You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have two identical GPUs in my machine but when fine tune resnet56 it seems only one of them is used.h
i had try the two cmd
$caffe train -solver solver.prototxt -weight restnet56.caffemodel -gpu all
$caffe train -solver solver.prototxt -weight restnet56.caffemodel -gpu 0,1
watch the cmd
$nvidia-smi
it only show gpu 0 use 89% memory ,and the 1 is 0%
at last, get the error: check failed:error == cudasuccess(2 vs. 0) out of memory
thanks for u help!
The text was updated successfully, but these errors were encountered:
I have two identical GPUs in my machine but when fine tune resnet56 it seems only one of them is used.h
i had try the two cmd
$caffe train -solver solver.prototxt -weight restnet56.caffemodel -gpu all
$caffe train -solver solver.prototxt -weight restnet56.caffemodel -gpu 0,1
watch the cmd
$nvidia-smi
it only show gpu 0 use 89% memory ,and the 1 is 0%
at last, get the error: check failed:error == cudasuccess(2 vs. 0) out of memory
thanks for u help!
The text was updated successfully, but these errors were encountered: