search

PrincetonUniversity / DP_GP_cluster

BSD 3-Clause "New" or "Revised" License
73 stars 25 forks source link

ValueError: need at least one array to concatenate #11

Open cartal opened 6 years ago

cartal commented 6 years ago

Hi,

I'm running DP_GP_cluster with 19 samples and ~ 8000 genes. This has worked in the past, although it took a long time. However, this time I get the following error:

screen shot 2018-06-04 at 09 48 25

Any ideas of what I may be doing wrong?

annashcherbina commented 5 years ago

I had the same error when all gene values in a row were 0 and also when all gene values in a row where the same (i.e. variance was 0). Once I removed those rows, the error went away.

MattPinder commented 2 years ago

Sorry to add to a years-old thread, but I'm receiving the same error message under what seems to be slightly different circumstances, and wondered if anyone could offer suggestions. I'm running DP_GP_cluster on two datasets, each with six timepoints and ~20,000 genes. If any NA values are present, I receive the above error, and it works fine if I remove all NAs from the input. As a test I substitited a single value in the 'no NAs' subset with NA, and once again I received the above error. As the main page for DP_GP_cluster says that it can handle missing values, I wondered if the problem is with my setup. The versions of the relevant software on my system are:

dstern commented 1 year ago

I had a similar error with a matrix with blank space instead of tab separation. Tab separated file fixed error.