训练D-NeRF模型时报错
报错内容：

Traceback (most recent call last):
  File "run_dnerf.py", line 957, in <module>
    train()
  File "run_dnerf.py", line 694, in train
    savedir=testsavedir, render_factor=args.render_factor, save_also_gt=True)
  File "run_dnerf.py", line 185, in render_path
    rgb, disp, acc, _ = render(H, W, focal, chunk=chunk, c2w=c2w[:3,:4], frame_time=frame_time, **render_kwargs)
  File "run_dnerf.py", line 151, in render
    all_ret = batchify_rays(rays, chunk, **kwargs)
  File "run_dnerf.py", line 83, in batchify_rays
    ret = render_rays(rays_flat[i:i+chunk], **kwargs)
  File "run_dnerf.py", line 449, in render_rays
    z_samples = sample_pdf(z_vals_mid, weights[...,1:-1], N_importance, det=(perturb==0.), pytest=pytest)
  File "E:NeRFD-NeRF-main
un_dnerf_helpers.py", line 332, in sample_pdf
    inds = searchsorted(cdf, u, side='right')
  File "E:Anacondaenvsdnerflibsite-packages	orchsearchsortedsearchsorted.py", line 41, in searchsorted
    raise Exception('torchsearchsorted on CUDA device is asked, but it seems '
Exception: torchsearchsorted on CUDA device is asked, but it seems that it is not available. Please install it

查了一下torchsearchsorted官方的说明，是找不到对应的CUDA版本导致。
后面又查了一下，发现pytorch在1.7以后已经内置了一个torch.searchsorted模块，不用再另外单独安装torchsearchsorted了，因此此处只需要修改一下代码就行。
找到run_dnerf_helpers.py中的报错位置，将：

inds = searchsorted(cdf, u, side='right')

改为：

inds = torch.searchsorted(cdf, u, right=True)

然后注释掉run_dnerf_helpers.py开头导入的torchsearchsorted包。
之后就能成功运行。