Some question about the results of x_far,x_close,z_far,z_close?

[betterCZ101]

when i use the model you release on the test dataset, i find that the error of x_close is larger than the that of x_far, as well as the z_far and z_close, though the accuracy of F1 score is nearly the same as you release, i want to know if the parameters i have forgotten to set.

[ilnehc]

@betterCZ101 Could you provide more details such as numbers? As shown in the table on the homepage README, these two errors are quite close, or x_c larger than x_far. Thus I think it could be possible.