Skip to content

[FEATURE] - Dependency Injection for Optimization Calculations #65

Description

@coltonbh

Is your feature request related to a problem? Please describe.
For any optimization calculation, one may want to modify the energy/gradient in some arbitrary way.

Describe the solution you'd like
Rather than try to build all cases as parameters or keywords, I could let users pass a modification function that can accept a ProgramOutput (the per-step computed energy/gradient by the "external" program--external to the optimizer), and then return a new ProgramOutput (with modified .energy and .gradient values according to any bespoke algorithm).

results = compute("geometric", ..., mod_func=myfunc)

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions