Model angelo not using GPUs about model-angelo HOT 5 CLOSED

Hi, thank you for using ModelAngelo. You have CUDA 10.1 installed, so you need to modify the installation script in the line

conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch

it should be

conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch

Could you try removing the model_angelo conda environment with conda env remove -n model_angelo and then try running your modified installation script again?

from model-angelo.

Dear Kiarash,
Hi,
thanks to your tip, it worked perfectly, but now I've encountered an another issue, seems the program somehow decided to stop running itself, the output are as follows:

model_angelo build -v Refs/Sharp.mrc -f Fastas/sequence.fa -o OutPut
---------------------------- ModelAngelo -----------------------------
By Kiarash Jamali, Scheres Group, MRC Laboratory of Molecular Biology
--------------------- Initial C-alpha prediction ---------------------
100%|█████████████████████████████████████████| 512/512 [11:55<00:00, 1.40s/it]
------------------ GNN model refinement, round 1 / 3 ------------------
(model_angelo)

It kind of stopped at round 1, any idea what might cause this problem?

Thank you again!
Sincerely,
Linhua Tai

from model-angelo.

Hi,
Yes, seems like it crashed in the GNN round. Can you send me the file OutPut/model_angelo.log please

from model-angelo.

Dear Kiarash,
Hi,
I'm sorry for the stupid mistake, I went into the log file and found out the resason was model angelo doesn't recognize a fasta file with extension name of '.fa', changing the extension to '.fasta' solved the problem, and now the program are running perfectly, I'll let you know if the program ends well.

Thank you again!
Sincerely,
Linhua Tai

from model-angelo.

Great, thank you very much! The very latest version should accept ".fa" extensions as well.

from model-angelo.