Merge pull request #1559 from pytorch/frank-wei-patch-1

Update test_flatten_aten and test_reshape_aten due to PT2.0 changed tracer behavior for these ops

Author: peri044

py/torch_tensorrt/fx/test/converters/aten_op/test_flatten_aten.py

@@ -27,7 +27,7 @@ def forward(self, x):
         self.run_test(
             Flatten(start_dim, end_dim),
             inputs,
-            expected_ops={torch.ops.aten._reshape_alias.default},
+            expected_ops={torch.ops.aten.view.default},
             test_implicit_batch_dim=(start_dim != 0),
         )
 

py/torch_tensorrt/fx/test/converters/aten_op/test_reshape_aten.py

@@ -25,7 +25,7 @@ def forward(self, x):
         self.run_test(
             TestModule(target_shape),
             inputs,
-            expected_ops={torch.ops.aten._reshape_alias.default},
+            expected_ops={torch.ops.aten.view.default},
         )
 
     ## TODO: proxytensor tracer does not support output size containing -1. If dim=0 is set to -1 for dynamic batch,