Hi,
When I examined the concentration in ICET “reference_data.db”, a single concentration value was sufficient because it was a binary system. I examined this to make a similar database, but how will I process it in the database and training process for ternary systems, for example (Mg-Cr-O)? I would be glad if you could inform me.
You don’t need to specify the concentration yourself for training, the cluster expansion will figure it out based on number of atoms of each type in the structure.
Or maybe I’m misunderstanding the question?
What I actually want to ask is this: How should ternary structures (For example: (Mg-Cr-O) be defined in the database when creating a database? When I examined the concentration in “reference_data.db”, a single concentration value was sufficient since it was a binary system, but how should it be done for the ternary system?
If you have your ternary structures as ASE Atoms, then these can be written to file by db.write.
You dont need to specify concentration in a database to construct a cluster expansion, you dont even need to use a databases if that makes it easier.
Maybe u can go through the tutorials on icet homepage and point to specific line of code that is giving you problems?
First of all, thank you for your interest. I don’t want to do just cluster expansion. I aim to adapt ICET’s “Compiling a structure container” tutorial (for example, for the Mg-Cr-Fe-O system) for myself. I also want to create a training set for this. I aim to create my own database for this purpose.
As mentioned above there are tutorials on how to do this on the icet website, see e.g. this one Structure enumeration — icet documentation