API Reference / agent / tools / data /

select_features.py

Feature selection.

Functions

Func

execute_select_features

Line 12
execute_select_features(**kwargs) -> Dict[str, Any]

Run feature selection on a dataset.

Backs the tuiml_select_features tool. The reduced dataset is written to a temporary CSV for use by downstream tools.

Parameters

data
str
Dataset to select features from: dataset_id, file path, or built-in name (arrives via **kwargs, like all parameters below).
method
str
Selector class name: SelectKBestSelector, SelectPercentileSelector, VarianceThresholdSelector, CFSSelector, WrapperSelector, SelectFprSelector, or SelectThresholdSelector.
method_params
dict = None
Extra constructor parameters for the selector.
k
int = None
Number of features to keep (selectors that accept k).
threshold
float = None
Threshold value (selectors that accept threshold).
target
str = 'target'
Column name for the label in the output CSV.

Returns

result
dict
On success: status ('success'), method, n_original, n_selected, selected_features, file_path (reduced CSV) and, when available, scores (feature -> score). On failure: status ('error'), error and optionally suggestion / error_type.