Finally got it to compile properly, but the most basic inference example fails:
(plenoxel) ubuntu$: ~/svox2/opt$ python render_imgs.py ../../256_to_512_fasttv/hotdog/ckpt.npz ../../nerf_synthetic/hotdog/val/
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /home/ubuntu/.cache/torch/hub/checkpoints/vgg16-397923af.pth
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████| 528M/528M [00:01<00:00, 280MB/s]
Loading model from: /home/ubuntu/.conda/envs/plenoxel/lib/python3.8/site-packages/lpips/weights/v0.1/vgg.pth
Defaulting to extended NSVF dataset
LOAD NSVF DATA ../../nerf_synthetic/hotdog/val/ split test
Traceback (most recent call last):
File "render_imgs.py", line 111, in <module>
dset = datasets[args.dataset_type](args.data_dir, split="test_train" if args.train else "test",
File "/home/ubuntu/svox2/opt/util/dataset.py", line 20, in auto_dataset
return NSVFDataset(root, *args, **kwargs)
File "/home/ubuntu/svox2/opt/util/nsvf_dataset.py", line 82, in __init__
img_dir_name = look_for_dir(["images", "image", "rgb"])
File "/home/ubuntu/svox2/opt/util/nsvf_dataset.py", line 79, in look_for_dir
assert False, "None of " + str(cands) + " found in data directory"
AssertionError: None of ['images', 'image', 'rgb'] found in data directory
The documentation is lacking here, do I have the wrong files or command structure?
Finally got it to compile properly, but the most basic inference example fails:
The documentation is lacking here, do I have the wrong files or command structure?