- nmf_matrix
a matrix
used for NMF decomposition with rows indicate samples and columns indicate components.
- n_sig
number of signature. Please run sig_estimate to select a suitable value.
- nrun
a numeric
giving the number of run to perform for each value in range
, nrun
set to 30~50 is
enough to achieve robust result.
- cores
number of cpu cores to run NMF.
- method
specification of the NMF algorithm. Use 'brunet' as default.
Available methods for NMF decompositions are 'brunet', 'lee', 'ls-nmf', 'nsNMF', 'offset'.
- optimize
if TRUE
, then refit the denovo signatures with QP method, see sig_fit.
- pynmf
if TRUE
, use Python NMF driver Nimfa.
The seed currently is not used by this implementation.
- use_conda
if TRUE
, create an independent conda environment to run NMF.
- py_path
path to Python executable file, e.g. '/Users/wsx/anaconda3/bin/python'. In my
test, it is more stable than use_conda=TRUE
. You can install the Nimfa package by yourself
or set use_conda
to TRUE
to install required Python environment, and then set this option.
- seed
specification of the starting point or seeding method, which will compute a starting point,
usually using data from the target matrix in order to provide a good guess.
- ...
other arguments passed to NMF::nmf()
.