Error when saving images using ensamble evaluator

[cugwu]

Hi, I've implemented the tutorial https://github.com/Project-MONAI/tutorials/blob/main/modules/cross_validation_models_ensemble.ipynb
related to the ensemble evaluator. When I try to save the predicted images, following the suggestion I received in a previous discussion I've opened (#4806):
from monai.transforms import SaveImaged
mean_post_transforms = Compose([
......
SaveImaged(keys="pred", output_dir="./out", output_postfix="seg", resample=False)
])

I have the following error:

Traceback (most recent call last):
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 89, in apply_transform
return _apply_transform(transform, data, unpack_items)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 53, in _apply_transform
return transform(parameters)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/io/dictionary.py", line 271, in call
meta_data = d[meta_key] if meta_key is not None else None
KeyError: 'pred_meta_dict'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 89, in apply_transform
return _apply_transform(transform, data, unpack_items)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 53, in apply_transform
return transform(parameters)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/compose.py", line 173, in call
input = apply_transform(transform, input, self.map_items, self.unpack_items, self.log_stats)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 113, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.io.dictionary.SaveImaged object at 0x7f78ecf53eb0>

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "test.py", line 349, in
main()
File "test.py", line 287, in main
ensemble_evaluate(mean_post_transforms, models, torch.device(device), test_loader,
File "test.py", line 132, in ensemble_evaluate
evaluator.run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/evaluator.py", line 147, in run
super().run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/workflow.py", line 290, in run
super().run(data=self.data_loader, max_epochs=self.state.max_epochs)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 704, in run
return self._internal_run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 783, in _internal_run
self._handle_exception(e)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 466, in _handle_exception
raise e
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 753, in _internal_run
time_taken = self._run_once_on_dataset()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 854, in _run_once_on_dataset
self._handle_exception(e)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 466, in _handle_exception
raise e
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 840, in _run_once_on_dataset
self.state.output = self._process_function(self, self.state.batch)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/evaluator.py", line 442, in _iteration
engine.fire_event(IterationEvents.MODEL_COMPLETED)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 445, in fire_event
return self._fire_event(event_name)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 421, in _fire_event
func(*first, *(event_args + others), **kwargs)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/workflow.py", line 234, in _run_postprocessing
engine.state.batch[i], engine.state.output[i] = engine_apply_transform(b, o, posttrans)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/utils.py", line 247, in engine_apply_transform
transformed_data = apply_transform(transform, data)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 113, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.compose.Compose object at 0x7f78ecf53d60>

I've tried to change keys="preds" with keys=["preds"] but I still get the error.
Can someone help me?

[KumoLiu]

Hi @cugwu, what's your output_key in MeanEnsembled?

[cugwu]

I also tried using SavedImaged in this way:

        SaveImaged(keys=["pred"], output_dir=output_directory, output_postfix="seg", output_ext=".nii.gz",
                   resample=False)

But I got this error:
Traceback (most recent call last):
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 89, in apply_transform
return _apply_transform(transform, data, unpack_items)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 53, in _apply_transform
return transform(parameters)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/io/dictionary.py", line 271, in call
meta_data = d[meta_key] if meta_key is not None else None
KeyError: 'pred_meta_dict'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 89, in apply_transform
return _apply_transform(transform, data, unpack_items)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 53, in apply_transform
return transform(parameters)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/compose.py", line 173, in call
input = apply_transform(transform, input, self.map_items, self.unpack_items, self.log_stats)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 113, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.io.dictionary.SaveImaged object at 0x7f6d202f3eb0>

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "test.py", line 350, in
main()
File "test.py", line 288, in main
ensemble_evaluate(mean_post_transforms, models, torch.device(device), test_loader,
File "test.py", line 132, in ensemble_evaluate
evaluator.run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/evaluator.py", line 147, in run
super().run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/workflow.py", line 290, in run
super().run(data=self.data_loader, max_epochs=self.state.max_epochs)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 704, in run
return self._internal_run()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 783, in _internal_run
self._handle_exception(e)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 466, in _handle_exception
raise e
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 753, in _internal_run
time_taken = self._run_once_on_dataset()
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 854, in _run_once_on_dataset
self._handle_exception(e)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 466, in _handle_exception
raise e
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 840, in _run_once_on_dataset
self.state.output = self._process_function(self, self.state.batch)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/evaluator.py", line 442, in _iteration
engine.fire_event(IterationEvents.MODEL_COMPLETED)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 445, in fire_event
return self._fire_event(event_name)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/ignite/engine/engine.py", line 421, in _fire_event
func(*first, *(event_args + others), **kwargs)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/workflow.py", line 234, in _run_postprocessing
engine.state.batch[i], engine.state.output[i] = engine_apply_transform(b, o, posttrans)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/engines/utils.py", line 247, in engine_apply_transform
transformed_data = apply_transform(transform, data)
File "/home/clusterusers/cugwu/.conda/envs/ai3d/lib/python3.8/site-packages/monai/transforms/transform.py", line 113, in apply_transform
raise RuntimeError(f"applying transform {transform}") from e
RuntimeError: applying transform <monai.transforms.compose.Compose object at 0x7f6d202f3d60>

[KumoLiu]

May I know your MONAI version? Maybe update to MONAI0.9.1 and try again! Also check if your input data's type is MetaTensor.

[cugwu]

My MONAI version is 0.9.0

[KumoLiu]

If 0.9.0, maybe try like this:

SaveImaged(keys="pred", meta_keys="image_meta_dict", output_dir=output_directory, output_postfix="seg", output_ext=".nii.gz", resample=False)

[cugwu]

Thanks, it works!