search

VITA-Group / GNT

[ICLR 2023] "Is Attention All NeRF Needs?" by Mukund Varma T*, Peihao Wang* , Xuxi Chen, Tianlong Chen, Subhashini Venugopalan, Zhangyang Wang
https://vita-group.github.io/GNT
MIT License
340 stars 25 forks source link

loaded state dict contains a parameter group that doesn't match the size of optimizer's group #17

Closed leo-frank closed 1 year ago

leo-frank commented 1 year ago

Descrpiton

I use the pretrained generalization model at https://drive.google.com/file/d/1AMN0diPeHvf2fw53IO5EE2Qp4os5SkoX/view?usp=share_link. The command is CUDA_VISIBLE_DEVICES=2 python3 eval.py --config configs/gnt_blender.txt \ --eval_dataset nerf_synthetic \ --eval_scenes mic --run_val \ --expname gnt_author_pretrained_cross_mic \ --ckpt_path /home/cbe/lwy/GNT/out/pretrained_cross_720000.pth

Bug

  File "/home/cbe/lwy/GNT/gnt/model.py", line 133, in load_from_ckpt
    self.load_model(fpath, load_opt, load_scheduler)
  File "/home/cbe/lwy/GNT/gnt/model.py", line 102, in load_model
    self.optimizer.load_state_dict(to_load["optimizer"])
  File "/home/cbe/miniconda3/envs/vision/lib/python3.8/site-packages/torch/optim/optimizer.py", line 146, in load_state_dict
    raise ValueError("loaded state dict contains a parameter group "
ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group

Questions

I wonder if you can run the pretrained generalization model ? or why I met such problem?

Related Issue

before renamed, #8 also met same problem when he/she runs render.py, but i am running eval.py.

Thanks for your great work !!!

leo-frank commented 1 year ago

sorry to bother you , i found the reason, it's because that the config file is not mismatched, should use --config configs/gnt_full.txt

Thanks for your great work again !!!