SeismicSearchConfig

class lightning_ir.retrieve.seismic.seismic_searcher.SeismicSearchConfig(k: int = 10, candidate_k: int = 100, imputation_strategy: 'min' | 'gather' | 'zero' = 'min', query_cut: int = 10, heap_factor: float = 0.7, num_threads: int = 1)

Bases: ApproximateSearchConfig

Configuration for SeismicSearcher.

__init__(k: int = 10, candidate_k: int = 100, imputation_strategy: 'min' | 'gather' | 'zero' = 'min', query_cut: int = 10, heap_factor: float = 0.7, num_threads: int = 1) → None

Initialize the SeismicSearchConfig.

Parameters:

• k (int) – Number of top candidates to retrieve. Defaults to 10.

• candidate_k (int) – Number of candidates to consider for each query. Defaults to 100.

• imputation_strategy (Literal["min", "gather", "zero"]) – Strategy for handling missing values. Defaults to “min”.

• query_cut (int) – Maximum number of components per query. Defaults to 10.

• heap_factor (float) – Factor to control the size of the heap used in the search. Defaults to 0.7.

• num_threads (int) – Number of threads to use for parallel processing. Defaults to 1.

Raises:

ValueError – If imputation_strategy is “gather”, as it is not supported for SeismicSearcher.

Methods

__init__([k, candidate_k, ...])

Initialize the SeismicSearchConfig.

Attributes

SUPPORTED_MODELS

search_class

alias of SeismicSearcher