Skip to content

Residual slayer possible?  #101

@zhaxidele

Description

@zhaxidele

Hi, guys,

Slayer works impressively on datasets that have shallow features, but anybody knows if there is somebody who tried going deeper with slayer using residual block? I tried two CNN and two FC on my dataset, it gets only around 45% accuracy for classification, when I give one more CNN, it gets worse with almost no decrease in loss. With ResNet of 21 layers CNN, my data set got over 90% accuracy. So I want to using residual form of SLAYER, and anybody has any resource of this? papers, repositories, etc. ?

Thanks : )

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