mosaicmpi.cnmf.cNMF.factorize

Contents

mosaicmpi.cnmf.cNMF.factorize#

cNMF.factorize(worker_i=0, total_workers=1, verbose=True)#

Iteratively run NMF with prespecified parameters. Use the worker_i and total_workers parameters for parallelization. Generic kwargs for NMF are loaded from self.paths[‘nmf_run_parameters’], defaults below:

non_negative_factorization default arguments:
    alpha=0.0
    l1_ratio=0.0
    beta_loss='kullback-leibler'
    solver='mu'
    tol=1e-4,
    max_iter=200
    regularization=None
    init='random'
    random_state, n_components are both set by the prespecified self.paths['nmf_replicate_parameters'].
Parameters:

  • worker_i (int, optional) – worker index, defaults to 0

  • total_workers (int, optional) – total number of workers, defaults to 1

  • verbose (bool, optional) – verbose, defaults to True

Contents