can i use this in windows 10?

[vincentchen2017]

i use this in windows10，conda，python 3.10，when i run pip install -r requirements.txt ,it works well，and then i run python app.py error occur，
Traceback (most recent call last):
File "D:\simon\project\One-2-3-45\demo\app.py", line 639, in
fire.Fire(run_demo)
File "C:\Users\u.conda\envs\one2345\lib\site-packages\fire\core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "C:\Users\u.conda\envs\one2345\lib\site-packages\fire\core.py", line 475, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "C:\Users\u.conda\envs\one2345\lib\site-packages\fire\core.py", line 691, in _CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "D:\simon\project\One-2-3-45\demo\app.py", line 449, in run_demo
models = init_model(device, os.path.join(code_dir, 'zero123-xl.ckpt'), half_precision=HALF_PRECISION)
File "..\utils\zero123_utils.py", line 45, in init_model
models['turncam'] = torch.compile(load_model_from_config(config, ckpt, device=device)).half()
File "C:\Users\u.conda\envs\one2345\lib\site-packages\torch_init.py", line 1441, in compile
return torch._dynamo.optimize(backend=backend, nopython=fullgraph, dynamic=dynamic, disable=disable)(model)
File "C:\Users\u.conda\envs\one2345\lib\site-packages\torch_dynamo\eval_frame.py", line 413, in optimize
check_if_dynamo_supported()
File "C:\Users\u.conda\envs\one2345\lib\site-packages\torch_dynamo\eval_frame.py", line 375, in check_if_dynamo_supported
raise RuntimeError("Windows not yet supported for torch.compile")
RuntimeError: Windows not yet supported for torch.compile。
what should i do ？my torch is 2.0.1+cu118

[vincentchen2017]

i change python version to python 3.9,python 3.8，and install it again，but more error occur。

[Dustinpro]

You may delete the torch.compile().

models['turncam'] = load_model_from_config(config, ckpt, device=device).half()

[vincentchen2017]

thanks,i will try it later。now i run this in docker and it works well.

[MHassan1122]

@vincentchen2017
can you please tell me about the format of input image i tried with obj and glb format but doesn't work it works for png and jpg 2d images.

[vincentchen2017]

I use images in png format

[vincentchen2017]

if you use on windows,do this:

1. modify One-2-3-45\app\demo.py, change with open('instructions_12345.md', 'r') as f: to
   with open('instructions_12345.md', 'r', encoding='utf-8') as f:

2.modify One-2-3-45\utils\zero123_utils.py,change
if half_precision:
models['turncam'] = torch.compile(load_model_from_config(config, ckpt, device=device)).half()
else:
models['turncam'] = torch.compile(load_model_from_config(config, ckpt, device=device))
to
if half_precision:
# models['turncam'] = torch.compile(load_model_from_config(config, ckpt, device=device)).half()
models['turncam'] = load_model_from_config(config, ckpt, device=device).half()
else:
# models['turncam'] = torch.compile(load_model_from_config(config, ckpt, device=device))
models['turncam'] = load_model_from_config(config, ckpt, device=device)

3.download this file https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2net.onnx and put it at C:\Users\your user.u2net

[MHassan1122]

dear I followed your steps but it give me the this
as you can see in last line mesh saved to the directory but there is no mesh.ply image in the mentioned directory and the other things is 'CUDA_VISIBLE_DEVICES' is not recognized as an internal or external command,
operable program or batch file.
i have cuda toolkit but confused how this not recongnized cuda....

C:\Users\admin\One-2-3-45-master>python run.py --img_path ./building.jpg --half_precision
Instantiating LatentDiffusion...
Loading model from zero123-xl.ckpt
C:\Users\admin\anaconda3\envs\flexicubes\lib\site-packages\torchaudio\backend\utils.py:74: UserWarning: No audio backend is available.
warnings.warn("No audio backend is available.")
Global Step: 122000
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.53 M params.
Keeping EMAs of 688.
making attention of type 'vanilla' with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type 'vanilla' with 512 in_channels
Instantiating StableDiffusionSafetyChecker...
C:\Users\admin\anaconda3\envs\flexicubes\lib\site-packages\torch_utils.py:776: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
return self.fget.get(instance, owner)()
SAM Time: 3.513s
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 76 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 76/76 [00:11<00:00, 6.84it/s]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [00:34<00:00, 1.41it/s]
2024-01-26 20:03:17.630 | INFO | elevation_estimate.utils.elev_est_api:get_feature_matcher:25 - Loading feature matcher...
2024-01-26 20:03:17.787 | INFO | elevation_estimate.utils.elev_est_api:mask_out_bkgd:48 - Image has no alpha channel, using thresholding to mask out background
2024-01-26 20:03:17.789 | INFO | elevation_estimate.utils.elev_est_api:mask_out_bkgd:48 - Image has no alpha channel, using thresholding to mask out background
2024-01-26 20:03:17.791 | INFO | elevation_estimate.utils.elev_est_api:mask_out_bkgd:48 - Image has no alpha channel, using thresholding to mask out background
2024-01-26 20:03:17.794 | INFO | elevation_estimate.utils.elev_est_api:mask_out_bkgd:48 - Image has no alpha channel, using thresholding to mask out background
2024-01-26 20:03:19.368 | WARNING | elevation_estimate.utils.elev_est_api:elev_est_api:199 - K is not provided, using default K
Estimated polar angle: 99
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 76 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 76/76 [01:34<00:00, 1.25s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [00:31<00:00, 1.56it/s]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [01:14<00:00, 1.52s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [01:37<00:00, 1.99s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [01:02<00:00, 1.28s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [01:42<00:00, 2.08s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [00:59<00:00, 1.21s/it]
Data shape for DDIM sampling is (4, 4, 32, 32), eta 1.0
Running DDIM Sampling with 49 timesteps
DDIM Sampler: 100%|████████████████████████████████████████████████████████████████████| 49/49 [01:37<00:00, 1.99s/it]
CUDA_VISIBLE_DEVICES=0 python exp_runner_generic_blender_val.py --specific_dataset_name C:\Users\admin\One-2-3-45-master\exp\building --mode export_mesh --conf confs/one2345_lod0_val_demo.conf --resolution 256
'CUDA_VISIBLE_DEVICES' is not recognized as an internal or external command,
operable program or batch file.
Mesh saved to: C:\Users\admin\One-2-3-45-master\exp\building\mesh.ply

[MHassan1122]

here you can see in picture it will give me different 2d images but mesh.ply has some problems in saving

[Dustinpro]

CUDA_VISIBLE_DEVICES=idx led to an error. You may delete it.