holisticai.datasets.Dataset#

class holisticai.datasets.Dataset(_data: DataFrame | None = None, _metadata=None, **kargs)[source]#

Represents a dataset.

Parameters

data: pd.DataFrame: The underlying data of the dataset.
features: list[str]: The list of features in the dataset.
num_rows: int: The number of rows in the dataset.
random_state: np.random.RandomState: The random state used for sampling.

filter(fn)[source]#: Returns a new dataset with rows filtered based on the given function.

groupby(key: list[str] | str)[source]#: Returns a new GroupByDataset object based on the given key.

map(fn, vectorized=True)[source]#

Applies a function to the dataset and returns a new dataset.

Parameters

fn: function: The function to apply to the dataset.
vectorized: bool: Whether to apply the function in a vectorized manner or not.

remove_columns(columns: str | list)[source]#: Returns a new dataset with the given columns removed.

rename(renames)[source]#: Returns a new dataset with renamed columns.

sample(n, random_state=None)[source]#: Returns a random sample of n rows from the dataset.

select(indices: Iterable)[source]#: Returns a new dataset with selected rows based on the given indices.

train_test_split(test_size=0.3, **kargs)[source]#: Splits the dataset into train and test datasets.