RuntimeError: one_hot is only applicable to index tensor.

[talhaanwarch]

Ground truth shape : torch.Size([4, 1, 512, 512])
Predicted shape: torch.Size([4, 59, 512, 512])
Loss: smp.losses.DiceLoss(mode='multiclass')
Model:

smp.Unet(
               encoder_name='resnet18',   
               encoder_weights=None,  
               in_channels=3,       
               classes=59,  
               activation='softmax2d' ) 

I also tried removing activation='softmax2d' but the error is still there.
Here is the complete traceback

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<timed eval> in <module>

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in fit(self, model, train_dataloader, val_dataloaders, datamodule)
    458         )
    459 
--> 460         self._run(model)
    461 
    462         assert self.state.stopped

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in _run(self, model)
    756 
    757         # dispatch `start_training` or `start_evaluating` or `start_predicting`
--> 758         self.dispatch()
    759 
    760         # plugin will finalized fitting (e.g. ddp_spawn will load trained model)

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in dispatch(self)
    797             self.accelerator.start_predicting(self)
    798         else:
--> 799             self.accelerator.start_training(self)
    800 
    801     def run_stage(self):

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/accelerators/accelerator.py in start_training(self, trainer)
     94 
     95     def start_training(self, trainer: 'pl.Trainer') -> None:
---> 96         self.training_type_plugin.start_training(trainer)
     97 
     98     def start_evaluating(self, trainer: 'pl.Trainer') -> None:

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py in start_training(self, trainer)
    142     def start_training(self, trainer: 'pl.Trainer') -> None:
    143         # double dispatch to initiate the training loop
--> 144         self._results = trainer.run_stage()
    145 
    146     def start_evaluating(self, trainer: 'pl.Trainer') -> None:

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in run_stage(self)
    807         if self.predicting:
    808             return self.run_predict()
--> 809         return self.run_train()
    810 
    811     def _pre_training_routine(self):

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in run_train(self)
    842             self.progress_bar_callback.disable()
    843 
--> 844         self.run_sanity_check(self.lightning_module)
    845 
    846         self.checkpoint_connector.has_trained = False

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in run_sanity_check(self, ref_model)
   1110 
   1111             # run eval step
-> 1112             self.run_evaluation()
   1113 
   1114             self.on_sanity_check_end()

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py in run_evaluation(self, on_epoch)
    965                 # lightning module methods
    966                 with self.profiler.profile("evaluation_step_and_end"):
--> 967                     output = self.evaluation_loop.evaluation_step(batch, batch_idx, dataloader_idx)
    968                     output = self.evaluation_loop.evaluation_step_end(output)
    969 

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/trainer/evaluation_loop.py in evaluation_step(self, batch, batch_idx, dataloader_idx)
    172             model_ref._current_fx_name = "validation_step"
    173             with self.trainer.profiler.profile("validation_step"):
--> 174                 output = self.trainer.accelerator.validation_step(args)
    175 
    176         # capture any logged information

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/accelerators/accelerator.py in validation_step(self, args)
    224 
    225         with self.precision_plugin.val_step_context(), self.training_type_plugin.val_step_context():
--> 226             return self.training_type_plugin.validation_step(*args)
    227 
    228     def test_step(self, args: List[Union[Any, int]]) -> Optional[STEP_OUTPUT]:

/opt/conda/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py in validation_step(self, *args, **kwargs)
    159 
    160     def validation_step(self, *args, **kwargs):
--> 161         return self.lightning_module.validation_step(*args, **kwargs)
    162 
    163     def test_step(self, *args, **kwargs):

<ipython-input-148-9c5092280cd4> in validation_step(self, batch, batch_idx)
     54     outputs=self(image)
     55     print('before',outputs.shape,segment.shape,segment.max())
---> 56     loss=self.criterion(outputs,segment)
     57     print('loss',loss)
     58 #     dice=dice_coef_multilabel(labels,segment)

/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py in _call_impl(self, *input, **kwargs)
    725             result = self._slow_forward(*input, **kwargs)
    726         else:
--> 727             result = self.forward(*input, **kwargs)
    728         for hook in itertools.chain(
    729                 _global_forward_hooks.values(),

/opt/conda/lib/python3.7/site-packages/segmentation_models_pytorch/losses/dice.py in forward(self, y_pred, y_true)
     93                 y_true = y_true.permute(0, 2, 1) * mask.unsqueeze(1)  # H, C, H*W
     94             else:
---> 95                 y_true = F.one_hot(y_true, num_classes)  # N,H*W -> N,H*W, C
     96                 y_true = y_true.permute(0, 2, 1)  # H, C, H*W
     97 

RuntimeError: one_hot is only applicable to index tensor.

[qubvel]

Try to squeeze ground truth tensor

[talhaanwarch]

I squeeze the shape of the ground label to torch.Size([4, 512, 512]) but still error exists

[talhaanwarch]

i need to convert ground truth to .long()

[ljb-1]

I have the same problem. Can you ask me how to solve it?

[ljb-1]

I have the same problem. Can you ask me how to solve it?

[MKaczkow]

For anyone looking for answer: gt input tensor (y_pred) must have type torch.LongTensor (or torch.cuda.LongTensor). If not, this line:

y_true = F.one_hot(y_true, num_classes)  # N,H*W -> N,H*W, C

from smp.losses module fails. This is also described on SO. Though it's unclear for me why LongTensor is specifically required in torch.nn.functional.one_hot (any integer type would do, I guess?), the docs state it is and .long() solves problem.