Rotate image(s) counter-clockwise by 90 degrees.
tf.image.rot90(
    image, k=1, name=None
)
For example:
a=tf.constant([[[1],[2]],[[3],[4]]])
# rotating `a` counter clockwise by 90 degrees
a_rot=tf.image.rot90(a,k=1) #rotated `a`
print(a_rot) # [[[2],[4]],[[1],[3]]]
Args:
  image: 4-D Tensor of shape [batch, height, width, channels] or 3-D Tensor
    of shape [height, width, channels].
  k: A scalar integer. The number of times the image is rotated by 90 degrees.
  name: A name for this operation (optional).
Returns | 
A rotated tensor of the same type and shape as image.
 | 
Raises | 
ValueError
 | 
if the shape of image not supported.
 |