How to add val Loss to tensor board while training

[AHarouni]

I would like to see the val loss in tensorboard so I can compare the train loss and the val loss and see if the mode is over fitting

[AHarouni]

I managed to figure something out. I added this to the trainner

def val_additional_metrics(self, context: Context):
    from ignite.metrics import Loss
    return {"val_loss": Loss(loss_fn=DiceCELoss(to_onehot_y=False, softmax=True)
                            ,output_transform=from_engine(["pred", "label"]))}
 

It should be to_onehot_y = True but it gave me an error so I had to change it to False to make it work