Dropout3d Type
A model which during training, randomly zero out entire channels. Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution.
Constructors
| Constructor | Description |
|
|
![[diffsharp-logo-text.png]](https://diffsharp.github.io/img/diffsharp-logo-text.png){kind=link}
![[copy-md.png]](https://diffsharp.github.io/content/img/copy-md.png){kind=link}
![[copy-md-hover.png]](https://diffsharp.github.io/content/img/copy-md-hover.png){kind=link}
![[copy-xml.png]](https://diffsharp.github.io/content/img/copy-xml.png){kind=link}
![[copy-xml-hover.png]](https://diffsharp.github.io/content/img/copy-xml-hover.png){kind=link}
![[github.png]](https://diffsharp.github.io/content/img/github.png){kind=link}
![[github-hover.png]](https://diffsharp.github.io/content/img/github-hover.png){kind=link}
Copyright 2021, DiffSharp Contributors.