nms不存在！！！我在win系统下边实现这个examples/instruct_pix2pix的报错RuntimeError: operator torchvision::nms does not exist

[CSolaris]

Describe the bug

(img2img) PS C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix> accelerate launch --mixed_precision="fp16" train_instruct_pix2pix.py `

--pretrained_model_name_or_path=$env:MODEL_NAME --dataset_name=$env:DATASET_ID
--enable_xformers_memory_efficient_attention --resolution=256 --random_flip
--train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing --max_train_steps=15000
--checkpointing_steps=5000 --checkpoints_total_limit=1 --learning_rate=5e-05 --max_grad_norm=1 --lr_warmup_steps=0
--conditioning_dropout_prob=0.05 --mixed_precision=fp16
--seed=42
Traceback (most recent call last):
File "C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix\train_instruct_pix2pix.py", line 43, in
from torchvision import transforms
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torchvision_init_.py", line 10, in
from torchvision import _meta_registrations, datasets, io, models, ops, transforms, utils # usort:skip
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torchvision_meta_registrations.py", line 163, in
@torch.library.register_fake("torchvision::nms")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch\library.py", line 1023, in register
use_lib._register_fake(op_name, func, _stacklevel=stacklevel + 1)
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch\library.py", line 214, in _register_fake
handle = entry.fake_impl.register(func_to_register, source)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch_library\fake_impl.py", line 31, in register
if torch._C._dispatch_has_kernel_for_dispatch_key(self.qualname, "Meta"):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: operator torchvision::nms does not exist
Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in run_code
File "C:\Study\Study\MiniConda\envs\img2img\Scripts\accelerate.exe_main.py", line 6, in
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\accelerate_cli.py", line 50, in main
args.func(args)
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\launch.py", line 1235, in launch_command
simple_launcher(args)
File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\launch.py", line 823, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['C:\Study\Study\MiniConda\envs\img2img\python.exe', 'train_instruct_pix2pix.py', '--pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5', '--dataset_name=fusing/instructpix2pix-1000-samples', '--enable_xfor
mers_memory_efficient_attention', '--resolution=256', '--random_flip', '--train_batch_size=4', '--gradient_accumulation_steps=4', '--gradient_checkpointing', '--max_train_steps=15000', '--checkpointing_steps=5000', '--checkpoints_total_limit=1', '--learning_rate=5e-05', '--max_grad_norm=1', '--lr_warmup_steps=0', '--conditioning_dropout_prob=0.05', '--mixed_precision=fp16', '--seed=42']' returned non-zero exit status 1.
哪个大佬可以帮帮我

Reproduction

accelerate launch --mixed_precision="fp16" train_instruct_pix2pix.py --pretrained_model_name_or_path="$env:MODEL_NAME"
--dataset_name="$env:DATASET_ID" --enable_xformers_memory_efficient_attention
--resolution=256 --random_flip --train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing
--max_train_steps=15000 --checkpointing_steps=5000 --checkpoints_total_limit=1
--learning_rate=5e-05 --max_grad_norm=1 --lr_warmup_steps=0 --conditioning_dropout_prob=0.05
--mixed_precision=fp16 `
--seed=42

Logs

(img2img) PS C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix> $env:MODEL_NAME = "runwayml/stable-diffusion-v1-5"
(img2img) PS C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix> $env:DATASET_ID = "fusing/instructpix2pix-1000-samples"
(img2img) PS C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix> accelerate launch --mixed_precision="fp16" train_instruct_pix2pix.py `
>>   --pretrained_model_name_or_path=$env:MODEL_NAME `
>>   --dataset_name=$env:DATASET_ID `
>>   --enable_xformers_memory_efficient_attention `
>>   --resolution=256 --random_flip `
>>   --train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing `
>>   --max_train_steps=15000 `
>>   --checkpointing_steps=5000 --checkpoints_total_limit=1 `
>>   --learning_rate=5e-05 --max_grad_norm=1 --lr_warmup_steps=0 `
>>   --conditioning_dropout_prob=0.05 `
>>   --mixed_precision=fp16 `
>>   --seed=42
Traceback (most recent call last):
  File "C:\Study\Study_Documents\PyCharm\diffusers-main\examples\instruct_pix2pix\train_instruct_pix2pix.py", line 43, in <module>
    from torchvision import transforms
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torchvision\__init__.py", line 10, in <module>
    from torchvision import _meta_registrations, datasets, io, models, ops, transforms, utils  # usort:skip
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torchvision\_meta_registrations.py", line 163, in <module>
    @torch.library.register_fake("torchvision::nms")
     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch\library.py", line 1023, in register
    use_lib._register_fake(op_name, func, _stacklevel=stacklevel + 1)
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch\library.py", line 214, in _register_fake
    handle = entry.fake_impl.register(func_to_register, source)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\torch\_library\fake_impl.py", line 31, in register
    if torch._C._dispatch_has_kernel_for_dispatch_key(self.qualname, "Meta"):
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: operator torchvision::nms does not exist
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "C:\Study\Study\MiniConda\envs\img2img\Scripts\accelerate.exe\__main__.py", line 6, in <module>
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\accelerate_cli.py", line 50, in main
    args.func(args)
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\launch.py", line 1235, in launch_command
    simple_launcher(args)
  File "C:\Study\Study\MiniConda\envs\img2img\Lib\site-packages\accelerate\commands\launch.py", line 823, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['C:\\Study\\Study\\MiniConda\\envs\\img2img\\python.exe', 'train_instruct_pix2pix.py', '--pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5', '--dataset_name=fusing/instructpix2pix-1000-samples', '--enable_xformers_memory_efficient_attention', '--resolution=256', '--random_flip', '--train_batch_size=4', '--gradient_accumulation_steps=4', '--gradient_checkpointing', '--max_train_steps=15000', '--checkpointing_steps=5000', '--checkpoints_total_limit=1', '--learning_rate=5e-05', '--max_grad_norm=1', '--lr_warmup_steps=0', '--conditioning_dropout_prob=0.05', '--mixed_precision=fp16', '--seed=42']' returned non-zero exit status 1.

System Info

diffusers 0.36.0.dev0
win11
pip3 install --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/cu130
5090显卡
python3.11.13

Who can help?

No response

[SahilCarterr]

Hey @CSolaris runwayml/stable-diffusion-v1-5 is depretiated use Lykon/dreamshaper-8 or stable-diffusion-v1-5/stable-diffusion-v1-5

[CSolaris]

thanks，it was solved At 2025-10-30 19:27:57, "SahilCarterr" ***@***.***> wrote: SahilCarterr left a comment (huggingface/diffusers#12559) Hey @CSolarisrunwayml/stable-diffusion-v1-5 is depretiated use Lykon/dreamshaper-8 or stable-diffusion-v1-5/stable-diffusion-v1-5 — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>