&Modulus{nat: a}) } else { v.sub(u) C.Add(A, &Modulus{nat: m}) D.Add(B, &Modulus{nat: a}) } } // Exactly one of u and v is now even. if u.IsOdd() == v.IsOdd() { panic("bigmod: internal error: u and v are not in the expected state") } // Halve the even one and adjust the corresponding coefficient. if u.IsOdd() == no { rshift1(u, 0) if A.IsOdd() == yes || B.IsOdd() == yes { rshift1(A, A.add(m)) rshift1(B, B.add(a)) } else { rshift1(A, 0) rshift1(B, 0) } } else { // v.IsOdd() == no rshift1(v, 0) if C.IsOdd() ==...

&Modulus{nat: a}) } else { v.sub(u) C.Add(A, &Modulus{nat: m}) D.Add(B, &Modulus{nat: a}) } } // Exactly one of u and v is now even. if u.IsOdd() == v.IsOdd() { panic("bigmod: internal error: u and v are not in the expected state") } // Halve the even one and adjust the corresponding coefficient. if u.IsOdd() == no { rshift1(u, 0) if A.IsOdd() == yes || B.IsOdd() == yes { rshift1(A, A.add(m)) rshift1(B, B.add(a)) } else { rshift1(A, 0) rshift1(B, 0) } } else { // v.IsOdd() == no rshift1(v, 0) if C.IsOdd() ==...

    the final state instead of the list of all states.
*   `tf.scatter_update` now no longer guarantees that lexicographically largest
    index be used for update when duplicate entries exist.
*   `tf.image.random_crop(image, [height, width])` is now `tf.random_crop(image,
    [height, width, depth])`, and `tf.random_crop` works for any rank (not just