update to support torch 2.0

main.py

@@ -11,7 +11,7 @@ def get_option():
     parser.add_argument('--load_from_raw', type=str2bool, default=True, help='If True, parse and load from benchmark files. If False, load from pt')
     parser.add_argument('--run_all', type=str2bool, default=False, help='If True, run all designs in the given dataset. If False, run the given design_name only.')
     parser.add_argument('--seed', type=int, default=0, help='seed to initialize all the random modules')
-    parser.add_argument('--gpu', type=int, default=1, help='gpu id')
+    parser.add_argument('--gpu', type=int, default=0, help='gpu id')
     parser.add_argument('--num_threads', type=int, default=20, help='threads')
     parser.add_argument('--deterministic', type=str2bool, default=True, help='use deterministic mode')
 

src/calculator.py

@@ -19,7 +19,7 @@ def apply_precond(mov_node_pos: torch.Tensor, ps: ParamScheduler, args):
     mov_node_pos.grad /= ps.precond_weight
     return mov_node_pos.grad
 
-# For nesterov
+
 def calc_obj_and_grad(
     mov_node_pos,
     constraint_fn=None,
@@ -103,18 +103,17 @@ def calc_obj_and_grad(
         grad = apply_precond(mov_node_pos, ps, args)
     return loss, grad
 
-# For Adam
 
 def calc_grad(
     optimizer: torch.optim.Optimizer, mov_node_pos: torch.Tensor, wl_loss, density_loss
 ):
-    optimizer.zero_grad()
+    optimizer.zero_grad(set_to_none=False)
     wl_loss.backward(retain_graph=True)
     wl_grad = mov_node_pos.grad.detach().clone()
-    optimizer.zero_grad()
+    optimizer.zero_grad(set_to_none=False)
     density_loss.backward(retain_graph=True)
     density_grad = mov_node_pos.grad.detach().clone()
-    optimizer.zero_grad()
+    optimizer.zero_grad(set_to_none=False)
     return wl_grad, density_grad