attribench.result.DeletionResult

class attribench.result.DeletionResult(method_names, maskers, activation_fns, mode, num_samples, num_steps)[source]

Bases: MetricResult

Represents results from running the Deletion metric.

Parameters:

method_namesList[str]: Names of attribution methods tested by Deletion.
maskersList[str]: Names of maskers used by Deletion.
activation_fnsList[str]: Names of activation functions used by Deletion.
modestr: Indicates if Deletion-MoRF or Deletion-LeRF was used. Options: “morf”, “lerf”
num_samplesint: Number of samples on which Deletion was run.
num_stepsint: Number of steps used by Deletion.

Methods

`add`	Adds a BatchResult to the result object.
`get_df`	Retrieves a dataframe from the result for a given masker and activation function.
`load`	Load a result from an HDF5 file or a directory of CSV files.
`save`	Save the result to an HDF5 file or a nested directory of CSV files.

add(batch_result)

Adds a BatchResult to the result object.

Parameters:

batch_resultBatchResult: BatchResult to add to the result object.

get_df(masker, activation_fn, agg_fn='auc', methods=None, columns=None)[source]

Retrieves a dataframe from the result for a given masker and activation function. The dataframe contains a row for each sample and a column for each method. Each value is the AUC/AOC for the given method on the given sample.

Parameters:

maskerstr: The masker to use.
activation_fnstr: The activation function to use.
agg_fnstr: Either “auc” for AUC or “aoc” for AOC.
methodsOptional[List[str]]: The methods to include. If None, includes all methods.
columnsOptional[npt.NDArray]: The columns used in the AUC/AOC calculation. If None, uses all columns.

Returns:

Tuple[pd.DataFrame, bool]: dataframe containing results, and boolean indicating if higher is better.

Return type:

Tuple[DataFrame, bool]

classmethod load(path)

Load a result from an HDF5 file or a directory of CSV files. The format is inferred from the path: if the path is a directory, the result is loaded from a directory of CSV files, otherwise the result is loaded from an HDF5 file.

The specific subclass of MetricResult is inferred from the metadata stored in the file or directory, and the appropriate load method is called.

Parameters:

pathstr: Path to the file or directory.

Returns:

MetricResult: The loaded result.

Return type:

MetricResult

save(path, format='hdf5')[source]

Save the result to an HDF5 file or a nested directory of CSV files.

Parameters:

pathstr: Path to the file.
formatstr: Format to save the result in. Options: hdf5, dir. If hdf5, the full result is stored in a single HDF5 file. If csv, the result is stored in a nested directory of CSV files.