Loading donut transformers model getting error

vision
1

self.model = self.model.to(self.device)
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 1145, in to
return self._apply(convert)
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 797, in _apply
module._apply(fn)
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 797, in _apply
module._apply(fn)
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 797, in _apply
module._apply(fn)
[Previous line repeated 1 more time]
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 820, in _apply
param_applied = fn(param)
File “/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py”, line 1143, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
RuntimeError: CUDA error: device-side assert triggered
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

my code is
class Donut:
def init(self):
self.processor = DonutProcessor.from_pretrained(“naver-clova-ix/donut-base-finetuned-docvqa”)
self.model = VisionEncoderDecoderModel.from_pretrained(“naver-clova-ix/donut-base-finetuned-docvqa”)
self.device = “cuda” if torch.cuda.is_available() else “cpu”
self.model.to(self.device)

torch version :- torch==2.0.1+cu118
torchaudio==2.0.2+cu118
torchvision==0.15.2+cu118
Nvidia Driver :- Driver Version: 550.144.03 CUDA Version: 12.4

2

Often device asserts are triggered by failing indexing operations so check if e.g. an embedding layer received an invalid input or any other indexing fails.

3

Thanks for reply. It has resolve , my max_position_embeddings mismatching.