openvino_genai.TextEmbeddingPipeline#

class openvino_genai.TextEmbeddingPipeline#

Bases: pybind11_object

Text embedding pipeline

__init__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, models_path: os.PathLike | str | bytes, device: str, config: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.Config | None = None, **kwargs) None#

Constructs a pipeline from xml/bin files, tokenizer and configuration in the same dir models_path (os.PathLike): Path to the directory containing model xml/bin files and tokenizer device (str): Device to run the model on (e.g., CPU, GPU). config: (TextEmbeddingPipeline.Config): Optional pipeline configuration kwargs: Plugin and/or config properties

Methods

__delattr__(name, /)

Implement delattr(self, name).

__dir__()

Default dir() implementation.

__eq__(value, /)

Return self==value.

__format__(format_spec, /)

Default object formatter.

__ge__(value, /)

Return self>=value.

__getattribute__(name, /)

Return getattr(self, name).

__getstate__()

Helper for pickle.

__gt__(value, /)

Return self>value.

__hash__()

Return hash(self).

__init__(self, models_path, device[, config])

Constructs a pipeline from xml/bin files, tokenizer and configuration in the same dir models_path (os.PathLike): Path to the directory containing model xml/bin files and tokenizer device (str): Device to run the model on (e.g., CPU, GPU).

__init_subclass__

This method is called when a class is subclassed.

__le__(value, /)

Return self<=value.

__lt__(value, /)

Return self<value.

__ne__(value, /)

Return self!=value.

__new__(**kwargs)

__reduce__()

Helper for pickle.

__reduce_ex__(protocol, /)

Helper for pickle.

__repr__()

Return repr(self).

__setattr__(name, value, /)

Implement setattr(self, name, value).

__sizeof__()

Size of object in memory, in bytes.

__str__()

Return str(self).

__subclasshook__

Abstract classes can override this to customize issubclass().

_pybind11_conduit_v1_

embed_documents(self, texts)

Computes embeddings for a vector of texts

embed_query(self, text)

Computes embeddings for a query

start_embed_documents_async(self, texts)

Asynchronously computes embeddings for a vector of texts

start_embed_query_async(self, text)

Asynchronously computes embeddings for a query

wait_embed_documents(self)

Waits computed embeddings of a vector of texts

wait_embed_query(self)

Waits computed embeddings for a query

Attributes

__annotations__

class Config#

Bases: pybind11_object

Structure to keep TextEmbeddingPipeline configuration parameters.

Attributes:
max_length (int, optional):

Maximum length of tokens passed to the embedding model.

pad_to_max_length (bool, optional):

If ‘True’, model input tensors are padded to the maximum length.

batch_size (int, optional):

Batch size for the embedding model. Useful for database population. If set, the pipeline will fix model shape for inference optimization. Number of documents passed to pipeline should be equal to batch_size. For query embeddings, batch_size should be set to 1 or not set.

pooling_type (TextEmbeddingPipeline.PoolingType, optional):

Pooling strategy applied to the model output tensor. Defaults to PoolingType.CLS.

normalize (bool, optional):

If True, L2 normalization is applied to embeddings. Defaults to True.

query_instruction (str, optional):

Instruction to use for embedding a query.

embed_instruction (str, optional):

Instruction to use for embedding a document.

padding_side (str, optional):

Side to use for padding “left” or “right”

__annotations__ = {}#
__class__#

alias of pybind11_type

__delattr__(name, /)#

Implement delattr(self, name).

__dir__()#

Default dir() implementation.

__eq__(value, /)#

Return self==value.

__format__(format_spec, /)#

Default object formatter.

Return str(self) if format_spec is empty. Raise TypeError otherwise.

__ge__(value, /)#

Return self>=value.

__getattribute__(name, /)#

Return getattr(self, name).

__getstate__()#

Helper for pickle.

__gt__(value, /)#

Return self>value.

__hash__()#

Return hash(self).

__init__(*args, **kwargs)#

Overloaded function.

  1. __init__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.Config) -> None

  2. __init__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.Config, **kwargs) -> None

__init_subclass__()#

This method is called when a class is subclassed.

The default implementation does nothing. It may be overridden to extend subclasses.

__le__(value, /)#

Return self<=value.

__lt__(value, /)#

Return self<value.

__ne__(value, /)#

Return self!=value.

__new__(**kwargs)#
__reduce__()#

Helper for pickle.

__reduce_ex__(protocol, /)#

Helper for pickle.

__repr__()#

Return repr(self).

__setattr__(name, value, /)#

Implement setattr(self, name, value).

__sizeof__()#

Size of object in memory, in bytes.

__str__()#

Return str(self).

