Closed lixiaopi1985 closed 4 years ago
Thanks for the report and sorry for the delay, I've been on vacation. This looks like a bug, but I'm having trouble reproducing it without seeing the input data set. Would it be possible for you to share it in an obfuscated/reduced form that still reproduces the error?
Thank you for getting back to me. I can't find the data anymore to reproduce the issue, I must have deleted it or over-written it. I'll just close this issue for now.
Sorry to hear that, feel free to re-open if the issue should arise again.
Hi,
Thank you for developing this tool.
I ran into this error during normalization steps:
┌ Warning: 1 factor variable with more than two categories were detected (cul_type), splitting it into separate dummy variables (One Hot) └ @ FlashWeave C:\Users\Xiaoping.julia\packages\FlashWeave\16aV0\src\preprocessing.jl:63 ┌ Warning: adaptive pseudo-counts for 1 samples were lower than machine precision due to insufficient counts, removing them └ @ FlashWeave C:\Users\Xiaoping.julia\packages\FlashWeave\16aV0\src\preprocessing.jl:125
ArgumentError: number of rows of each array must match (got (46, 47)).
Just wondering if the program should have taken care of it internally?
my code for running: learn_network(otu, meta, sensitive=true, heterogeneous=false, normalize=true, alpha=0.05, FDR=true)
Thanks