removed logging of every horizon

model/pytorch/dcrnn_supervisor.py

@@ -109,20 +109,10 @@ class DCRNNSupervisor:
             val_iterator = self._data['{}_loader'.format(dataset)].get_iterator()
             losses = []
 
-            per_timestep_loss = torch.zeros(12)  # hardcoded batch size, horizon
-            num_batches = 0
-
-            for batch_i, (x, y) in enumerate(val_iterator):
+            for _, (x, y) in enumerate(val_iterator):
                 x, y = self._prepare_data(x, y)
 
                 output = self.dcrnn_model(x)
-
-                # (horizon, batch_size, num_sensor * output_dim)
-                for t in range(y.size(0)):
-                    per_timestep_loss[t] += self._compute_loss(y[t], output[t])
-
-                num_batches += 1
-
                 loss = self._compute_loss(y, output)
                 losses.append(loss.item())
 
@@ -130,11 +120,6 @@ class DCRNNSupervisor:
 
             self._writer.add_scalar('{} loss'.format(dataset), mean_loss, batches_seen)
 
-            per_timestep_loss /= num_batches
-
-            for i, val in enumerate(per_timestep_loss):
-                self._logger.info("Dataset:{}, Timestep: {}, MAE:{:.4f}".format(dataset, i, val.item()))
-
             return mean_loss
 
     def _train(self, base_lr,
@@ -148,8 +133,6 @@ class DCRNNSupervisor:
         lr_scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=steps,
                                                             gamma=lr_decay_ratio)
 
-        self.dcrnn_model = self.dcrnn_model.train()
-
         self._logger.info('Start training ...')
 
         # this will fail if model is loaded with a changed batch_size