fix bug2

src/sparcscore/ml/plmodels.py

@@ -219,10 +219,12 @@ def training_step(self, batch):
  target = target.unsqueeze(1)
  output = self.network(data) # Forward pass, only one output
 
+ loss = self.configure_loss()
+
  if self.hparams["loss"] == "huber": # Huber loss
- loss = self.configure_loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
+ loss = loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
  else: # MSE
- loss = self.configure_loss(output, target)
+ loss = loss(output, target)
 
  self.log('loss/train', loss, on_step=False, on_epoch=True, prog_bar=True)
  self.log('mse/train', self.mse(output, target), on_epoch=True, prog_bar=True)
@@ -234,11 +236,13 @@ def validation_step(self, batch):
  data, target = batch
  target = target.unsqueeze(1)
  output = self.network(data)
+
+ loss = self.configure_loss()
 
  if self.hparams["loss"] == "huber": # Huber loss
- loss = self.configure_loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
+ loss = loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
  else: # MSE
- loss = self.configure_loss(output, target)
+ loss = loss(output, target)
 
  self.log('loss/val', loss, on_step=False, on_epoch=True, prog_bar=True)
  self.log('mse/val', self.mse(output, target), on_epoch=True, prog_bar=True)
@@ -250,11 +254,13 @@ def test_step(self, batch):
  data, target = batch
  target = target.unsqueeze(1)
  output = self.network(data)
+
+ loss = self.configure_loss()
 
  if self.hparams["loss"] == "huber": # Huber loss
- loss = self.configure_loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
+ loss = loss(output, target, delta=self.hparams["huber_delta"], reduction='mean')
  else: # MSE
- loss = self.configure_loss(output, target)
+ loss = loss(output, target)
 
  self.log('loss/test', loss, on_step=False, on_epoch=True, prog_bar=True)
  self.log('mse/test', self.mse(output, target), on_epoch=True, prog_bar=True)