You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
> 1. assuming your individual submodules were trained with a checkpoint directory such as "/foo/submodule-0", /foo/submodule-1", etc, you'd pass in "/foo/submodule-" as the prefix and the merging script will iterate through all of the trained submodule directories.
My input checkpoint directory is "--ckpt_prefix logs/exp/ex-", but the root seems error:
Traceback (most recent call last):
File "/home/paper/Desktop/meganerf/scripts/merge_submodules.py", line 113, in <module>
main(_get_merge_opts())
File "/home/paper/anaconda3/envs/mega-nerf/lib/python3.9/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "/home/paper/Desktop/meganerf/scripts/merge_submodules.py", line 51, in main
version_dirs = sorted([int(x.name) for x in list(centroid_path.iterdir())], reverse=True)
File "/home/paper/Desktop/meganerf/scripts/merge_submodules.py", line 51, in <listcomp>
version_dirs = sorted([int(x.name) for x in list(centroid_path.iterdir())], reverse=True)
ValueError: invalid literal for int() with base 10: 'hparams.txt'
what the meaning of version_dirs ?
Could you please help me solve this problem? Thanks a lot.
The text was updated successfully, but these errors were encountered:
SSXXSSXX
changed the title
> 1. assuming your individual submodules were trained with a checkpoint directory such as "/foo/submodule-0", /foo/submodule-1", etc, you'd pass in "/foo/submodule-" as the prefix and the merging script will iterate through all of the trained submodule directories.
ValueError: invalid literal for int() with base 10: 'hparams.txt'
Apr 25, 2024
My input checkpoint directory is "--ckpt_prefix logs/exp/ex-", but the root seems error:
what the meaning of
version_dirs
?Could you please help me solve this problem? Thanks a lot.
Originally posted by @Kikihqq in #31 (comment)
The text was updated successfully, but these errors were encountered: