statsmodels.base.distributed_estimation.DistributedModel.fit¶

DistributedModel.fit(data_generator, fit_kwds=None, parallel_method='sequential', parallel_backend=None, init_kwds_generator=None)[source]¶

Performs the distributed estimation using the corresponding DistributedModel

Parameters:

data_generatorgenerator: A generator that produces a sequence of tuples where the first element in the tuple corresponds to an endog array and the element corresponds to an exog array.
fit_kwdsdict-like or None: Keywords needed for the model fitting.
parallel_methodstr: type of distributed estimation to be used, currently “sequential”, “joblib” and “dask” are supported.
parallel_backendNone or joblib parallel_backend object: used to allow support for more complicated backends, ex: dask.distributed
init_kwds_generatorgenerator or None: Additional keyword generator that produces model init_kwds that may vary based on data partition. The current usecase is for WLS and GLS

Returns:

join_method result. For the default, _join_debiased, it returns a
p length array.