ERROR in compute NMF method 'brunet'

[Donatella93]

Hi! I keep getting this error while computing NMF with different methods.
This ERROR occur for methods 'brunet', 'KL' and others, but NOT for 'lee' and 'Frobenius' methods that work OK

I have a matrix with 130 rows (samples/obs) x 79 columns (features)

meth <- nmfAlgorithm(all = TRUE)
estim.r <- nmf(data, 2:6, method = meth, nrun=200, seed=123456)

Compute NMF method 'brunet' [2/16] ... Timing stopped at: 1.83 0.01 17.47
Timing stopped at: 1.92 0.02 17.27
Timing stopped at: 1.89 0.03 16.29
Timing stopped at: 1.83 0 16.64
Timing stopped at: 1.87 0.03 15.86
ERROR

Did anyone already deal with this issue?

Thanks

[vfincke]

I have the same problem. Did you solve it?

[Donatella93]

Hi!
I edited my dataset (containing binary data only), by removing columns and rows where the sum of values was 0 and I did not get anymore that error.
I can suggest you to check your data, and maybe it will help!

Good luck!

[Donatella93]

Hi! I edited my dataset (containing binary data only), by removing columns and rows where the sum of values was 0 and I did not get anymore that error. I can suggest you to check your data, and maybe it will help! Good luck! Il giorno ven 19 apr 2024 alle ore 17:11 Victoria Fincke < ***@***.***> ha scritto:

…

I have the same problem. Did you solve it? — Reply to this email directly, view it on GitHub <#180 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A2GCF6KVOJ4J6YSYZVAH633Y6EXYRAVCNFSM6AAAAABC5QCPI6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANRWG43TSNRVHE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Donatella Talotta, MSc, PhD Division of Hematology Department of Translational Medicine University of Eastern Piedmont Via Solaroli 17 28100 Novara, Italy Tel (39) (0321) 660-577 E-mail ***@***.***