Skip to content

Loss becoming infinite #11

@Mohdwajtech

Description

@Mohdwajtech

Hi,
I have tried training on new set of 80 images following the steps but somewhere in between the training I get this error which I have just attached within this issue as an image. I tried it multiple times but looks like it gets stuck at any random iteration. The only solution is if I keep the dataset small i.e. 10 - 15 pages I am sure not to get this error. Is there a possible workaround to this issue
training_error
?

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