mosaicmpi.dataset.Dataset.initialize_cnmf

mosaicmpi.dataset.Dataset.initialize_cnmf#

Dataset.initialize_cnmf(cnmf_output_dir: str, cnmf_name: str, kvals: Collection = range(2, 61), n_iter: int = 200, beta_loss: str = 'kullback-leibler', seed: int | None = None) → cNMF#

Initialize a cNMF run for subsequent factorization.

Parameters:

  • cnmf_output_dir (str) – Output directory for cNMF results

  • cnmf_name (str) – Name of the cNMF results. Files will be output to [cnmf_output_dir]/[cnmf_name]/

  • kvals (Collection, optional) – Ranks for cNMF factorization, defaults to range(2, 61)

  • n_iter (int, optional) – Number of iterations from which to build a consensus solution, defaults to 200

  • beta_loss (str, optional) – beta-loss function, either “kullback-leibler” or “frobenius”. Defaults to “kullback-leibler”

  • seed (Optional[int], optional) – Random seed for reproducibility, defaults to None

Returns:

cNMF object

Return type:

mosaicmpi.cnmf.cNMF

Contents