apply Black 25.11.0 style in fbcode/deeplearning/projects (21/92)

Summary:
Formats the covered files with pyfmt.

paintitblack

Reviewed By: itamaro

Differential Revision: D90476315

fbshipit-source-id: ee94c471788b8e7d067813d8b3e0311214d17f3f

sam3/perflib/connected_components.py

@@ -36,9 +36,9 @@ def connected_components_cpu(input_tensor: torch.Tensor):
     if input_tensor.dim() == 4 and input_tensor.shape[1] == 1:
         input_tensor = input_tensor.squeeze(1)
     else:
-        assert (
-            input_tensor.dim() == 3
-        ), "Input tensor must be (B, H, W) or (B, 1, H, W)."
+        assert input_tensor.dim() == 3, (
+            "Input tensor must be (B, H, W) or (B, 1, H, W)."
+        )
 
     batch_size = input_tensor.shape[0]
     labels_list = []
@@ -67,9 +67,9 @@ def connected_components(input_tensor: torch.Tensor):
     if input_tensor.dim() == 3:
         input_tensor = input_tensor.unsqueeze(1)
 
-    assert (
-        input_tensor.dim() == 4 and input_tensor.shape[1] == 1
-    ), "Input tensor must be (B, H, W) or (B, 1, H, W)."
+    assert input_tensor.dim() == 4 and input_tensor.shape[1] == 1, (
+        "Input tensor must be (B, H, W) or (B, 1, H, W)."
+    )
 
     if input_tensor.is_cuda:
         if HAS_CC_TORCH:

sam3/perflib/nms.py

@@ -6,7 +6,6 @@ import logging
 
 import numpy as np
 import torch
-
 from sam3.perflib.masks_ops import mask_iou
 
 

sam3/perflib/triton/connected_components.py

@@ -407,16 +407,16 @@ def connected_components_triton(input_tensor: torch.Tensor):
             - A BxHxW output tensor with dense labels. Background is 0.
             - A BxHxW tensor with the size of the connected component for each pixel.
     """
-    assert (
-        input_tensor.is_cuda and input_tensor.is_contiguous()
-    ), "Input tensor must be a contiguous CUDA tensor."
+    assert input_tensor.is_cuda and input_tensor.is_contiguous(), (
+        "Input tensor must be a contiguous CUDA tensor."
+    )
     out_shape = input_tensor.shape
     if input_tensor.dim() == 4 and input_tensor.shape[1] == 1:
         input_tensor = input_tensor.squeeze(1)
     else:
-        assert (
-            input_tensor.dim() == 3
-        ), "Input tensor must be (B, H, W) or (B, 1, H, W)."
+        assert input_tensor.dim() == 3, (
+            "Input tensor must be (B, H, W) or (B, 1, H, W)."
+        )
 
     B, H, W = input_tensor.shape
     numel = B * H * W