Hi,
thanks for releasing v0.2.0, I'm finding it to be an improvement and useful! One feature that would be nice is the ability to select the GPUs you want to run training and/or prediction on directly through cryoCARE (unless there is a way to do this that I am unfamiliar with?).
Currently, I set GPU_VISIBLE_DEVICES to the GPU ID I want to use seperately in bash, but a more universal method, i.e. including the feature within cryoCARE itself , may work slightly better for users who can't control their running environment/don't know how. I envision using a flag to specify this, e.g.
cryoCARE_train.py --conf [path] --gpu_id 0,1
Alternatively, it could be entered in the json files.
I can give implementing this a go and submit a PR if you think this is a good idea and are willing to merge if it works?
Thanks,
Euan
Hi, thanks for releasing v0.2.0, I'm finding it to be an improvement and useful! One feature that would be nice is the ability to select the GPUs you want to run training and/or prediction on directly through cryoCARE (unless there is a way to do this that I am unfamiliar with?).
Currently, I set GPU_VISIBLE_DEVICES to the GPU ID I want to use seperately in bash, but a more universal method, i.e. including the feature within cryoCARE itself , may work slightly better for users who can't control their running environment/don't know how. I envision using a flag to specify this, e.g.
cryoCARE_train.py --conf [path] --gpu_id 0,1
Alternatively, it could be entered in the json files.
I can give implementing this a go and submit a PR if you think this is a good idea and are willing to merge if it works? Thanks, Euan