kimjingu / nonnegfac-python Goto Github PK

hello, I'm using NMF_ANLS_BLOCKPIVOT algorithm on a 200x990 matrix with this code
W1, H1, info1 = NMF_ANLS_BLOCKPIVOT().run(X1, 56, max_iter=500, verbose=1)
but after running for about 15 minutes, it shows an error for singular matrix:

File "nnmf.py", line 120, in
W1, H1, info1 = nmf().run(X1, rank, max_iter=500, verbose=1)
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/nonnegfac/nmf.py", line 97, in run
(W, H) = self.iter_solver(A, W, H, k, i)
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/nonnegfac/nmf.py", line 195, in iter_solver
Sol, info = nnlsm_blockpivot(W, A, init=H.T)
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/nonnegfac/nnls.py", line 120, in nnlsm_blockpivot
AtA, AtB[:, not_opt_cols], PassSet[:, not_opt_cols])
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/nonnegfac/nnls.py", line 322, in normal_eq_comb
Z[ix1] = nla.solve(AtA[ix2], AtB[ix1])
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/numpy/linalg/linalg.py", line 394, in solve
r = gufunc(a, b, signature=signature, extobj=extobj)
File "/home/navid/anaconda3/envs/fr/lib/python3.6/site-packages/numpy/linalg/linalg.py", line 89, in _raise_linalgerror_singular
raise LinAlgError("Singular matrix")
numpy.linalg.linalg.LinAlgError: Singular matrix

I searched the singular matrix error and I got this error happens if a numpy matrix isn't invertible.
now what should I do?
sorry for my bad english and I thank you.

Thanks ton for the wonderful package! I was wondering if there is some way in which missing data can be handled? The trivial application being MovieLens like dataset. I was wondering if instead of the usual 0 rating for missing entries in the matrix, we could just let those entries be np.NaN?

Hello Mr. Kim thanks for the package, My name is Viqi Nurhaqiqi I'm student at Department of Informatics UIN Sunan Gunung Djati Bandung, I have written the following code from the example in your package :

`W_org = random.rand(3, 3)
H_org = random.rand(3, 3)

A = np.array([[0,1,0],[1,0,1],[1,1,0]])

W, H, info = NMF().run(A, 2)`

My question is how to show matrix decomposition from A ?