langtest.transform.robustness.AddContext#

class AddContext#

Bases: BaseRobustness

A class for adding context to text samples.

__init__()#

Methods

__init__()
async_run(sample_list, model, **kwargs)	Creates a task to run the robustness measure.
run(sample_list, model, **kwargs)	Abstract method that implements the robustness measure.
transform(sample_list[, prob, ...])	Adds context to the input sentences.

Attributes

alias_name
supported_tasks
test_types

async classmethod async_run(sample_list: List[Sample], model: ModelAPI, **kwargs)#

Creates a task to run the robustness measure.

Parameters:

• sample_list (List[Sample]) – The input data to be transformed.

• model (ModelAPI) – The model to be used for evaluation.

• **kwargs – Additional arguments to be passed to the robustness measure.

Returns:

The task that runs the robustness measure.

Return type:

asyncio.Task

abstract async static run(sample_list: List[Sample], model: ModelAPI, **kwargs) → List[Sample]#

Abstract method that implements the robustness measure.

Parameters:

• sample_list (List[Sample]) – The input data to be transformed.

• model (ModelAPI) – The model to be used for evaluation.

• **kwargs – Additional arguments to be passed to the robustness measure.

Returns:

The transformed data based on the implemented robustness measure.

Return type:

List[Sample]

static transform(sample_list: List[Sample], prob: float | None = 1.0, starting_context: List[str] | None = None, ending_context: List[str] | None = None, strategy: str | None = None, count: int = 1) → List[Sample]#

Adds context to the input sentences.

Parameters:

• sample_list (List[Sample]) – A list of samples to be transformed.

• prob (Optional[float]) – The probability of adding context to each sample. Defaults to 1.0, which means all samples will have context added.

• starting_context (Optional[List[str]]) – A list of terms (context) to be added at the start of sentences.

• ending_context (Optional[List[str]]) – A list of terms (context) to be added at the end of sentences.

• strategy (str) – Config method to adjust where the context tokens are added. Options: ‘start’, ‘end’, or ‘combined’.

• count – Number of variations to create.

Returns:

The transformed sample list with context added.

Return type:

List[Sample]