Struct chromadb::v1::collection::ChromaCollection

pub struct ChromaCollection { /* private fields */ }

A collection representation for interacting with the associated ChromaDB collection.

Implementations

impl ChromaCollection

pub fn id(&self) -> &str

Get the UUID of the collection.

pub fn name(&self) -> &str

Get the name of the collection.

pub fn metadata(&self) -> Option<&Map<String, Value>>

Get the metadata of the collection.

pub fn count(&self) -> Result<usize>

The total number of embeddings added to the database.

pub fn modify( &self, name: Option<&str>, metadata: Option<&Map<String, Value>> ) -> Result<()>

Modify the name/metadata of a collection.

Arguments

• name - The new name of the collection. Must be unique.
• metadata - The new metadata of the collection. Must be a JSON object with keys and values that are either numbers, strings or floats.

Errors

• If the collection name is invalid

pub fn add( &self, collection_entries: CollectionEntries<'_>, embedding_function: Option<Box<dyn EmbeddingFunction>> ) -> Result<Value>

Add embeddings to the data store. Ignore the insert if the ID already exists.

Arguments

• ids - The ids to associate with the embeddings.
• embeddings - The embeddings to add. If None, embeddings will be computed based on the documents using the provided embedding_function. Optional.
• metadata - 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.
• embedding_function - The function to use to compute the embeddings. If None, embeddings must be provided. Optional.

Errors

• If you don’t provide either embeddings or documents
• If the length of ids, embeddings, metadatas, or documents don’t match
• If you provide duplicates in ids, empty ids
• If you provide documents and don’t provide an embedding function when embeddings is None
• If you provide an embedding function and don’t provide documents
• If you provide both embeddings and embedding_function

pub fn upsert( &self, collection_entries: CollectionEntries<'_>, embedding_function: Option<Box<dyn EmbeddingFunction>> ) -> Result<Value>

Add embeddings to the data store. Update the entry if an ID already exists.

Arguments

• ids - The ids to associate with the embeddings.
• embeddings - The embeddings to add. If None, embeddings will be computed based on the documents using the provided embedding_function. Optional.
• metadata - 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.
• embedding_function - The function to use to compute the embeddings. If None, embeddings must be provided. Optional.

Errors

• If you don’t provide either embeddings or documents
• If the length of ids, embeddings, metadatas, or documents don’t match
• If you provide duplicates in ids, empty ids
• If you provide documents and don’t provide an embedding function when embeddings is None
• If you provide an embedding function and don’t provide documents
• If you provide both embeddings and embedding_function

pub fn get(&self, get_options: GetOptions) -> Result<GetResult>

Get embeddings and their associated data from the collection. If no ids or filter is provided returns all embeddings up to limit starting at offset.

Arguments

• ids - The ids of the embeddings to get. Optional..
• where_metadata - Used to filter results by metadata. E.g. { "$and": [{"foo": "bar"}, {"price": {"$gte": 4.20}}] }. See https://docs.trychroma.com/usage-guide#filtering-by-metadata for more information on metadata filters. Optional.
• limit - The maximum number of documents to return. Optional.
• offset - The offset to start returning results from. Useful for paging results with limit. Optional.
• where_document - Used to filter by the documents. E.g. {“$contains”: “hello”}. See https://docs.trychroma.com/usage-guide#filtering-by-document-contents for more information on document content filters. 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.

pub fn update( &self, collection_entries: CollectionEntries<'_>, embedding_function: Option<Box<dyn EmbeddingFunction>> ) -> Result<bool>

Update the embeddings, metadatas or documents for provided ids.

Arguments

• ids - The ids to associate with the embeddings.
• embeddings - The embeddings to add. If None, embeddings will be computed based on the documents using the provided embedding_function. Optional.
• metadata - 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.
• embedding_function - The function to use to compute the embeddings. If None, embeddings must be provided. Optional.

Errors

• If the length of ids, embeddings, metadatas, or documents don’t match
• If you provide duplicates in ids, empty ids
• If you provide documents and don’t provide an embedding function when embeddings is None
• If you provide an embedding function and don’t provide documents
• If you provide both embeddings and embedding_function

pub fn query( &self, query_options: QueryOptions<'_>, embedding_function: Option<Box<dyn EmbeddingFunction>> ) -> Result<QueryResult>

Get the n_results nearest neighbor embeddings for provided query_embeddings or query_texts.

Arguments

• query_embeddings - The embeddings to get the closest neighbors of. Optional.
• query_texts - The document texts to get the closest neighbors of. Optional.
• n_results - The number of neighbors to return for each query_embedding or query_texts. Optional.
• where_metadata - Used to filter results by metadata. E.g. {“$and”: [“color” : “red”, “price”: {“$gte”: 4.20}]}. Optional.
• where_document - Used to filter results by documents. E.g. {$contains: “some text”}. 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.
• embedding_function - The function to use to compute the embeddings. If None, embeddings must be provided. Optional.

Errors

• If you don’t provide either query_embeddings or query_texts
• If you provide both query_embeddings and query_texts
• If you provide query_texts and don’t provide an embedding function when embeddings is None

pub fn peek(&self, limit: usize) -> Result<GetResult>

Get the first entries in the collection up to the limit

Arguments

• limit - The number of entries to return.

pub fn delete( &self, ids: Option<Vec<&str>>, where_metadata: Option<Value>, where_document: Option<Value> ) -> Result<Vec<String>>

Delete the embeddings based on ids and/or a where filter. Deletes all the entries if None are provided

Arguments

• ids - The ids of the embeddings to delete. Optional
• where_metadata - Used to filter deletion by metadata. E.g. {“$and”: [“color” : “red”, “price”: {“$gte”: 4.20}]}. Optional.
• where_document - Used to filter the deletion by the document content. E.g. {$contains: “some text”}. Optional.. Optional.

Trait Implementations

impl Debug for ChromaCollection

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ChromaCollection

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.