Skip to content

One question about using rrcf in the stream environment #107

@linguiqiu

Description

@linguiqiu

Hi mdb,
Thanks for your RRCF package which is an outstanding method in detecting anomaly point. However, when I use it in the stream environment some problem apper.
I need save RRCF model to Redis and Redis can only save bytes for model information as we know. Therefore, the serialize and pickle are necessary steps. I use to_dict() function to serializes RCTree to a nested dict and use pickle.dumps() function to change RRCF to bytes. The core challenge, however, lies in the significant time variation between these two steps during streaming processing. The execution duration can vary dramatically—ranging from mere tens of milliseconds to hundreds of milliseconds—depending on the specific computational context. I even can't find out the reason. Do you have any advice about this problem? I will appreciate it if you can give us some useful advice.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions