Cleaned up code

main.py

@@ -205,40 +205,41 @@ def run(rank, world_size, args,
     print(model)
 
     if args.use_dist:
-        device = rank #{'cuda:%d' % 0: 'cuda:%d' % rank}
-        ddp_model = model
-        ddp_model.to(rank)
+        # device = rank #{'cuda:%d' % 0: 'cuda:%d' % rank}
+        device = torch.device(rank)
+        model.to(rank)
         #ddp_model = DDP(model, device_ids=[rank])
     else:
         device = get_device(args)
-        ddp_model = model
-        ddp_model.to(device)
+        model.to(device)
 
 
+    # if args.load_model:
+    #     if args.use_dist:  
+    #         map_location = {'cuda:%d' % 0: 'cuda:%d' % rank} 
+    #         ddp_model.load_state_dict(
+    #             torch.load(args.checkpoint_path, map_location=device)
+    #         )
+    #     else: 
+    #         ddp_model.load_state_dict(torch.load(args.checkpoint_path))
+
     if args.load_model:
-        if args.use_dist:  
-            map_location = {'cuda:%d' % 0: 'cuda:%d' % rank} 
-            ddp_model.load_state_dict(
-                torch.load(args.checkpoint_path, map_location=device)
-            )
-        else: 
-            model.load_state_dict(torch.load(args.checkpoint_path))
+        model.load_state_dict(torch.load(args.checkpoint_path))
 
-
     # PDE loss function
     def loss_fn(data, u_scatter, data_2): 
         return get_pde_loss(data, 
                             args.wavelength,
                             args.n_background,
                             u_scatter,
-                            ddp_model,
+                            model,
                             device,
                             args.use_pde_cl,
                             args.two_d,
                             data_2=data_2,
                             ) 
 
-    optimizer = torch.optim.Adam(ddp_model.parameters(), lr=args.learning_rate)
+    optimizer = torch.optim.Adam(model.parameters(), lr=args.learning_rate)
     optimizer.zero_grad()
 
     # Train the PINN
@@ -251,7 +252,7 @@ def loss_fn(data, u_scatter, data_2):
         test_loss_vec.append(test_loss)
         # Automatically synced here, don't need barrier
         if rank == 0:
-            torch.save(ddp_model.state_dict(), args.checkpoint_path) # save model
+            torch.save(model.state_dict(), args.checkpoint_path) # save model
         print("Saved PyTorch Model State to: " + args.checkpoint_path)
     torch.save(test_loss_vec, "test_loss_vec_" + str(rank) + ".pth") # save test loss
     print("Done! Rank: " + str(rank))