Training Iterations Never Start

[bennett-taniguchi]

I previously ran ns-train nerfacto on chair and then I ran ns-train in2n --data chair --load-dir outputs\chair\nerfacto\2023-09-26_201523\nerfstudio_models --prompt "Make the chair made of leather" on the folder. After this happens I get the following :

Saving config to: outputs\chair\in2n\2023-09-28_192314\config.yml                    experiment_config.py:141
           Saving checkpoints to: outputs\chair\in2n\2023-09-28_192314\nerfstudio_models                  trainer.py:134
           Auto image downscale factor of 1                                                 nerfstudio_dataparser.py:349
            Dataset is overriding orientation method to none                                nerfstudio_dataparser.py:229
            Dataset is overriding orientation method to none                                nerfstudio_dataparser.py:229
Setting up training dataset...
Caching all 26 images.
Setting up evaluation dataset...                                                                                                                                                                               
Caching all 2 images.
unet\diffusion_pytorch_model.safetensors not found                                                                                                                                                             
Loading pipeline components...: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 6/6 [00:02<00:00,  2.64it/s]
InstructPix2Pix loaded!
╭─────────────────────────────────────────── Viewer ───────────────────────────────────────────╮
│        ╷                                                                                     │
│   HTTP │ https://viewer.nerf.studio/versions/23-05-15-1/?websocket_url=ws://localhost:7007   │
│        ╵                                                                                     │
╰──────────────────────────────────────────────────────────────────────────────────────────────╯
[NOTE] Not running eval iterations since only viewer is enabled.
Use --vis {wandb, tensorboard, viewer+wandb, viewer+tensorboard} to run with eval.
Loading latest Nerfstudio checkpoint from load_dir...
Done loading Nerfstudio checkpoint from outputs\chair\nerfacto\2023-09-26_201523\nerfstudio_models\step-000029999.ckpt
Disabled comet/tensorboard/wandb event writers
C:\Python\lib\site-packages\torch\functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ..\aten\src\ATen\native\TensorShape.cpp:3484.)
  return _VF.meshgrid(tensors, **kwargs

which doesn't update with training iterations and when I look on the browser. Also the browser says Iteration:0 & Resolution:?

[ayaanzhaque]

This is quite odd... did the viewer work when you were training the regular nerfacto model?

[bennett-taniguchi]

Yeah, the viewer works on the regular one and it seems like it's stalled on something for me.

[ayaanzhaque]

Is this still an issue?