CUDA out of memory on 8xv100 32G GPUs for r101 training

[rambleramble]

CUDA out of memory with the current source code on 8xv100 32G

However, it seems the following code, need to use r101 resolution, instead of r50 to fit into 32G of V100.

VisTR/datasets/ytvos.py

Line 167 in 554e711

 #T.RandomResize([280], max_size=504),#for r101 

My question is that the claimed result of r101, is it achieved under the r101 resolution, aka
T.RandomResize([280], max_size=504),#for r101

or the default source code one:
T.RandomResize([300], max_size=540),#for r50

thanks in advance!!

[Yuqing Wang]

Hi @rambleramble, the result is achieved by the resolution of r101