search

ubc-vision / nf-soft-mining

(CVPR 2024) Accelerating Neural Field Training via Soft Mining
https://ubc-vision.github.io/nf-soft-mining/
Apache License 2.0
35 stars 1 forks source link

Several Issues about execution #1

Closed RPFey closed 10 months ago

RPFey commented 10 months ago

Hi, thank you for sharing the code ! There are still several mistakes in the current version.

  1. I cannot find LLFF_NDC_SCENES in examples.utils , which is imported in examples/utils.py.

  2. I cannot find nerfacc.losses, which is imported in examples/losses.py.

  3. When running the LMC sampling method, I get the error when setting the proposal_requires_grad variable to True

RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward.

Thank you in advance if you could fix these bugs !

shakibakh commented 10 months ago

@RPFey Thank you for the comment. A few local changes haven't been pushed before. The current main branch should be ok. I will close the issue. Feel free to reopen it if there is any problem.