formatting changes from black 22.3.0

Summary:
Applies the black-fbsource codemod with the new build of pyfmt.

paintitblack

Reviewed By: lisroach

Differential Revision: D36324783

fbshipit-source-id: 280c09e88257e5e569ab729691165d8dedd767bc

fvcore/common/param_scheduler.py

@@ -116,7 +116,7 @@ def __init__(
         self._decay = decay
 
     def __call__(self, where: float) -> float:
-        return self._start_value * (self._decay ** where)
+        return self._start_value * (self._decay**where)
 
 
 class LinearParamScheduler(ParamScheduler):

fvcore/nn/smooth_l1_loss.py

@@ -64,7 +64,7 @@ def smooth_l1_loss(
     else:
         n = torch.abs(input - target)
         cond = n < beta
-        loss = torch.where(cond, 0.5 * n ** 2 / beta, n - 0.5 * beta)
+        loss = torch.where(cond, 0.5 * n**2 / beta, n - 0.5 * beta)
 
     if reduction == "mean":
         loss = loss.mean() if loss.numel() > 0 else 0.0 * loss.sum()

tests/test_flop_count.py

@@ -361,8 +361,8 @@ def _test_conv(
                 batch_size
                 * input_dim
                 * output_dim
-                * (kernel_size ** conv_dim)
-                * (spatial_size ** conv_dim)
+                * (kernel_size**conv_dim)
+                * (spatial_size**conv_dim)
                 / group_size
                 / 1e9
             )
@@ -905,7 +905,7 @@ def test_upsample(self):
         nodes = self._count_function(
             F.interpolate, (torch.rand(2, 2, 2, 2), None, 2, "bilinear", False), op_name
         )
-        self.assertEqual(counter(*nodes), 2 ** 4 * 4 * 4)
+        self.assertEqual(counter(*nodes), 2**4 * 4 * 4)
 
     def test_complicated_einsum(self):
         op_name = "aten::einsum"

tests/test_smooth_l1_loss.py

@@ -17,7 +17,7 @@ def test_smooth_l1_loss(self) -> None:
         targets = torch.tensor([1.1, 2, 4.5], dtype=torch.float32)
         beta = 0.5
         loss = smooth_l1_loss(inputs, targets, beta=beta, reduction="none").numpy()
-        self.assertTrue(np.allclose(loss, [0.5 * 0.1 ** 2 / beta, 0, 1.5 - 0.5 * beta]))
+        self.assertTrue(np.allclose(loss, [0.5 * 0.1**2 / beta, 0, 1.5 - 0.5 * beta]))
 
         beta = 0.05
         loss = smooth_l1_loss(inputs, targets, beta=beta, reduction="none").numpy()

tests/test_weight_init.py

@@ -63,9 +63,9 @@ def test_conv_weight_init(self) -> None:
                 if layer is nn.Conv1d:
                     spatial_dim = k_size
                 elif layer is nn.Conv2d:
-                    spatial_dim = k_size ** 2
+                    spatial_dim = k_size**2
                 elif layer is nn.Conv3d:
-                    spatial_dim = k_size ** 3
+                    spatial_dim = k_size**3
 
                 # Calculate fan_in and fan_out.
                 # pyre-fixme[61]: `spatial_dim` may not be initialized here.