Skip to content

Database #30

@kj7rrv

Description

@kj7rrv

Because the models are getting large, we should probably store them in a database. Probably one table for logs (storing text and category) and another for the compiled model? Maybe we should just compile the model every time CS it starts? It doesn't seem to take too long, and doing that would make it easier to load a new model (just restart CedarSentinel). We might have to stop CS when retraining if we use a DB; do you think that will be a problem? I guess since CS doesn't read logs while running, we could have the model editor remove the logs from the DB (or mark them as "being processed" in a new DB column)? @fire219

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions