ValueError: Input size must have a shape of (*, 3, H, W).Got torch.Size([1, 4, 258, 369])

Question

I tried to run the following code on Google collab:

x_bgr: torch.Tensor = K.utils.image_to_tensor(np.array(PIL.Image.open(str(covid[1]))))  
x_bgr = x_bgr[None,...].float() / 255.

x_rgb: torch.Tensor = K.color.bgr_to_rgb(x_bgr)
x_gray = K.color.rgb_to_grayscale(x_rgb)

I got this error:

ValueError                                Traceback (most recent call last)
<ipython-input-16-49608288d875> in <module>()
      3 x_bgr = x_bgr[None,...].float() / 255.
      4 
----> 5 x_rgb: torch.Tensor = K.color.bgr_to_rgb(x_bgr)
      6 x_gray = K.color.rgb_to_grayscale(x_rgb)

/usr/local/lib/python3.7/dist-packages/kornia/color/rgb.py in bgr_to_rgb(image)
     46 
     47     if len(image.shape) < 3 or image.shape[-3] != 3:
---> 48         raise ValueError(f"Input size must have a shape of (*, 3, H, W).Got {image.shape}")
     49 
     50     # flip image channels

ValueError: Input size must have a shape of (*, 3, H, W).Got torch.Size([1, 4, 258, 369])

How should i fix it?