promptmeteo.selector package#

Submodules#

promptmeteo.selector.base module#

class promptmeteo.selector.base.BaseSelector(language: str, embeddings: Embeddings, selector_k: int, selector_algorithm: str)#

Bases: ABC

Base Selector Interface

SELECTOR = None#
load_example_selector(model_path: str, **kwargs) → Self#

Load a vectorstore database from a disk file

property template: str#

Selector Template

property vectorstore#

Selector Vectorstore.

class promptmeteo.selector.base.BaseSelectorSupervised(language: str, embeddings: Embeddings, selector_k: int, selector_algorithm: str)#

Bases: BaseSelector

run(sample: str) → FewShotPromptTemplate#

Creates the FewShotPromptTemplate from the samples of the vectorstore.

train(examples: List[str], annotations: List[str]) → Self#

Creates the vectorstor with the training samples.

class promptmeteo.selector.base.BaseSelectorUnsupervised(language: str, embeddings: Embeddings, selector_k: int, selector_algorithm: str)#

Bases: BaseSelector

run() → FewShotPromptTemplate#

Creates the FewShotPromptTemplate from the samples of the vectorstore.

train(examples: List[str], annotations: List[str] | None) → Self#

Creates the vectorstore with the training samples.

class promptmeteo.selector.base.SelectorAlgorithms(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)#

Bases: str, Enum

Enum with the avaialable selector algorithms.

RELEVANCE: str = 'relevance'#
SIMILARITY: str = 'similarity'#
SIMILARITY_CLASS_BALANCED: str = 'similarity_class_balanced'#

promptmeteo.selector.custom_selectors module#

Custom selectors

class promptmeteo.selector.custom_selectors.BalancedSemanticSamplesSelector(*, vectorstore: VectorStore, k: int = 6, example_keys: List[str] | None = None, input_keys: List[str] | None = None, vectorstore_kwargs: Dict[str, Any] | None = None, class_list: List[str], class_key: str)#

Bases: BaseExampleSelector, BaseModel

Example selector that selects examples based on SemanticSimilarity in a balanced way.

class Config#

Bases: object

Configuration for this pydantic object.

arbitrary_types_allowed = True#
add_example(example: Dict[str, str]) → str#

Add new example to vectorstore.

class_key: str#
class_list: List[str]#

element of examples metadata which contains the class

example_keys: List[str] | None#

Optional keys to filter examples to.

classmethod from_examples(examples: List[dict], class_list: List[str], class_key: str, embeddings: Embeddings, vectorstore_cls: Type[VectorStore], k: int = 6, input_keys: List[str] | None = None, **vectorstore_cls_kwargs: Any)#

Create k-shot example selector using example list and embeddings.

Reshuffles examples dynamically based on query similarity.

Parameters:

  • examples – List of examples to use in the prompt.

  • class_list – list of classes of the classification problem

  • class_key – key which refers to category field in the example dictionary

  • embeddings – An initialized embedding API interface, e.g. OpenAIEmbeddings().

  • vectorstore_cls – A vector store DB interface class, e.g. FAISS.

  • input_keys – If provided, the search is based on the input variables instead of all variables.

  • vectorstore_cls_kwargs – optional kwargs containing url for vector store

Returns:

The ExampleSelector instantiated, backed by a vector store.

input_keys: List[str] | None#

Optional keys to filter input to. If provided, the search is based on the input variables instead of all variables.

k: int#

Number of examples to select per class.

select_examples(input_variables: Dict[str, str]) → List[dict]#

Select which examples to use based on semantic similarity.

vectorstore: VectorStore#

Vectorstore

vectorstore_kwargs: Dict[str, Any] | None#

Extra arguments passed to similarity_search function of the vectorstore.

promptmeteo.selector.custom_selectors.sorted_values(values: Dict[str, str]) → List[Any]#

Return a list of values in dict sorted by key.

Module contents#

class promptmeteo.selector.SelectorFactory#

Bases: object

Factory of Selectors

classmethod factory_method(language: str, embeddings: Embeddings, selector_k: int, selector_type: str, selector_algorithm: str) → BaseSelector#

Returns and instance of a BaseSelector object depending on the selector_algorithm.

class promptmeteo.selector.SelectorTypes(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)#

Bases: str, Enum

Enum with the avaialable selector types.

SUPERVISED: str = 'supervised'#
UNSUPERVISED: str = 'unsupervised'#