__subclasshook__()#

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

_pybind11_conduit_v1_()#
property batch_size#
property embed_instruction#
property max_length#
property normalize#
property pad_to_max_length#
property padding_side#
property pooling_type#
property query_instruction#
validate(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.Config) None#

Checks that are no conflicting parameters. Raises exception if config is invalid.

class PoolingType#

Bases: pybind11_object

Members:

CLS : First token embeddings

MEAN : The average of all token embeddings

LAST_TOKEN : Last token embeddings

CLS = <PoolingType.CLS: 0>#
LAST_TOKEN = <PoolingType.LAST_TOKEN: 2>#
MEAN = <PoolingType.MEAN: 1>#
__annotations__ = {}#
__class__#

alias of pybind11_type

__delattr__(name, /)#

Implement delattr(self, name).

__dir__()#

Default dir() implementation.

__eq__(self: object, other: object, /) bool#
__format__(format_spec, /)#

Default object formatter.

Return str(self) if format_spec is empty. Raise TypeError otherwise.

__ge__(value, /)#

Return self>=value.

__getattribute__(name, /)#

Return getattr(self, name).

__getstate__(self: object, /) int#
__gt__(value, /)#

Return self>value.

__hash__(self: object, /) int#
__index__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.PoolingType, /) int#
__init__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.PoolingType, value: SupportsInt) None#
__init_subclass__()#

This method is called when a class is subclassed.

The default implementation does nothing. It may be overridden to extend subclasses.

__int__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.PoolingType, /) int#
__le__(value, /)#

Return self<=value.

__lt__(value, /)#

Return self<value.

__members__ = {'CLS': <PoolingType.CLS: 0>, 'LAST_TOKEN': <PoolingType.LAST_TOKEN: 2>, 'MEAN': <PoolingType.MEAN: 1>}#
__ne__(self: object, other: object, /) bool#
__new__(**kwargs)#
__reduce__()#

Helper for pickle.

__reduce_ex__(protocol, /)#

Helper for pickle.

__repr__(self: object, /) str#
__setattr__(name, value, /)#

Implement setattr(self, name, value).

__setstate__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.PoolingType, state: SupportsInt, /) None#
__sizeof__()#

Size of object in memory, in bytes.

__str__(self: object, /) str#
__subclasshook__()#

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

_pybind11_conduit_v1_()#
property name#
property value#
__annotations__ = {}#
__class__#

alias of pybind11_type

__delattr__(name, /)#

Implement delattr(self, name).

__dir__()#

Default dir() implementation.

__eq__(value, /)#

Return self==value.

__format__(format_spec, /)#

Default object formatter.

Return str(self) if format_spec is empty. Raise TypeError otherwise.

__ge__(value, /)#

Return self>=value.

__getattribute__(name, /)#

Return getattr(self, name).

__getstate__()#

Helper for pickle.

__gt__(value, /)#

Return self>value.

__hash__()#

Return hash(self).

__init__(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, models_path: os.PathLike | str | bytes, device: str, config: openvino_genai.py_openvino_genai.TextEmbeddingPipeline.Config | None = None, **kwargs) None#

Constructs a pipeline from xml/bin files, tokenizer and configuration in the same dir models_path (os.PathLike): Path to the directory containing model xml/bin files and tokenizer device (str): Device to run the model on (e.g., CPU, GPU). config: (TextEmbeddingPipeline.Config): Optional pipeline configuration kwargs: Plugin and/or config properties

__init_subclass__()#

This method is called when a class is subclassed.

The default implementation does nothing. It may be overridden to extend subclasses.

__le__(value, /)#

Return self<=value.

__lt__(value, /)#

Return self<value.

__ne__(value, /)#

Return self!=value.

__new__(**kwargs)#
__reduce__()#

Helper for pickle.

__reduce_ex__(protocol, /)#

Helper for pickle.

__repr__()#

Return repr(self).

__setattr__(name, value, /)#

Implement setattr(self, name, value).

__sizeof__()#

Size of object in memory, in bytes.

__str__()#

Return str(self).

__subclasshook__()#

Abstract classes can override this to customize issubclass().

This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).

_pybind11_conduit_v1_()#
embed_documents(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, texts: collections.abc.Sequence[str]) list[list[float]] | list[list[int]] | list[list[int]]#

Computes embeddings for a vector of texts

embed_query(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, text: str) list[float] | list[int] | list[int]#

Computes embeddings for a query

start_embed_documents_async(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, texts: collections.abc.Sequence[str]) None#

Asynchronously computes embeddings for a vector of texts

start_embed_query_async(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline, text: str) None#

Asynchronously computes embeddings for a query

wait_embed_documents(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline) list[list[float]] | list[list[int]] | list[list[int]]#

Waits computed embeddings of a vector of texts

wait_embed_query(self: openvino_genai.py_openvino_genai.TextEmbeddingPipeline) list[float] | list[int] | list[int]#

Waits computed embeddings for a query