diel in OCEAN

[FCMeng]

In the basic jupyter note book, there is a code block like:

ocean_params = OCEANParameters(
    edge="K",
    diel=6.84,
    cards=mycards,
    kpoints=lightshow.common.kpoints.GenericEstimatorKpoints(cutoff=33),
    nbands=lightshow.common.nbands.UnitCellVolumeEstimate(e_range=40)
)

If there are multiple materials provided, diel should not set to a given value, since this will be assigned to all the materials. The easiest way is to get this value from MP. @deyulu I remember you had collaboration work using OCEAN. Please double check if this setting is correct or not.

[matthewcarbone]

@deyulu this looks like a bug which we can easily fix by setting diel=None in the parameters and then, if it is None, using the provided value from the Materials Project. Thoughts?

[deyulu]

@matthewcarbone dielectric constant is a required parameter in ocean. If there is no value retrieved from MP, we need to throw an error message, so that user can fix it. Does it make sense? @jtv3, what do you think?

[matthewcarbone]

@deyulu Yeah all sounds good. So I think the plan should be to

1. If diel=None, set the value from the MP
2. If diel is not None, then use that value for all materials (maybe throw a warning that this is not recommended)

[jtv3]

It shouldn’t be required in ocean 3On Sep 27, 2023, at 08:34, Matthew Carbone ***@***.***> wrote: @deyulu Yeah all sounds good. So I think the plan should be to If diel=None, set the value from the MP If diel is not None, then use that value for all materials (maybe throw a warning that this is not recommended) —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[matthewcarbone]

@jtv3 does that mean you think we shouldn't provide a sensible default from the MP? What do you recommend?

[jtv3]

If one is available from MP that is a good option. But if there isn’t one then lightshow shouldn’t worry. On Sep 27, 2023, at 09:07, Matthew Carbone ***@***.***> wrote: @jtv3 does that mean you think we shouldn't provide a sensible default from the MP? What do you recommend? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[matthewcarbone]

Seems then like this is a request for a new feature, basically. Anyone want to do it? If not I'll get to it, eventually.