Horizontal flip of a row is like swapping the values at pairs of indices starting from the two ends going towards the center. So, if values of a pair of such indices are different, a flip followed by an invert behaves like two inverts in sequence — resulting in zero changes. Likewise, same values behaves like an invert. So, we actually have a single operation per pair of elements in a row.
Time: 
class Solution:
def flip_and_invert(self, row, n):
left, right = 0, n-1
while left < right:
if row[left] == row[right]:
row[left] = 1 - row[left]
row[right] = 1 - row[right]
left += 1
right -= 1
if left == right:
# If a center exists, invert
row[left] = 1 - row[left]
def flipAndInvertImage(self, image: List[List[int]]) -> List[List[int]]:
n = len(image)
for row in image:
self.flip_and_invert(row, n)
return image
unreasonably effective 
Leave a comment