Closed yurikleb closed 7 years ago
I'm not super familiar with Tensorflow's allocator and scheduler, so I'm not sure.
On Jun 13, 2017, at 12:22 AM, yurikleb notifications@github.com wrote:
Hi, this is amazing piece of work! I have a question regarding GPU memory allocation when running neural_style.py it indicates i have 1.5gb available, but right after it starts iteration 1 it says: ran out of memory trying to allocate 975.59MiB ran out of memory trying to allocate 278.55MiB ect..
the script runs fine but I wonder if this memory allocation issue affects performance?
please see attached screen shot.
Many thanks!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.
Hi, this is amazing piece of work! I have a question regarding GPU memory allocation when running neural_style.py it indicates I have 1.5gb available, but right after it starts iteration 1 it says:
ran out of memory trying to allocate 975.59MiB
ran out of memory trying to allocate 278.55MiB
ect..the script runs fine but I wonder if this memory allocation issue affects performance?
please see attached screen shot.
Many thanks!