Without GUI how to use it in Linux? how to use it by command line？

[fanzz1208]

just do it , but i don`t know how to save mesh or where to save? i need command of different function without gui

CUDA_VISIBLE_DEVICES=1 python run.py /workspace/instant-ngp/data/rewjj/ --save_mesh PLY 02:05:22 SUCCESS Initialized CUDA. Active GPU is #0: NVIDIA TITAN Xp [61] 02:05:22 INFO Loading NeRF dataset from 02:05:22 INFO /workspace/instant-ngp/data/rewjj/transforms.json 02:05:22 SUCCESS Loaded 100 images after 0s 02:05:22 INFO cam_aabb=[min=[-0.622653,-1.1264,-0.525948], max=[2.14657,1.67218,1.17911]] Training: 0%| | 0/35000 [00:00<?, ?steps/s$ 02:05:23 INFO Loading network config from: /workspace/instant-ngp/configs/nerf/base.json 02:05:23 INFO GridEncoding: Nmin=16 b=2 F=4 T=2^19 L=8 Warning: FullyFusedMLP is not supported for the selected architecture 61. Falling back to CutlassMLP. For maximum performance, raise the target GP U architecture to 75+. Warning: FullyFusedMLP is not supported for the selected architecture 61. Falling back to CutlassMLP. For maximum performance, raise the target GP U architecture to 75+. Warning: FullyFusedMLP is not supported for the selected architecture 61. Falling back to CutlassMLP. For maximum performance, raise the target GP U architecture to 75+. 02:05:23 INFO Density model: 3--[HashGrid]-->32--[FullyFusedMLP(neurons=64,layers=3)]-->1 02:05:23 INFO Color model: 3--[Composite]-->16+16--[FullyFusedMLP(neurons=64,layers=4)]-->3 02:05:23 INFO total_encoding_params=11681792 total_network_params=10240 Training: 100%|____________________________________________________________________________| 34997/35000 [26:45<00:00, 21.80steps/s, loss=0.00296] Generating mesh via marching cubes and saving to PLY. Resolution=[256,256,256], Density Threshold=2.5 02:32:09 INFO #vertices=587121 #triangles=1120198

[maomaocun]

ME TOO!