folktexts.acs package

Submodules

folktexts.acs.acs_columns module

Module to hold ACS column mappings from values to natural text.

folktexts.acs.acs_columns.describe_income_to_poverty_ratio_pct(ratio_pct)[source]

Describe the income-to-poverty ratio in natural language.

Return type:

str

folktexts.acs.acs_dataset module

Module to access ACS data using the folktables package.

class folktexts.acs.acs_dataset.ACSDataset(data, full_acs_data, task, test_size=0.1, val_size=0.1, subsampling=None, seed=42)[source]

Bases: Dataset

Wrapper for ACS folktables datasets.

Construct a Dataset object.

Parameters:

  • data (pd.DataFrame) – The dataset’s data in pandas DataFrame format.

  • task (TaskMetadata) – The metadata for the prediction task.

  • test_size (float, optional) – The size of the test set, as a fraction of the total dataset size, by default 0.1.

  • val_size (float, optional) – The size of the validation set, as a fraction of the total dataset size, by default 0.1.

  • subsampling (float, optional) – Whether to use sub-sampling, and which fraction of the data to keep. By default will not use sub-sampling (subsampling=None).

  • seed (int, optional) – The random state seed, by default 42.

classmethod make_from_task(task, cache_dir=None, survey_year='2018', horizon='1-Year', survey='person', seed=42, **kwargs)[source]

Construct an ACSDataset object from a given ACS task.

Can customize survey sample parameters (survey year, horizon, survey type).

Parameters:

  • task (str | ACSTaskMetadata) – The name of the ACS task or the task object itself.

  • cache_dir (str | Path, optional) – The directory where ACS data is (or will be) saved to, by default uses DEFAULT_DATA_DIR.

  • survey_year (str, optional) – The year from which to load survey data, by default DEFAULT_SURVEY_YEAR.

  • horizon (str, optional) – The time horizon of survey data to load, by default DEFAULT_SURVEY_HORIZON.

  • survey (str, optional) – The name of the survey unit to load, by default DEFAULT_SURVEY_UNIT.

  • seed (int, optional) – The random seed, by default DEFAULT_SEED.

  • **kwargs – Extra key-word arguments to be passed to the Dataset constructor.

property task: ACSTaskMetadata

folktexts.acs.acs_questions module

A collection of instantiated ACS column objects and ACS tasks.

folktexts.acs.acs_tasks module

A collection of ACS prediction tasks based on the folktables package.

class folktexts.acs.acs_tasks.ACSTaskMetadata(name, features, target, cols_to_text, sensitive_attribute=None, target_threshold=None, multiple_choice_qa=None, direct_numeric_qa=None, cot_qa=None, description=None, _use_numeric_qa=False, _use_cot_qa=False, folktables_obj=None)[source]

Bases: TaskMetadata

A class to hold information on an ACS prediction task.

folktables_obj: BasicProblem = None
classmethod make_folktables_task(name, target_threshold=None, description=None)[source]
Return type:

ACSTaskMetadata

classmethod make_task(name, features, target=None, sensitive_attribute=None, target_threshold=None, multiple_choice_qa=None, direct_numeric_qa=None, description=None, folktables_obj=None)[source]

Create an ACS task object from the given parameters.

Return type:

ACSTaskMetadata

folktexts.acs.acs_thresholds module

Threshold instances for ACS / folktables tasks.

Module contents