xailib.data_loaders package

Submodules

xailib.data_loaders.dataframe_loader module

DataFrame loading and preparation utilities for XAI-Lib.

This module provides utilities for loading and preparing pandas DataFrames for use with XAI-Lib tabular data explainers.

Functions:

prepare_dataframe: Prepare a DataFrame for use with explainers.

Example

>>> from xailib.data_loaders.dataframe_loader import prepare_dataframe
>>>
>>> df, feature_names, class_values, numeric_columns, \
...     rdf, real_feature_names, features_map = prepare_dataframe(df, 'target')
xailib.data_loaders.dataframe_loader.prepare_dataframe(df, class_field)[source]

Prepare a pandas DataFrame for use with XAI-Lib explainers.

This function wraps the LORE library’s prepare_dataset function to process a DataFrame for use with various XAI-Lib explainers. It handles feature encoding, categorical variable mapping, and other preprocessing steps.

Parameters:

  • df (pd.DataFrame) – Input DataFrame containing features and target.

  • class_field (str) – Name of the target/class column in the DataFrame.

Returns:

A tuple containing:

  • df: Processed DataFrame

  • feature_names: List of feature column names

  • class_values: List of unique class values

  • numeric_columns: List of numeric column names

  • rdf: Reconstructed DataFrame with original encoding

  • real_feature_names: Original feature names before encoding

  • features_map: Mapping of encoded to original features

Return type:

tuple

Example

>>> import pandas as pd
>>> df = pd.DataFrame({
...     'age': [25, 30, 35],
...     'income': [50000, 60000, 70000],
...     'approved': [0, 1, 1]
... })
>>> result = prepare_dataframe(df, 'approved')
>>> df_processed, feature_names, class_values, *rest = result

Module contents

Data loading utilities for XAI-Lib.

This subpackage provides utilities for loading and preparing data for use with XAI-Lib explainers.

Available Loaders

Example

>>> from xailib.data_loaders.dataframe_loader import prepare_dataframe
>>>
>>> df, feature_names, class_values, numeric_columns, \
...     rdf, real_feature_names, features_map = prepare_dataframe(df, 'target')