EmbeddingsVectorizer#

class langchain_redis.cache.EmbeddingsVectorizer[source]#

Bases: BaseVectorizer

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

param cache: EmbeddingsCache | None = None#
param dims: Annotated[int | None, Field(strict=True, gt=0)] = None#
Constraints:

  • strict = True

  • gt = 0

param dtype: str = 'float32'#
param embeddings: Embeddings [Required]#
param model: str = 'custom_embeddings'#
async aembed(
text: str,
dtype: str | VectorDataType = 'float32',
**kwargs: Any,
) → List[float][source]#

Asynchronously generate a vector embedding for a text string.

Parameters:

  • text (str) – The text to convert to a vector embedding

  • preprocess – Function to apply to the text before embedding

  • as_buffer – Return the embedding as a binary buffer instead of a list

  • skip_cache – Bypass the cache for this request

  • **kwargs (Any) – Additional model-specific parameters

  • dtype (str | VectorDataType)

  • **kwargs

Returns:

The vector embedding as either a list of floats or binary buffer

Return type:

List[float]

Examples

>>> embedding = await vectorizer.aembed("Hello world")
async aembed_many(
texts: List[str],
dtype: str | VectorDataType = 'float32',
**kwargs: Any,
) → List[List[float]][source]#

Asynchronously generate vector embeddings for multiple texts efficiently.

Parameters:

  • texts (List[str]) – List of texts to convert to vector embeddings

  • preprocess – Function to apply to each text before embedding

  • batch_size – Number of texts to process in each API call

  • as_buffer – Return embeddings as binary buffers instead of lists

  • skip_cache – Bypass the cache for this request

  • **kwargs (Any) – Additional model-specific parameters

  • dtype (str | VectorDataType)

  • **kwargs

Returns:

List of vector embeddings in the same order as the input texts

Return type:

List[List[float]]

Examples

>>> embeddings = await vectorizer.aembed_many(["Hello", "World"], batch_size=2)
batchify(
seq: list,
size: int,
preprocess: Callable | None = None,
)#

Split a sequence into batches of specified size.

Parameters:

  • seq (list) – Sequence to split into batches

  • size (int) – Batch size

  • preprocess (Callable | None) – Optional function to preprocess each item

Yields:

Batches of the sequence

embed(
text: str,
dtype: str | VectorDataType = 'float32',
**kwargs: Any,
) → List[float][source]#

Generate a vector embedding for a text string.

Parameters:

  • text (str) – The text to convert to a vector embedding

  • preprocess – Function to apply to the text before embedding

  • as_buffer – Return the embedding as a binary buffer instead of a list

  • skip_cache – Bypass the cache for this request

  • **kwargs (Any) – Additional model-specific parameters

  • dtype (str | VectorDataType)

  • **kwargs

Returns:

The vector embedding as either a list of floats or binary buffer

Return type:

List[float]

Examples

>>> embedding = vectorizer.embed("Hello world")
embed_many(
texts: List[str],
dtype: str | VectorDataType = 'float32',
**kwargs: Any,
) → List[List[float]][source]#

Generate vector embeddings for multiple texts efficiently.

Parameters:

  • texts (List[str]) – List of texts to convert to vector embeddings

  • preprocess – Function to apply to each text before embedding

  • batch_size – Number of texts to process in each API call

  • as_buffer – Return embeddings as binary buffers instead of lists

  • skip_cache – Bypass the cache for this request

  • **kwargs (Any) – Additional model-specific parameters

  • dtype (str | VectorDataType)

  • **kwargs

Returns:

List of vector embeddings in the same order as the input texts

Return type:

List[List[float]]

Examples

>>> embeddings = vectorizer.embed_many(["Hello", "World"], batch_size=2)
encode(
texts: str | List[str],
dtype: str | VectorDataType,
**kwargs: Any,
) → ndarray[source]#
Parameters:

  • texts (str | List[str])

  • dtype (str | VectorDataType)

  • kwargs (Any)

Return type:

ndarray

property type: str#

Return the type of vectorizer.