Skip to content

Serialization changes proposed: remove "mask_write", move eggs/larva to new habitat if old one doesn't exist #113

@stitova-idm

Description

@stitova-idm

Serialization_Mask_Node_Write parameter is currently not useful. Originally there was an idea to filter some of the data so you can reduce your serialzed file, but the only 'filter/mask' that's available is the larval habitat data which takes up very little room. It's also just nice to always have all the data just in case and overwrite it if you don't need it using mask_read. It's also very hard to tell whether or not serialized file is missing data - you can only check by loading the file and seeing what it replaced.

Currently, even if you are loading your larval habitats from config, you have to keep all the same habitat types that you have in serialized file (and you can add more). The work-around if you want to "replace" the habitat is to set the habitat you don't want to 0, so it's still "present" in the config, but isn't used (except for it's still used, the vectors are still trying to lay eggs into it, it just keeps zeroign everything out every time step unless you have drought survival). This work-around also creates a temprorary dip in new vectors since the eggs and larva that already exist die out in the 0-set habitat and the new eggs and larva in the new habitat take around 20 days to become mature vectors. We require old habitats to be present because we serialize egg and larval queues and those queues know which habitat they belong to and try to get back into that habitat and we get an error if they can't find it: Proposed solutions - 1) (my favorite) If their home habitat is not present, put them into the next available habitat (or pick a random available habitat or even split them equally between the available habitats) - this offers egg and larva continuity without the dip, like picking up and moving your into a new house 2) zero out the eggs and larva if the habitat doesnt exist - works the same way as the work-around without the annoying propagation of the habitat that takes up (small, but) computer resources. -- what to do about larvicing intervention when the habitat is no longer there.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions