accessing UNet depth from config.py

Author: n0obcoder

train.py

@@ -87,7 +87,7 @@ def plot_losses(running_train_loss, running_val_loss, train_epoch_loss, val_epoc
 print('len(val_loader)  : {}  @bs={}'.format(len(val_loader), batch_size))
 
 # defining the model
-model = UNet(n_classes = 1, depth = 5, padding = True).to(device) # try decreasing the depth value if there is a memory error
+model = UNet(n_classes = 1, depth = cfg.depth, padding = True).to(device) # try decreasing the depth value if there is a memory error
 
 resume = cfg.resume