inverts the pixel when its value is greater or equal to the threshold, so one would think that higher augmentation magnitude needs lower threshold. However, the threshold increases linearly with magnitude:
Counterintuitively, it still works as expected with magnitude=_MAX_LEVEL because of integer overflow. Given
t = tf.constant([[[0,0,0]]], dtype=tf.uint8)
t < i evaluates to tf.Tensor([[[False False False]]], shape=(1, 1, 3), dtype=bool) iff not (i % 256). In other words, magnitude=_MAX_LEVEL means int((level/_MAX_LEVEL) * 256) = 256, which is equivalent to 0 in tf.uint8. Given the following tf_gradient that goes from (0, 0, 0) to (255, 255, 255) in alternating directions
Both solarize(tf_gradient, 256) and solarize(tf_gradient, 0) indeed fully invert the image:
But if magnitude is 9, int((9/10) * 256) = 230, and solarize(tf_gradient, 230) "abruptly" only inverts a small portion of the image:
I am not 100% sure about the intention but I do want to raise the alarm. The
solarize()transform herehttps://github.com/google-research/big_vision/blob/01edb81a4716f93a48be43b3a4af14e29cdb3a7f/big_vision/pp/autoaugment.py#L180-L184
inverts the pixel when its value is greater or equal to the threshold, so one would think that higher augmentation magnitude needs lower threshold. However, the threshold increases linearly with magnitude:
https://github.com/google-research/big_vision/blob/01edb81a4716f93a48be43b3a4af14e29cdb3a7f/big_vision/pp/autoaugment.py#L513
Counterintuitively, it still works as expected with
magnitude=_MAX_LEVELbecause of integer overflow. Givent < ievaluates totf.Tensor([[[False False False]]], shape=(1, 1, 3), dtype=bool)iffnot (i % 256). In other words,magnitude=_MAX_LEVELmeansint((level/_MAX_LEVEL) * 256) = 256, which is equivalent to 0 intf.uint8. Given the followingtf_gradientthat goes from(0, 0, 0)to(255, 255, 255)in alternating directionsBoth
solarize(tf_gradient, 256)andsolarize(tf_gradient, 0)indeed fully invert the image:But if magnitude is
9,int((9/10) * 256) = 230, andsolarize(tf_gradient, 230)"abruptly" only inverts a small portion of the image: