Summary
I haven't necessarily thought through all aspects of this, but I'm wondering if we ought to have a data directory independent to the code i.e.
ml_peg
calcs
analysis
app
data
calcs
app
This would simplify git tracking, which currently has to ignore a huge amount of data, and potentially simplifies data that's shared between the calc and app e.g. phonon reference data, although I'm still not fully sure if this is easily solved.
We'd also need to check this isn't a problem for the mlipx tests, although I think it should be fine.
Example
No response
Alternatives
No response
Additional context
No response
Summary
I haven't necessarily thought through all aspects of this, but I'm wondering if we ought to have a
datadirectory independent to the code i.e.This would simplify git tracking, which currently has to ignore a huge amount of data, and potentially simplifies data that's shared between the calc and app e.g. phonon reference data, although I'm still not fully sure if this is easily solved.
We'd also need to check this isn't a problem for the mlipx tests, although I think it should be fine.
Example
No response
Alternatives
No response
Additional context
No response