Skip to content
This repository was archived by the owner on Nov 3, 2023. It is now read-only.
This repository was archived by the owner on Nov 3, 2023. It is now read-only.

Question about dataprocess of SeeKeR #5087

@ahxchzt

Description

@ahxchzt

I'd like to know how you handle data, as you have three different tasks (search query, knowledge, response). During the training process, do you mix these three different types of data together for training, or do you train one task after another once a task is completed? If it's the latter, I'm curious about how you address the issue of catastrophic forgetting.

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