Make test py run test inductor

test.py

@@ -36,14 +36,17 @@ def tearDown(self):
         gc.collect()
 
 
-def _create_example_model_instance(task: ModelTask, device: str):
+def _create_example_model_instance(task: ModelTask, device: str, mode: str):
     skip = False
+    extra_args = ["--accuracy"]
+    if mode == "inductor":
+        extra_args.append("--inductor")
     try:
-        task.make_model_instance(test="eval", device=device, extra_args=["--accuracy"])
+        task.make_model_instance(test="eval", device=device, extra_args=extra_args)
     except NotImplementedError:
         try:
             task.make_model_instance(
-                test="train", device=device, extra_args=["--accuracy"]
+                test="train", device=device, extra_args=extra_args
             )
         except NotImplementedError:
             skip = True
@@ -54,7 +57,7 @@ def _create_example_model_instance(task: ModelTask, device: str):
             )
 
 
-def _load_test(path, device):
+def _load_test(path, device, mode):
     model_name = os.path.basename(path)
 
     def _skip_cuda_memory_check_p(metadata):
@@ -70,7 +73,7 @@ def example_fn(self):
             skip=_skip_cuda_memory_check_p(metadata), assert_equal=self.assertEqual
         ):
             try:
-                _create_example_model_instance(task, device)
+                _create_example_model_instance(task, device, mode)
                 accuracy = task.get_model_attribute("accuracy")
                 assert (
                     accuracy == "pass"
@@ -96,7 +99,7 @@ def train_fn(self):
         ):
             try:
                 task.make_model_instance(
-                    test="train", device=device, batch_size=batch_size
+                    test="train", device=device, batch_size=batch_size, extra_args=["--inductor"] if mode == "inductor" else []
                 )
                 task.invoke()
                 task.check_details_train(device=device, md=metadata)
@@ -119,7 +122,7 @@ def eval_fn(self):
         ):
             try:
                 task.make_model_instance(
-                    test="eval", device=device, batch_size=batch_size
+                    test="eval", device=device, batch_size=batch_size, extra_args=["--inductor"] if mode == "inductor" else []
                 )
                 task.invoke()
                 task.check_details_eval(device=device, md=metadata)
@@ -136,7 +139,7 @@ def check_device_fn(self):
             skip=_skip_cuda_memory_check_p(metadata), assert_equal=self.assertEqual
         ):
             try:
-                task.make_model_instance(test="eval", device=device)
+                task.make_model_instance(test="eval", device=device, extra_args=["--inductor"] if mode == "inductor" else [])
                 task.check_device()
                 task.del_model_instance()
             except NotImplementedError as e:
@@ -152,9 +155,10 @@ def check_device_fn(self):
         # set exclude list based on metadata
         setattr(
             TestBenchmark,
-            f"test_{model_name}_{fn_name}_{device}",
+            f"test_{model_name}_{fn_name}_{device}_{mode}",
             (
                 unittest.skipIf(
+                    # This is expecting that models will never be skipped just based on backend, just on eval or train functions being implemented
                     skip_by_metadata(
                         test=fn_name, device=device, extra_args=[], metadata=metadata
                     ),
@@ -165,6 +169,7 @@ def check_device_fn(self):
 
 
 def _load_tests():
+    modes = ["eager", "inductor"]
     devices = ["cpu"]
     if torch.cuda.is_available():
         devices.append("cuda")
@@ -181,7 +186,8 @@ def _load_tests():
         if "quantized" in path:
             continue
         for device in devices:
-            _load_test(path, device)
+            for mode in modes:
+                _load_test(path, device, mode)
 
 
 _load_tests()