search

VinAIResearch / 3D-UCaps

3D-UCaps: 3D Capsules Unet for Volumetric Image Segmentation (MICCAI 2021)
Apache License 2.0
66 stars 11 forks source link

RuntimeError: CUDA error: device-side assert triggered #8

Open jsong0041 opened 2 years ago

jsong0041 commented 2 years ago

First of all, congratulations for your recent paper '3D-UCaps: 3D Capsules Unet for Volumetric Image Segmentation' accepted by MICCAI'21, it's really a great job, and thank you very much for your open source code in github.

As for codes, I used a new dataset as inputs with .tif format, but following errors are thrown:

Validation sanity check: 0%| | 0/1 [00:00<?, ?it/s]C :/w/b/windows/pytorch/aten/src/ATen/native/cuda/ScatterGatherKernel.cu:312: block: [1926,0,0], thread: [32,0,0] Assertion idx_dim >= 0 && idx_dim < index_size && "index out of bounds" failed. ... === Transform input info -- AsDiscrete === Traceback (most recent call last): File "C:\Python36\lib\site-packages\monai\transforms\transform.py", line 84, in apply_transform return _apply_transform(transform, data, unpack_items) File "C:\Python36\lib\site-packages\monai\transforms\transform.py", line 52, in _apply_transform return transform(parameters) File "C:\Python36\lib\site-packages\monai\transforms\post\array.py", line 174, in call img = one_hot(img, num_classes=_nclasses, dim=0) File "C:\Python36\lib\site-packages\monai\networks\utils.py", line 86, in onehot labels = o.scatter(dim=dim, index=labels.long(), value=1) RuntimeError: CUDA error: device-side assert triggered

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "train.py", line 129, in trainer.fit(net, datamodule=data_module) File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 741, in fit self._fit_impl, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 685, in _call_and_handle_interrupt return trainer_fn(*args, kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 777, in _fit_impl self._run(model, ckpt_path=ckpt_path) File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 1199, in _run self._dispatch() File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 1279, in _dispatch self.training_type_plugin.start_training(self) File "C:\Python36\lib\site-packages\pytorch_lightning\plugins\training_type\training_type_plugin.py", line 202, in start_training self._results = trainer.run_stage() File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 1289, in run_stage return self._run_train() File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 1311, in _run_train self._run_sanity_check(self.lightning_module) File "C:\Python36\lib\site-packages\pytorch_lightning\trainer\trainer.py", line 1375, in _run_sanity_check self._evaluation_loop.run() File "C:\Python36\lib\site-packages\pytorch_lightning\loops\base.py", line 145, in run self.advance(*args, *kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\loops\dataloader\evaluation_loop.py", line 110, in advance dl_outputs = self.epoch_loop.run(dataloader, dataloader_idx, dl_max_batches, self.num_dataloaders) File "C:\Python36\lib\site-packages\pytorch_lightning\loops\base.py", line 145, in run self.advance(args, kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\loops\epoch\evaluation_epoch_loop.py", line 122, in advance output = self._evaluation_step(batch, batch_idx, dataloader_idx) File "C:\Python36\lib\site-packages\pytorch_lightning\loops\epoch\evaluation_epoch_loop.py", line 217, in _evaluation_step output = self.trainer.accelerator.validation_step(step_kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\accelerators\accelerator.py", line 239, in validation_step return self.training_type_plugin.validation_step(step_kwargs.values()) File "C:\Python36\lib\site-packages\pytorch_lightning\plugins\training_type\dp.py", line 104, in validation_step return self.model(args, kwargs) File "C:\Python36\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result = self.forward(*input, *kwargs) File "C:\Python36\lib\site-packages\torch\nn\parallel\data_parallel.py", line 159, in forward return self.module(inputs[0], kwargs[0]) File "C:\Python36\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result = self.forward(*input, kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\overrides\data_parallel.py", line 63, in forward output = super().forward(*inputs, *kwargs) File "C:\Python36\lib\site-packages\pytorch_lightning\overrides\base.py", line 92, in forward output = self.module.validation_step(inputs, kwargs) File "E:#project_b\3d-ucaps-master\module\ucaps.py", line 265, in validation_step labels = [self.post_label(label) for label in decollate_batch(labels)] File "E:#project_b\3d-ucaps-master\module\ucaps.py", line 265, in labels = [self.post_label(label) for label in decollatebatch(labels)] File "C:\Python36\lib\site-packages\monai\transforms\compose.py", line 159, in call input = apply_transform(transform, input, self.map_items, self.unpack_items) File "C:\Python36\lib\site-packages\monai\transforms\transform.py", line 107, in apply_transform _log_stats(data=data) File "C:\Python36\lib\site-packages\monai\transforms\transform.py", line 98, in _log_stats datastats(img=data, data_shape=True, value_range=True, prefix=prefix) # type: ignore File "C:\Python36\lib\site-packages\monai\transforms\utility\array.py", line 524, in call lines.append(f"Value range: ({torch.min(img)}, {torch.max(img)})") RuntimeError: CUDA error: device-side assert triggered

Any help is much appreciated.

hoangtan96dl commented 1 year ago

hello @jsong0041 , sorry for my late reply. Based on my experience, you are getting out of memory error in the validation step but it was logged in another way. You can reference issue #6 and the README to find which arguments you can change to make it works.