Round gaussian_blur outputs before OpenCV reference compare.

OpenCV ground truth in gaussian_blur_opencv_results.pt is uint8 (discrete
gray levels). torchvision returns float blur values, so compare rounded
outputs to match integer reference semantics at atol=1.0.

Example (small image, kernel [3, 5], sigma 0.8, channel 0 at (9, 0)):
- OpenCV uint8 reference: 62
- OpenCV float / torchvision fp32: ~61.05 (already within atol=1 vs 62)
- torchvision fp16 on ROCm raw: ~60.94 -> |62 - 60.94| = 1.0625 fails atol=1.0
- after round: 61 -> |62 - 61| = 1.0 passes

Use round, not a bare .to(torch.uint8) on float outputs: PyTorch truncates
float-to-uint8 (60.9375 -> 60), which gives |62 - 60| = 2.0 and fails.
round(out) is enough here and keeps assert_close aligned with true_out
dtype (.to(tensor) leaves float refs as float).

All test_gaussian_blur cases that hit the reference pass with round + atol=1.0.

Author: dnikolaev-amd

test/test_functional_tensor.py

@@ -1053,7 +1053,8 @@ def test_gaussian_blur(device, image_size, dt, ksize, sigma, fn):
     )
 
     out = fn(tensor, kernel_size=ksize, sigma=sigma)
-    torch.testing.assert_close(out, true_out, rtol=0.0, atol=1.0, msg=f"{ksize}, {sigma}")
+    # OpenCV references are uint8; round float outputs before comparing levels.
+    torch.testing.assert_close(torch.round(out), true_out, rtol=0.0, atol=1.0, msg=f"{ksize}, {sigma}")
 
 
 @pytest.mark.parametrize("device", cpu_and_cuda())