@jeremiedb Vgg16 would require more than 12GB of GPU memory size and vgg19 requires more than 15GB. This numbers are the GPU memory footprint at the time of crash. However, I tried the gc() fix which you mentioned above and the transfer learning do work fine with GPU memory footprint constant at 1.6GB. I think it makes sense to add the gc() fix in model.R to avoid these crashes. Do you have any better suggestions?
[ Full content available at: https://github.com/apache/incubator-mxnet/issues/7968 ] This message was relayed via gitbox.apache.org for [email protected]
