ee5bb2a6b05ef2e7a0bc60adf1e2c8641d64eac9,tmtoolkit/lda_utils/evaluation_sklearn.py,,evaluate_topic_models,#Any#Any#Any#Any#Any#,22

Before Change

    if n_folds > 1:
        split_folds = get_split_folds_array(n_folds, data.shape[0])
    else:
        split_folds = None

    initializer_args = (shared_full_data_base, shared_sparse_data_base, shared_sparse_rows_base,
                        shared_sparse_cols_base, data.shape[0], data.shape[1], n_folds, split_folds)
    eval_results = start_multiproc_eval(n_workers, _init_shared_data, initializer_args,
                                        _fit_model_using_params, merged_params)

    return eval_results


def _init_shared_data(shared_full_data_base, shared_sparse_data_base, shared_sparse_rows_base,
                      shared_sparse_cols_base, n_rows, n_cols, n_folds, split_folds):

After Change

    mp_eval = MultiprocEvaluation(MultiprocEvaluationWorkerSklearn, data, varying_parameters, constant_parameters,
                                  n_max_processes=n_workers, n_folds=n_folds)

    return mp_eval.evaluate()

Instances