Embeddings

zenpyre.embeddings

Contain embedding models.

zenpyre.embeddings.inspect_embeddings

inspect_embeddings(
    vector_store: VectorStore, n: int = 2
) -> None

Log a summary of documents retrieved from a vector store.

Retrieves the first n documents via :meth:~langchain_core.vectorstores.VectorStore.similarity_search and logs each document's ID, source metadata, and text content.

.. note:: Unlike the Chroma-specific implementation, this function works on any :class:~langchain_core.vectorstores.VectorStore backend but does not show embedding vectors, as there is no standard cross-provider API for retrieving raw embeddings.

Parameters:

Name	Type	Description	Default
vector_store	VectorStore	Any :class:~langchain_core.vectorstores.VectorStore instance to inspect.	required
n	int	Number of documents to preview. Defaults to 2.	2

zenpyre.embeddings.resolve_embeddings

resolve_embeddings(
    embeddings: Embeddings | dict[str, Any] | BaseConfig,
) -> Embeddings

Resolve a LangChain :class:~langchain_core.embeddings.Embeddings instance from an existing object, a configuration dictionary, or a :class:~zenpyre.utils.config.BaseConfig.

If embeddings is already an :class:~langchain_core.embeddings.Embeddings instance it is returned as-is. If it is a :class:dict or a :class:~zenpyre.utils.config.BaseConfig, it is treated as an objectory factory configuration and instantiated via :func:objectory.factory. See :func:~zenpyre.utils.resolve.resolve_object for details.

Parameters:

Name	Type	Description	Default
embeddings	Embeddings | dict[str, Any] | BaseConfig	Either a fully configured :class:~langchain_core.embeddings.Embeddings instance, a :class:dict containing an objectory factory specification (must include a "_target_" key pointing to the fully-qualified class name), or a :class:~zenpyre.utils.config.BaseConfig whose to_kwargs() includes a "_target_" key.	required

Returns:

Type	Description
Embeddings	A configured :class:~langchain_core.embeddings.Embeddings
Embeddings	instance.

Raises:

Type	Description
TypeError	If the resolved object is not a :class:~langchain_core.embeddings.Embeddings instance.

Example

>>> from zenpyre.embeddings import resolve_embeddings
>>> # From an existing instance:
>>> from langchain_core.embeddings.fake import FakeEmbeddings
>>> embeddings = resolve_embeddings(FakeEmbeddings(size=128))
>>> # From a configuration dictionary:
>>> embeddings = resolve_embeddings(
...     {"_target_": "langchain_ollama.OllamaEmbeddings", "model": "nomic-embed-text"}
... )

zenpyre.embeddings.factory

Contain embedding factories.

zenpyre.embeddings.factory.BaseEmbeddingsFactory

Bases: ABC

Abstract base class for embedding model factories.

Subclasses implement :meth:make_embedding to instantiate and return a configured :class:~langchain_core.factory.Embeddings object. This pattern decouples embedding creation from the rest of the codebase, making it easy to swap providers (e.g. Ollama, OpenAI, HuggingFace) without changing call sites.

Example

>>> from langchain_ollama import OllamaEmbeddings
>>> from zenpyre.embeddings.factory import BaseEmbeddingsFactory
>>> class OllamaEmbeddingsFactory(BaseEmbeddingsFactory):
...     def make_embeddings(self) -> OllamaEmbeddings:
...         return OllamaEmbeddings(model="nomic-embed-text")
...

zenpyre.embeddings.factory.BaseEmbeddingsFactory.make_embeddings abstractmethod

make_embeddings() -> Embeddings

Create and return a configured embedding model instance.

Returns:

Name	Type	Description
A	Embeddings	class:~langchain_core.factory.Embeddings instance ready for use.

zenpyre.embeddings.factory.ConfigurableEmbeddingsFactory

Bases: BaseEmbeddingsFactory, MultilineDisplayMixin

A concrete embedding factory that accepts either a pre-built :class:~langchain_core.embeddings.Embeddings instance or a configuration dictionary.

When a dict is provided it is resolved at each :meth:make_embeddings call via :func:~resolve_embeddings, which uses objectory to instantiate the configured class. When an instance is provided it is returned as-is.

Parameters:

Name	Type	Description	Default
embeddings	Embeddings | dict[str, Any]	A fully configured :class:~langchain_core.embeddings.Embeddings instance, or a :class:dict containing an objectory factory specification (must include a "_target_" key pointing to the fully-qualified class name).	required

Example

>>> from langchain_core.embeddings.fake import FakeEmbeddings
>>> from zenpyre.embeddings.factory import ConfigurableEmbeddingsFactory
>>> factory = ConfigurableEmbeddingsFactory(FakeEmbeddings(size=128))
>>> embeddings = factory.make_embeddings()

zenpyre.embeddings.factory.EmbeddingsFactory

Bases: BaseEmbeddingsFactory, MultilineDisplayMixin

A concrete embedding factory that wraps a pre-built :class:~langchain_core.embeddings.Embeddings instance.

Use this when the embedding model is already instantiated and you simply want to wrap it in the :class:~BaseEmbeddingsFactory interface — for example, when injecting a fixed embedding model into a component that expects a factory.

Parameters:

Name	Type	Description	Default
embeddings	Embeddings	A fully configured :class:~langchain_core.embeddings.Embeddings instance to return from :meth:make_embeddings.	required

Example

>>> from langchain_ollama import OllamaEmbeddings
>>> from zenpyre.embeddings.factory import EmbeddingsFactory
>>> factory = EmbeddingsFactory(OllamaEmbeddings(model="nomic-embed-text"))
>>> embeddings = factory.make_embeddings()

zenpyre.embeddings.factory.HuggingFaceEmbeddingsFactory

Bases: BaseEmbeddingsFactory, InlineDisplayMixin

A concrete embedding factory that wraps a pre-built :class:~langchain_core.embeddings.Embeddings instance.

Use this when the embedding model is already instantiated and you simply want to wrap it in the :class:~BaseEmbeddingsFactory interface — for example, when injecting a fixed embedding model into a component that expects a factory.

Parameters:

Name	Type	Description	Default
embeddings		A fully configured :class:~langchain_core.embeddings.Embeddings instance to return from :meth:make_embeddings.	required

Example

>>> from zenpyre.embeddings.factory import HuggingFaceEmbeddingsFactory
>>> factory = HuggingFaceEmbeddingsFactory(model_name="all-MiniLM-L6-v2")
>>> embeddings = factory.make_embeddings()  # doctest: +SKIP