Spaces:
Build error
Build error
File size: 12,482 Bytes
60e3a80 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 |
from typing import TYPE_CHECKING, Optional, Union
import numpy as np
from chromadb.api.models.CollectionCommon import CollectionCommon
from chromadb.api.types import (
URI,
CollectionMetadata,
Embedding,
PyEmbedding,
Include,
Metadata,
Document,
Image,
Where,
IDs,
GetResult,
QueryResult,
ID,
OneOrMany,
WhereDocument,
)
import logging
logger = logging.getLogger(__name__)
if TYPE_CHECKING:
from chromadb.api import ServerAPI # noqa: F401
class Collection(CollectionCommon["ServerAPI"]):
def count(self) -> int:
"""The total number of embeddings added to the database
Returns:
int: The total number of embeddings added to the database
"""
return self._client._count(collection_id=self.id)
def add(
self,
ids: OneOrMany[ID],
embeddings: Optional[ # type: ignore[type-arg]
Union[
OneOrMany[Embedding],
OneOrMany[PyEmbedding],
]
] = None,
metadatas: Optional[OneOrMany[Metadata]] = None,
documents: Optional[OneOrMany[Document]] = None,
images: Optional[OneOrMany[Image]] = None,
uris: Optional[OneOrMany[URI]] = None,
) -> None:
"""Add embeddings to the data store.
Args:
ids: The ids of the embeddings you wish to add
embeddings: The embeddings to add. If None, embeddings will be computed based on the documents or images using the embedding_function set for the Collection. Optional.
metadatas: The metadata to associate with the embeddings. When querying, you can filter on this metadata. Optional.
documents: The documents to associate with the embeddings. Optional.
images: The images to associate with the embeddings. Optional.
uris: The uris of the images to associate with the embeddings. Optional.
Returns:
None
Raises:
ValueError: If you don't provide either embeddings or documents
ValueError: If the length of ids, embeddings, metadatas, or documents don't match
ValueError: If you don't provide an embedding function and don't provide embeddings
ValueError: If you provide both embeddings and documents
ValueError: If you provide an id that already exists
"""
(
ids,
embeddings,
metadatas,
documents,
uris,
) = self._validate_and_prepare_embedding_set(
ids, embeddings, metadatas, documents, images, uris
)
self._client._add(ids, self.id, embeddings, metadatas, documents, uris)
def get(
self,
ids: Optional[OneOrMany[ID]] = None,
where: Optional[Where] = None,
limit: Optional[int] = None,
offset: Optional[int] = None,
where_document: Optional[WhereDocument] = None,
include: Include = ["metadatas", "documents"],
) -> GetResult:
"""Get embeddings and their associate data from the data store. If no ids or where filter is provided returns
all embeddings up to limit starting at offset.
Args:
ids: The ids of the embeddings to get. Optional.
where: A Where type dict used to filter results by. E.g. `{"$and": [{"color" : "red"}, {"price": {"$gte": 4.20}}]}`. Optional.
limit: The number of documents to return. Optional.
offset: The offset to start returning results from. Useful for paging results with limit. Optional.
where_document: A WhereDocument type dict used to filter by the documents. E.g. `{$contains: {"text": "hello"}}`. Optional.
include: A list of what to include in the results. Can contain `"embeddings"`, `"metadatas"`, `"documents"`. Ids are always included. Defaults to `["metadatas", "documents"]`. Optional.
Returns:
GetResult: A GetResult object containing the results.
"""
(
valid_ids,
valid_where,
valid_where_document,
valid_include,
) = self._validate_and_prepare_get_request(ids, where, where_document, include)
get_results = self._client._get(
self.id,
valid_ids,
valid_where,
None,
limit,
offset,
where_document=valid_where_document,
include=valid_include,
)
return self._transform_get_response(get_results, include)
def peek(self, limit: int = 10) -> GetResult:
"""Get the first few results in the database up to limit
Args:
limit: The number of results to return.
Returns:
GetResult: A GetResult object containing the results.
"""
return self._transform_peek_response(self._client._peek(self.id, limit))
def query(
self,
query_embeddings: Optional[ # type: ignore[type-arg]
Union[
OneOrMany[Embedding],
OneOrMany[PyEmbedding],
]
] = None,
query_texts: Optional[OneOrMany[Document]] = None,
query_images: Optional[OneOrMany[Image]] = None,
query_uris: Optional[OneOrMany[URI]] = None,
n_results: int = 10,
where: Optional[Where] = None,
where_document: Optional[WhereDocument] = None,
include: Include = ["metadatas", "documents", "distances"],
) -> QueryResult:
"""Get the n_results nearest neighbor embeddings for provided query_embeddings or query_texts.
