Flux.jl/docs/src/models/nnlib.md

62 lines
935 B
Markdown
Raw Normal View History

2020-02-29 12:45:17 +00:00
# NNlib
2020-03-02 09:52:27 +00:00
2020-02-29 10:14:48 +00:00
Flux re-exports all of the functions exported by the [NNlib](https://github.com/FluxML/NNlib.jl) package.
## Activation Functions
2020-03-02 09:52:27 +00:00
2020-02-29 10:14:48 +00:00
Non-linearities that go between layers of your model. Note that, unless otherwise stated, activation functions operate on scalars. To apply them to an array you can call `σ.(xs)`, `relu.(xs)` and so on.
```@docs
NNlib.celu
2020-02-29 10:14:48 +00:00
NNlib.elu
NNlib.gelu
NNlib.hardsigmoid
NNlib.hardtanh
2020-02-29 10:14:48 +00:00
NNlib.leakyrelu
NNlib.lisht
2020-02-29 10:14:48 +00:00
NNlib.logcosh
NNlib.logsigmoid
NNlib.mish
2020-02-29 10:14:48 +00:00
NNlib.relu
NNlib.relu6
NNlib.rrelu
2020-02-29 10:14:48 +00:00
NNlib.selu
2020-03-03 06:52:20 +00:00
NNlib.sigmoid
2020-02-29 10:14:48 +00:00
NNlib.softplus
NNlib.softshrink
2020-02-29 10:14:48 +00:00
NNlib.softsign
NNlib.swish
NNlib.tanhshrink
NNlib.trelu
2020-02-29 10:14:48 +00:00
```
## Softmax
2020-03-02 09:52:27 +00:00
2020-02-29 10:14:48 +00:00
```@docs
NNlib.softmax
NNlib.logsoftmax
```
## Pooling
2020-03-02 09:52:27 +00:00
2020-02-29 10:14:48 +00:00
```@docs
NNlib.maxpool
NNlib.meanpool
```
## Convolution
2020-03-02 09:52:27 +00:00
2020-02-29 10:14:48 +00:00
```@docs
NNlib.conv
NNlib.depthwiseconv
2020-03-02 09:52:27 +00:00
```
## Batched Operations
```@docs
NNlib.batched_mul
NNlib.batched_mul!
NNlib.batched_adjoint
2020-03-03 06:52:20 +00:00
NNlib.batched_transpose
```