Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Errors saving models after optimization #2

Open
gradeeterna opened this issue Sep 25, 2024 · 0 comments
Open

Errors saving models after optimization #2

gradeeterna opened this issue Sep 25, 2024 · 0 comments

Comments

@gradeeterna
Copy link

Hey,

I'm getting the following errors at the saving model part after optimizing for 15k steps.

Saving model... Traceback (most recent call last): File "H:\Frosting\train.py", line 169, in <module> coarse_sugar_path = coarse_training_with_density_regularization_and_dn_consistency(coarse_args) File "H:\Frosting\frosting_trainers\coarse_density_and_dn_consistency.py", line 902, in coarse_training_with_density_regularization_and_dn_consistency sugar.save_model(path=model_path, File "H:\Frosting\frosting_scene\sugar_model.py", line 2394, in save_model torch.save(checkpoint, path) File "H:\anacondaH\frosting\lib\site-packages\torch\serialization.py", line 440, in save with _open_zipfile_writer(f) as opened_zipfile: File "H:\anacondaH\frosting\lib\site-packages\torch\serialization.py", line 315, in _open_zipfile_writer return container(name_or_buffer) File "H:\anacondaH\frosting\lib\site-packages\torch\serialization.py", line 288, in __init__ super().__init__(torch._C.PyTorchFileWriter(str(name))) RuntimeError: Parent directory ./output/coarse\allotment\sugarcoarse_3Dgs7000_densityestim02_sdfnorm02 does not exist.

I also had to use --gs_output_dir to load an already trained vanilla gs model, as when I try to train from scratch without that argument I get these errors:

Loading config output\vanilla_gs\data/allotment\... Traceback (most recent call last): File "H:\Frosting\train.py", line 169, in <module> coarse_sugar_path = coarse_training_with_density_regularization_and_dn_consistency(coarse_args) File "H:\Frosting\frosting_trainers\coarse_density_and_dn_consistency.py", line 377, in coarse_training_with_density_regularization_and_dn_consistency nerfmodel = GaussianSplattingWrapper( File "H:\Frosting\frosting_scene\gs_model.py", line 132, in __init__ cam_list = load_gs_cameras( File "H:\Frosting\frosting_scene\cameras.py", line 34, in load_gs_cameras with open(gs_output_path + 'cameras.json') as f: FileNotFoundError: [Errno 2] No such file or directory: 'output\\vanilla_gs\\data/allotment\\cameras.json'

Not sure if related, but I'm using pytorch3d 0.7.5 instead of 0.7.4, as "conda install pytorch3d==0.7.4 -c pytorch3d" did not find any packages and I failed to bild it myself, so I used this method instead which only has 0.7.5 up -facebookresearch/pytorch3d#1752

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant