wrap forward passes with torch.no_grad() (#19416)

tests/models/tapas/test_modeling_tapas.py

@@ -570,6 +570,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         table, queries = prepare_tapas_single_inputs_for_inference()
         inputs = tokenizer(table=table, queries=queries, return_tensors="pt")
         inputs = {k: v.to(torch_device) for k, v in inputs.items()}
+        with torch.no_grad():
             outputs = model(**inputs)
         # test the sequence output
         expected_slice = torch.tensor(
@@ -608,6 +609,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         table, queries = prepare_tapas_single_inputs_for_inference()
         inputs = tokenizer(table=table, queries=queries, return_tensors="pt")
         inputs = {k: v.to(torch_device) for k, v in inputs.items()}
+        with torch.no_grad():
             outputs = model(**inputs)
         # test the logits
         logits = outputs.logits
@@ -657,6 +659,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         table, queries = prepare_tapas_single_inputs_for_inference()
         inputs = tokenizer(table=table, queries=queries, return_tensors="pt")
         inputs = {k: v.to(torch_device) for k, v in inputs.items()}
+        with torch.no_grad():
             outputs = model(**inputs)
         # test the logits
         logits = outputs.logits
@@ -705,6 +708,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         inputs = tokenizer(table=table, queries=queries, padding="longest", return_tensors="pt")
         inputs_on_device = {k: v.to(torch_device) for k, v in inputs.items()}
 
+        with torch.no_grad():
             outputs = model(**inputs_on_device)
         # test the logits
         logits = outputs.logits
@@ -774,6 +778,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         float_answer = torch.FloatTensor(float_answer).to(torch_device)
 
         # forward pass to get loss + logits:
+        with torch.no_grad():
             outputs = model(
                 input_ids=input_ids,
                 attention_mask=attention_mask,
@@ -829,6 +834,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         table, queries = prepare_tapas_single_inputs_for_inference()
         inputs = tokenizer(table=table, queries=queries, return_tensors="pt")
         inputs = {k: v.to(torch_device) for k, v in inputs.items()}
+        with torch.no_grad():
             outputs = model(**inputs)
         # test the logits
         logits = outputs.logits
@@ -884,6 +890,7 @@ class TapasModelIntegrationTest(unittest.TestCase):
         table, queries = prepare_tapas_single_inputs_for_inference()
         inputs = tokenizer(table=table, queries=queries, padding="longest", return_tensors="pt")
         inputs = {k: v.to(torch_device) for k, v in inputs.items()}
+        with torch.no_grad():
             outputs = model(**inputs)
 
         # test the classification logits