Closed jake4844 closed 2 years ago
Hi Jake,
I believe this is OOM (out of memory) error, meaning you do not have enough GPU memory. I am curious what is your cube_size in subtomo.star, usually it should be 64
If the cubesize is normal, I wonder whether your set your tensorflow Channel-First format?
well that solved it! cube size was set to 96 instead of 64. definitely a ram error with the gpus
Hi,
I am having an issue on installing isonet on a certain workstation. The workstation has 2x 3080 Ti's. The driver version is 470.57.02, I've also tried a few different ones. It is also running cuda version 11.4. Tensorflow 2.6.0 with keras 2.6.0. I've installed isonet on many different computers but strangely only having this issue with the 3080 gpus. I attached a copy for the error if this helps!
Thanks, Jakeisonet.txt