error while training in multiple GPUs

[alaa-shubbak]

hello , I try to train faster-RCNN model on my dataset using the following script (which is the same on your repository):

for FOLD in 1 2 3 4 5;
do
  bash tools/dist_train_partially.sh semi ${FOLD} 10 8
done

but i got the following error during the training:

any help to understand this error and solve it , please ?

[mary-0830]

请问您解决了吗？

[alaa-shubbak]

yes, I did it . the problem was in two side , one related to my dataset which does not annotate by segmentation (mask) , and i used the code with segmentation by mistake. the other side was related to connect my code with wandb. you shall have your account there in case you want to use wandb other wise make it unused in the code.