langtest.augmentation.base.TemplaticAugment#

class TemplaticAugment(templates: str | List[str], task: TaskManager, generate_templates=False, show_templates=False)#

This class is used for templatic augmentation. It is a subclass of the BaseAugmentation class.

__templates#: A string or a list of strings or samples that represents the templates for the augmentation.

__generate_templates#: if set to True, generates sample templates from the given ones.

__init__(self, templates: Union[str, List[str]], task: str): Initializes the TemplaticAugment class.

fix(self, training_data: str, output_path: str, *args, **kwargs): Performs the templatic augmentation and exports the results to a specified path.

__init__(templates: str | List[str], task: TaskManager, generate_templates=False, show_templates=False) → None#

This constructor for the TemplaticAugment class.

Parameters:

templates (Union[str, List[str]]) – The templates to be used for the augmentation.
task (str) – The task for which the augmentation is being performed.
generate_templates (bool, optional) – if set to True, generates sample templates from the given ones.
show_templates (bool, optional) – if set to True, displays the used templates.

Methods

`__init__`(templates, task[, ...])	This constructor for the TemplaticAugment class.
`add_spaces_around_punctuation`(text)	This method is used to add spaces around punctuation in a string.
`extract_variable_names`(f_string)	This method is used to extract the variable names from the templates.
`fix`(training_data, output_path[, max_num, ...])	This method is used to perform the templatic augmentation.
`new_sample`(template)	This method is used to generate a new sample from a template.
`search_sample_results`(samples)	This method is used to search the results of the samples for the entities in the templates.
`str_to_sample`(template)	This method is used to convert a template string to a Sample object.

Attributes

`task`	Task getter
`templates`	Templates getter

static add_spaces_around_punctuation(text: str)#

This method is used to add spaces around punctuation in a string.

static extract_variable_names(f_string: str) → List[str]#

This method is used to extract the variable names from the templates.

fix(training_data: Dict[str, Any], output_path: str, max_num: int | None = None, append_original: bool = False, *args, **kwargs) → bool#

This method is used to perform the templatic augmentation.

It takes the input data, performs the augmentation and then saves the augmented data to the output path.

Parameters:

training_data (dict) – A dictionary containing the input data for augmentation.
output_path (str) – The path where the augmented data will be saved.
max_num (int) – Maximum number of new samples to generate
append_original (bool, optional) – If set to True, appends the original data to the augmented data. Defaults to False.
*args – Variable length argument list.
**kwargs – Arbitrary keyword arguments.

Returns:

Returns True upon successful completion of the method.

Return type:

bool

new_sample(template: Sample)#

This method is used to generate a new sample from a template.

Parameters:: template (Sample) – The template from which the new sample is to be generated.
Returns:: The new sample generated from the template.
Return type:: Sample

static search_sample_results(samples: List[Sample]) → Dict[str, List[NERPrediction | SequenceLabel]]#

This method is used to search the results of the samples for the entities in the templates.

Parameters:: samples (List[Sample]) – The samples for which the results are to be searched.
Returns:: A dictionary containing the search results.
Return type:: Dict[str, List[Union[NERPrediction, SequenceLabel]]]

str_to_sample(template: str)#

This method is used to convert a template string to a Sample object.