fix incorrect comment

Author: OckermanSethGVSU

examples/indexBatching/DCRNN/chicken_pox_main.py

@@ -70,7 +70,7 @@ def train(train_dataloader, val_dataloader, edge_index,edge_weight, epochs, seq_
             # Forward pass
             outputs = model(X_batch, edge_index, edge_weight)  # Shape: (batch_size, seq_length, num_nodes, out_channels)
 
-            # Calculate loss (use only the first output channel, assuming it's the target)
+            # Calculate loss
             loss = masked_mae_loss(outputs,y_batch )
 
             # Backward pass

examples/indexBatching/DCRNN/pems_allLA_main.py

@@ -68,7 +68,8 @@ def train(train_dataloader, val_dataloader, mean, std, edges, edge_weights, epoc
             # Forward pass
             outputs = model(X_batch, edge_index, edge_weight)  # Shape: (batch_size, seq_length, num_nodes, out_channels)
 
-            # Calculate loss (use only the first output channel, assuming it's the target)
+            
+            # Calculate loss
             loss = masked_mae_loss((outputs * std) + mean, (y_batch * std) + mean)
 
             # Backward pass

examples/indexBatching/DCRNN/pems_ddp.py

@@ -112,7 +112,7 @@ def train(args=None, epochs=None, batch_size=None, allGPU=False, debug=False, lo
             # Forward pass
             outputs = model(X_batch, edge_index, edge_weight)  # Shape: (batch_size, seq_length, num_nodes, out_channels)
 
-            # Calculate loss (use only the first output channel, assuming it's the target)
+            # Calculate loss
             loss = masked_mae_loss((outputs * std) + mean, (y_batch * std) + mean)
 
             # Backward pass

examples/indexBatching/DCRNN/pems_main.py

@@ -61,14 +61,13 @@ def train(train_dataloader, val_dataloader, mean, std, edges, edge_weights, epoc
             X_batch, y_batch = batch
 
             if allGPU == False:
-                # print("casting")
                 X_batch = X_batch.to(device).float()
                 y_batch = y_batch.to(device).float()
 
             # Forward pass
             outputs = model(X_batch, edge_index, edge_weight)  # Shape: (batch_size, seq_length, num_nodes, out_channels)
 
-            # Calculate loss (use only the first output channel, assuming it's the target)
+            # Calculate loss
             loss = masked_mae_loss((outputs * std) + mean, (y_batch * std) + mean)
 
             # Backward pass