Is there something wrong when i !=j?

[cciao]

In the code ,

dx = self._alpha*np.power(1 - pt, self._gamma - 1) * (self._gamma * (-1 * pt * pro_) * np.log(pt) + pro_ * (1 - pt)) * 1.0

however, in my inference, they should be
dx = self._alpha*np.power(1 - pt, self._gamma - 1) * (self._gamma * (-1 * pt * pro_) * np.log(pt) + pt * (1 - pt)) * 1.0

Is that right?

[cciao]

ok, now I'm totally confused.

If the label is j, the pro_ is equivalent to the p_i while the pt is equivalent to the p_j, right?
If that is right, then the code should be,
dx = self._alpha*np.power(1 - pro_, self._gamma - 1) * (self._gamma * (-1 * pt * pro_) * np.log(pro_) + pro_ * (1 - pro_)) * 1.0
when i !=j.

Help me, please.