"Out of memory: Killed process" when I trained the Tank&Template dataset.

[honey-zhao]

trafficstars

Hello! For NeRF and NVSF datasets I can train normally, but when I do Tank&Template training, it keeps reporting killed, and by checking the logs, I found that it is because of "out of memory". Later, by checking the memory usage through "top", I found that the memory is indeed overflowing, but the gpu memory is still largely remaining.

[kajalsanklecha]

Hello! I am facing a similar issue. I am using NVIDIA RTX 3090 with 32GB RAM and the system is stuck when I am using 'Truck' from 'TanksAndTemple' dataset. Reducing the batch size is also not working.

[Xzzit]

Try to add downsample_train = 8 in ./configs/truck.txt file. This helps you train the radiance field with 1/8-size images thus saving memory.