search

bmild / nerf

Code release for NeRF (Neural Radiance Fields)
http://tancik.com/nerf
MIT License
9.6k stars 1.34k forks source link

Failed to run using --config paper_configs/llff_config.txt #71

Closed shauidu closed 3 years ago

shauidu commented 3 years ago

Hi,

I want to see the best result using your code, so I download the data and run with "paper_configs/llff_config.txt".

Here's the command I use, "python run_nerf.py --config paper_configs/llff_config.txt --chunk 8 --netchunk 16". After a few minutes, I got this error massage, "OOM when allocating tensor with shape[16,256] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [Op:MatMul]".

It seems the GPU memory is not enough, my gpu is 1080Ti with 12GB memory.

BTW, run with the example config just works fine. Any idea what is wrong? Thank you.

shauidu commented 3 years ago

Use a lower N_rand solves the problem.