mosaicmpi.cnmf.cNMF

class mosaicmpi.cnmf.cNMF(output_dir='.', name=None)

Legacy cNMF object based off of the cNMF package

Parameters:

• output_dir (str, optional) – Place to put cNMF resutls, defaults to “.”

• name (str, optional) – A name for this analysis. Will be prefixed to all output files, defaults to automatically generated timestamp (and random string).

Methods

combine_nmf(k[, skip_missing_files, ...])
consensus(k[, density_threshold, ...])
factorize([worker_i, total_workers, verbose])	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::.
get_and_check_consensus(k, ...)
get_nmf_iter_params(ks[, n_iter, ...])	_summary_
postprocess([cpus, local_density_threshold, ...])
refit_spectra(X, usage)	Takes an input data matrix and a fixed usage matrix and uses NNLS to find the optimal spectra matrix.
refit_usage(X, spectra)	Takes an input data matrix and a fixed spectra and uses NNLS to find the optimal usage matrix.
save_nmf_iter_params(replicate_params, ...)