Skip to content

Switch Norm 1d for 3D tensors #19

@adrienchaton

Description

@adrienchaton

Hello,
Thank you for providing code implementation for your paper.

I am interested in trying your normalization in my current experiment which works on raw waveform and audio "style". It is thus of prime interest to adaptively modulate different feature normalizations and I hope your proposal would work good to my extent.

However, when I read your
Switchable-Normalization/devkit/ops/switchable_norm.py
the 1d normalization only applies to 2D tensors and the 2d normalization only applies to 4D tensors. Whereas pytorch implementations of BatchNorm1d and InstanceNorm1d applies to both 2D and 3D tensors.

If possible, how should I please apply your SwitchNorm1d to 3D tensors, as for instance the output of conv1d ?

thank you !

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