im experimenting with training of faster rcnn + Vgg16 model. At the moment i found a way to run it on a single GPU only, via --gpu parameter. i was wondering whether there is way to use multiple GPUs ( which i apparently have ).
I'm not sure about your configuration but a similar question has been asked and answered here #107 about the multi-gpu support. It seems like it isn't possible, due to the Python layer dependency.
Hello Ross !
im experimenting with training of faster rcnn + Vgg16 model. At the moment i found a way to run it on a single GPU only, via --gpu parameter. i was wondering whether there is way to use multiple GPUs ( which i apparently have ).
Please advise.
thank you in advance, Yulia