text)
            except Exception:
                # NOTE: deprecated legacy logic for backward compatibility
                metadata, node_info, relationships = legacy_metadata_dict_to_node(
                    metadata
                )

                node = TextNode(
                    text=text,
                    id_=node_id,
                    metadata=metadata,
                    start_char_idx=node_info.get("start", None),
                    end_char_idx=node_info.get("end", None),
                    relationships=relationships,
                )

            nodes.append(node)
            similarities.append(1.0 - math.exp(-distance))
            ids.append(node_id)

            logger.debug(f"node: {node}")
            logger.debug(f"similarity: {1.0 - math.exp(-distance)}")
            logger.debug(f"id: {node_id}")

        return VectorStoreQueryResult(nodes=nodes, similarities=similarities, ids=ids)

client property #

client: Any

Get the Bagel cluster.

add #

add(nodes: List[BaseNode], **add_kwargs: Any) -> List[str]

Add a list of nodes with embeddings to the vector store.

Parameters:

Name Type Description Default
nodes List[BaseNode]

List of nodes with embeddings.

 required
kwargs

Additional arguments.

 required

Returns:

Type Description
List[str]

List of document ids.
Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-bagel/llama_index/vector_stores/bagel/base.py 
 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
def add(self, nodes: List[BaseNode], **add_kwargs: Any) -> List[str]:
    """
    Add a list of nodes with embeddings to the vector store.

    Args:
        nodes: List of nodes with embeddings.
        kwargs: Additional arguments.

    Returns:
        List of document ids.
    """
    if not self._collection:
        raise ValueError("collection not set")

    ids = []
    embeddings = []
    metadatas = []
    documents = []

    for node in nodes:
        ids.append(node.node_id)
        embeddings.append(node.get_embedding())
        metadatas.append(
            node_to_metadata_dict(
                node,
                remove_text=True,
                flat_metadata=self.flat_metadata,
            )
        )
        documents.append(node.get_content(metadata_mode=MetadataMode.NONE) or "")

    self._collection.add(
        ids=ids, embeddings=embeddings, metadatas=metadatas, documents=documents
    )

    return ids

delete #

delete(ref_doc_id: str, **kwargs: Any) -> None

Delete a document from the vector store.

Parameters:

Name Type Description Default
ref_doc_id str

Reference document id.

 required
kwargs Any

Additional arguments.

 {}
Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-bagel/llama_index/vector_stores/bagel/base.py 
114
115
116
117
118
119
120
121
122
123
124
125
126
127
def delete(self, ref_doc_id: str, **kwargs: Any) -> None:
    """
    Delete a document from the vector store.

    Args:
        ref_doc_id: Reference document id.
        kwargs: Additional arguments.
    """
    if not self._collection:
        raise ValueError("collection not set")

    results = self._collection.get(where={"doc_id": ref_doc_id})
    if results and "ids" in results:
        self._collection.delete(ids=results["ids"])

query #

query(query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult

Query the vector store.

Parameters:

Name Type Description Default
query VectorStoreQuery

Query to run.

 required
kwargs Any

Additional arguments.

 {}

Returns:

Type Description
VectorStoreQueryResult

Query result.
Source code in llama-index-integrations/vector_stores/llama-index-vector-stores-bagel/llama_index/vector_stores/bagel/base.py 
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
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
def query(self, query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult:
    """
    Query the vector store.

    Args:
        query: Query to run.
        kwargs: Additional arguments.

    Returns:
        Query result.
    """
    if not self._collection:
        raise ValueError("collection not set")

    if query.filters is not None:
        if "where" in kwargs:
            raise ValueError("Cannot specify both filters and where")
        where = _to_bagel_filter(query.filters)
    else:
        where = kwargs.get("where", {})

    results = self._collection.find(
        query_embeddings=query.query_embedding,
        where=where,
        n_results=query.similarity_top_k,
        **kwargs,
    )

    logger.debug(f"query results: {results}")

    nodes = []
    similarities = []
    ids = []

    for node_id, text, metadata, distance in zip(
        results["ids"][0],
        results["documents"][0],
        results["metadatas"][0],
        results["distances"][0],
    ):
        try:
            node = metadata_dict_to_node(metadata)
            node.set_content(text)
        except Exception:
            # NOTE: deprecated legacy logic for backward compatibility
            metadata, node_info, relationships = legacy_metadata_dict_to_node(
                metadata
            )

            node = TextNode(
                text=text,
                id_=node_id,
                metadata=metadata,
                start_char_idx=node_info.get("start", None),
                end_char_idx=node_info.get("end", None),
                relationships=relationships,
            )

        nodes.append(node)
        similarities.append(1.0 - math.exp(-distance))
        ids.append(node_id)

        logger.debug(f"node: {node}")
        logger.debug(f"similarity: {1.0 - math.exp(-distance)}")
        logger.debug(f"id: {node_id}")

    return VectorStoreQueryResult(nodes=nodes, similarities=similarities, ids=ids)