Args:
query_embeddings: The embeddings to get the closes neighbors of. Optional.
query_texts: The document texts to get the closes neighbors of. Optional.
query_images: The images to get the closes neighbors of. Optional.
query_uris: The URIs to be used with data loader. Optional.
n_results: The number of neighbors to return for each query_embedding or query_texts. Optional.
where: A Where type dict used to filter results by. E.g. `{"$and": [{"color" : "red"}, {"price": {"$gte": 4.20}}]}`. Optional.
where_document: A WhereDocument type dict used to filter by the documents. E.g. `{$contains: {"text": "hello"}}`. Optional.
include: A list of what to include in the results. Can contain `"embeddings"`, `"metadatas"`, `"documents"`, `"distances"`. Ids are always included. Defaults to `["metadatas", "documents", "distances"]`. Optional.
Returns:
QueryResult: A QueryResult object containing the results.
Raises:
ValueError: If you don't provide either query_embeddings, query_texts, or query_images
ValueError: If you provide both query_embeddings and query_texts
ValueError: If you provide both query_embeddings and query_images
ValueError: If you provide both query_texts and query_images
"""
(
valid_query_embeddings,
valid_n_results,
valid_where,
valid_where_document,
) = self._validate_and_prepare_query_request(
query_embeddings,
query_texts,
query_images,
query_uris,
n_results,
where,
where_document,
include,
)
query_results = self._client._query(
collection_id=self.id,
query_embeddings=valid_query_embeddings,
n_results=valid_n_results,
where=valid_where,
where_document=valid_where_document,
include=include,
)
return self._transform_query_response(query_results, include)
def modify(
self, name: Optional[str] = None, metadata: Optional[CollectionMetadata] = None
) -> None:
"""Modify the collection name or metadata
Args:
name: The updated name for the collection. Optional.
metadata: The updated metadata for the collection. Optional.
Returns:
None
"""
self._validate_modify_request(metadata)
# Note there is a race condition here where the metadata can be updated
# but another thread sees the cached local metadata.
# TODO: fixme
self._client._modify(id=self.id, new_name=name, new_metadata=metadata)
self._update_model_after_modify_success(name, metadata)
def update(
self,
ids: OneOrMany[ID],
embeddings: Optional[ # type: ignore[type-arg]
Union[
OneOrMany[Embedding],
OneOrMany[np.ndarray],
]
] = None,
metadatas: Optional[OneOrMany[Metadata]] = None,
documents: Optional[OneOrMany[Document]] = None,
images: Optional[OneOrMany[Image]] = None,
uris: Optional[OneOrMany[URI]] = None,
) -> None:
"""Update the embeddings, metadatas or documents for provided ids.
Args:
ids: The ids of the embeddings to update
embeddings: The embeddings to update. If None, embeddings will be computed based on the documents or images using the embedding_function set for the Collection. Optional.
metadatas: The metadata to associate with the embeddings. When querying, you can filter on this metadata. Optional.
documents: The documents to associate with the embeddings. Optional.
images: The images to associate with the embeddings. Optional.
Returns:
None
"""
(
ids,
embeddings,
metadatas,
documents,
uris,
) = self._validate_and_prepare_update_request(
ids, embeddings, metadatas, documents, images, uris
)
self._client._update(self.id, ids, embeddings, metadatas, documents, uris)
def upsert(
self,
ids: OneOrMany[ID],
embeddings: Optional[ # type: ignore[type-arg]
Union[
OneOrMany[Embedding],
OneOrMany[PyEmbedding],
]
] = None,
metadatas: Optional[OneOrMany[Metadata]] = None,
documents: Optional[OneOrMany[Document]] = None,
images: Optional[OneOrMany[Image]] = None,
uris: Optional[OneOrMany[URI]] = None,
) -> None:
"""Update the embeddings, metadatas or documents for provided ids, or create them if they don't exist.
Args:
ids: The ids of the embeddings to update
embeddings: The embeddings to add. If None, embeddings will be computed based on the documents using the embedding_function set for the Collection. Optional.
metadatas: The metadata to associate with the embeddings. When querying, you can filter on this metadata. Optional.
documents: The documents to associate with the embeddings. Optional.
Returns:
None
"""
(
ids,
embeddings,
metadatas,
documents,
uris,
) = self._validate_and_prepare_upsert_request(
ids, embeddings, metadatas, documents, images, uris
)
self._client._upsert(
collection_id=self.id,
ids=ids,
embeddings=embeddings,
metadatas=metadatas,
documents=documents,
uris=uris,
)
def delete(
self,
ids: Optional[IDs] = None,
where: Optional[Where] = None,
where_document: Optional[WhereDocument] = None,
) -> None:
"""Delete the embeddings based on ids and/or a where filter
Args:
ids: The ids of the embeddings to delete
where: A Where type dict used to filter the delection by. E.g. `{"$and": [{"color" : "red"}, {"price": {"$gte": 4.20}]}}`. Optional.
where_document: A WhereDocument type dict used to filter the deletion by the document content. E.g. `{$contains: {"text": "hello"}}`. Optional.
Returns:
None
Raises:
ValueError: If you don't provide either ids, where, or where_document
"""
(ids, where, where_document) = self._validate_and_prepare_delete_request(
ids, where, where_document
)
self._client._delete(self.id, ids, where, where_document)
|