search

GoekeLab / m6anet

Detection of m6A from direct RNA-Seq data
https://m6anet.readthedocs.io/
MIT License
100 stars 17 forks source link

TypeError: argument should be integer or None, not 'numpy.float64' #160

Closed rugilemat closed 3 months ago

rugilemat commented 3 months ago

Hi, I've been able to run m6anet 2.1.0 without any issues. However, it suddenly started failing with TypeError: argument should be integer or None, not 'numpy.float64' when running m6anet inference on HPC.

My command is as follows: m6anet inference --input_dir ${input} --out_dir ${output} --pretrained_model HEK293T_RNA004 --n_processes 20 --num_iterations 5

I had m6anet installed through pip but the same error persists when running a conda installed version. My python version is 3.8.

I'm not entirely sure if it's something that's failing on our HPC end or something else is going on, so any advice would be greatly appreciated!

yuukiiwa commented 3 months ago

Hi @rugilemat,

Can you try removing --num_iterations 5 from your command and see whether that solves the problem?

m6anet inference --input_dir ${input} --out_dir ${output} --pretrained_model HEK293T_RNA004 --n_processes 20

Thanks!

Best wishes, Yuk Kei

rugilemat commented 3 months ago

Hi @yuukiiwa,

Thanks for this! I think the issue was with the files generated from m6anet dataprep as I didn't have this issue with a different dataset and rerunning m6anet dataprep on the file that was producing the issue solved it. If I encounter something similar next time, I'll try dropping --num_iterations 5.