sdv.SDV

class sdv.SDV(model=None, model_kwargs=None)[source]

Automated generative modeling and sampling tool.

Allows the users to generate synthetic data after creating generative models for their data.

Parameters
  • model (type) – Class of the model to use. Defaults to sdv.relational.HMA1.

  • model_kwargs (dict) – Keyword arguments to pass to the model. If no model is given, this defaults to using a GaussianCopula with gaussian distribution and categorical_fuzzy categorical transformer.

__init__(model=None, model_kwargs=None)[source]

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([model, model_kwargs])

Initialize self.

fit(metadata[, tables, root_path])

Fit this SDV instance to the dataset data.

load(path)

Load a SDV instance from a given path.

sample([table_name, num_rows, …])

Generate synthetic data for one table or the entire dataset.

sample_all([num_rows, reset_primary_keys])

Sample the entire dataset.

save(path)

Save this SDV instance to the given path using pickle.

Attributes

DEFAULT_MODEL_KWARGS