Equation 6 & 7 from the paper suggests that the scores are computed from $\hat{x}_{t_i}$ (not $\hat{x}'_{t_i}$).
However, in the implementation, the update (Eq. 6) is applied to x then passed to update_fn, which will then compute the scores from the updated x ($\hat{x}'_{t_i}$ in the paper), instead of $\hat{x}_{t_i}$.
Equation 6 & 7 from the paper suggests that the scores are computed from $\hat{x}_{t_i}$ (not $\hat{x}'_{t_i}$).
However, in the implementation, the update (Eq. 6) is applied to
xthen passed toupdate_fn, which will then compute the scores from the updatedx($\hat{x}'_{t_i}$ in the paper), instead of $\hat{x}_{t_i}$.https://github.com/yang-song/score_inverse_problems/blob/b56e3836b9d7e6a26d41d366203b8e56a6bb5d0b/cs.py#L181-L196
Is there any reason to prefer one method over the other?