fix: Flake8-BugBear code B-026 for PyTorch (#111362)

Fixes #106571

I have fixed the B-026 error codes for Flake8 tests on the codebase. Please review and tell me anything else to do.
Thanks and excited for this first contribution to PyTorch.

Also I refer this issue which introduced [B-026](https://github.com/PyCQA/flake8-bugbear/issues/286) in `pytest-bugbear` and discuss the error code.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111362
Approved by: https://github.com/Skylion007

.flake8

@@ -14,7 +14,7 @@ ignore =
     # to line this up with executable bit
     EXE001,
     # these ignores are from flake8-bugbear; please fix!
-    B007,B008,B017,B019,B023,B026,B028,B903,B904,B905,B906,B907
+    B007,B008,B017,B019,B023,B028,B903,B904,B905,B906,B907
     # these ignores are from flake8-comprehensions; please fix!
     C407,
     # these ignores are from flake8-logging-format; please fix!

pyproject.toml

@@ -29,7 +29,7 @@ ignore = [
     "B007", "B008", "B017",
     "B018", # Useless expression
     "B019",
-    "B023", "B026",
+    "B023",
     "B028", # No explicit `stacklevel` keyword argument found
     "B904",
     "E402",

test/distributed/fsdp/test_fsdp_checkpoint.py

@@ -100,7 +100,7 @@ class TestFSDPCheckpoint(FSDPTest):
                 l3 = ckpt_wrapper(l3)
 
             fsdp_wrapper = partial(
-                _maybe_wrap_fsdp, wrap_fsdp=wrap_fsdp, *fsdp_args, **fsdp_kwargs
+                _maybe_wrap_fsdp, *fsdp_args, wrap_fsdp=wrap_fsdp, **fsdp_kwargs
             )
             self.ffn = nn.Sequential(
                 fsdp_wrapper(l1),

test/distributed/fsdp/test_fsdp_state_dict.py

@@ -230,8 +230,8 @@ class TestFSDPStateDict(FSDPTest):
                 bn1 = checkpoint_wrapper(bn1)
                 lin2 = checkpoint_wrapper(lin2)
             seq = nn.Sequential(
-                FSDP(lin1, mixed_precision=lin_mp, *fsdp_args, **fsdp_kwargs),
-                FSDP(bn1, mixed_precision=bn_mp, *fsdp_args, **fsdp_kwargs),
+                FSDP(lin1, *fsdp_args, mixed_precision=lin_mp, **fsdp_kwargs),
+                FSDP(bn1, *fsdp_args, mixed_precision=bn_mp, **fsdp_kwargs),
                 lin2,
             )
             if checkpoint_wrap:

test/onnx/test_operators.py

@@ -48,7 +48,7 @@ _onnx_dep = True  # flag to import onnx package.
 
 def export_to_pbtxt(model, inputs, *args, **kwargs):
     return torch.onnx.export_to_pretty_string(
-        model, inputs, google_printer=True, *args, **kwargs
+        model, inputs, *args, google_printer=True, **kwargs
     )
 
 

test/test_ops.py

@@ -986,7 +986,7 @@ class TestCommon(TestCase):
             try:
                 if with_out:
                     out = torch.empty(0, dtype=torch.int32, device=device)
-                    op_to_test(inputs, out=out, *args, **kwargs)
+                    op_to_test(inputs, *args, out=out, **kwargs)
                 else:
                     out = op_to_test(inputs, *args, **kwargs)
                 self.assertFalse(expectFail)

torch/_functorch/aot_autograd.py

@@ -4716,7 +4716,7 @@ def aot_module(mod: nn.Module, *args, **kwargs) -> nn.Module:
     named_buffers = dict(mod.named_buffers(remove_duplicate=False))
     num_params_buffers = len(named_params) + len(named_buffers)
     compiled_f = aot_function(
-        functional_call, num_params_buffers=num_params_buffers, *args, **kwargs
+        functional_call, *args, num_params_buffers=num_params_buffers, **kwargs
     )
 
     class AOTModule(nn.Module):

torch/utils/data/datapipes/iter/combinatorics.py

@@ -40,7 +40,7 @@ class SamplerIterDataPipe(IterDataPipe[T_co]):
         self.sampler_args = () if sampler_args is None else sampler_args
         self.sampler_kwargs = {} if sampler_kwargs is None else sampler_kwargs
         # https://github.com/python/mypy/pull/9629 will solve
-        self.sampler = sampler(data_source=self.datapipe, *self.sampler_args, **self.sampler_kwargs)  # type: ignore[misc]
+        self.sampler = sampler(*self.sampler_args, data_source=self.datapipe, **self.sampler_kwargs)  # type: ignore[misc]
 
     def __iter__(self) -> Iterator[T_co]:
         return iter(self.sampler)