Replies: 4 comments 2 replies
-
@wanghan-iapcm decision required |
Beta Was this translation helpful? Give feedback.
-
The scale is applied after the calculation of model deviation. let's try to reproduce the bug first. |
Beta Was this translation helpful? Give feedback.
-
I agree with you. |
Beta Was this translation helpful? Give feedback.
-
Submitted #3036. |
Beta Was this translation helpful? Give feedback.
-
Given the following LAMMPS input snippet,
As you can see that the two configuration items of deepmd are used the same configuration except the output file, one might naturally assume that the same input would lead to the same output. But it turns out the result is different
I didn't dive into the code yet but I think it is obviously that the evaluation of
model_devi
takes the scale factor into account. I am not sure if it is a good idea to do it this way asmodel_devi
is supposed to be an objective metric for assessing the fitting degree of the model, and it should not be influenced by the scale factor.PS: This is a test run so the two deepmd items have the same configuraiton, in real use cases they are different.
Beta Was this translation helpful? Give feedback.
All reactions