Closed DavieHR closed 8 years ago
Hello, this is probably because if cudnn autotune, it need some memory to test methods, did you try nn backend? Does it also crash?
Yes i also tried nn backend,however, it still crashed
Got the same error, no idea, probably the implementation of something has been changed, -batch_size 15
works for me.
Hi. out of memory at /home/whr/torch-cl/extra/cutorch/lib/THC/generic/THCStorage.cu:40 when i run the "th texture_train.lua -texture data/textures/red-peppers256.o.jpg -model_name pyramid -backend cudnn -num_iterations 1500 -vgg_no_pad true -normalize_gradients true". My GPU is Nvidia Black Titan X. What's wrong? Thx.