search

renozao / NMF

NMF: A Flexible R package for Nonnegative Matrix Factorization
139 stars 41 forks source link

Error: NMF::nmf - Unexpected error: no partial result seem to have been saved. #68

Open lavkan opened 8 years ago

lavkan commented 8 years ago

Hi,

I get the following error, but my colleague does not get any error for the same eset. What could make it possible? Should I try and remove any known temporary folder NMF? If so, how do I locate those files.

Below is a minimal example:

expression.matrix <- exprs(eset) rank = 4 nrun = 100 nmf.out <- NMF::nmf(expression.matrix, rank=rank, nrun=nrun) ERROR Error: NMF::nmf - Unexpected error: no partial result seem to have been saved. Timing stopped at: 149.217 18.538 63.211

Thanks, Lavanya

omelnikov commented 8 years ago

In your minimal example, what is exprs(eset) ? Could you use some common data set in R or a randomly generated set? This example is not reproducible. Cheers!

lavkan commented 8 years ago

Please use this:

expression.matrix <- matrix(runif(500*500), ncol=500) nmf.out <- NMF::nmf(expression.matrix, rank=4, nrun=100) ERROR Error: NMF::nmf - Unexpected error: no partial result seem to have been saved. Timing stopped at: 80.825 14.639 37.558

On Tue, Jun 21, 2016 at 1:00 PM, OLEG MELNIKOV notifications@github.com wrote:

In your minimal example, what is exprs(eset) ? Could you use some common data set in R or a randomly generated set? This example is not reproducible. Cheers!

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/renozao/NMF/issues/68#issuecomment-227504044, or mute the thread https://github.com/notifications/unsubscribe/AFYMXgZMiI_7ostIcR1ZP5Ju516NJmvUks5qOBiagaJpZM4I68-i .

omelnikov commented 8 years ago

It seems to work fine in my R 3.3.0 32 bit on Win 7 OS in RStudio 0.99.902. My nmf version is 0.22. Make sure your system and tools are up to date. I'm not sure if nmf on CRAN is the same or lower version. Renaud (author) instructed me on downloading v0.22 in one of the prior threads. Good luck.

> expression.matrix <- matrix(runif(500*500), ncol=500)
> rank = 4
> nrun = 100
> nmf.out <- NMF::nmf(expression.matrix, rank=rank, nrun=nrun)
> nmf.out
<Object of class: NMFfitX1 >
  Method: brunet 
  Runs:  100 
  RNG:
   407L, 2028638175L, 1292011668L, 1418388101L, -1556273790L, -127884517L, 1892023200L 
  Total timing:
   user  system elapsed 
   2.75    1.16  926.53