Annex v0.2.1 Annex.Layer behaviour View Source

The Annex.Layer is the module that defines types, callbacks, and helper for Layers.

By implementing the Layer behaviour a struct/model can be used along side other Layers to compose the layers of a deep neural network.

Link to this section Summary

Link to this section Types

Link to this section Functions

Link to this function

backprop(layer, error, props) View Source

Link to this function

backward_shape(layer) View Source
backward_shape(t()) :: Annex.Shape.t() | nil

Link to this function

data_type(layer) View Source
data_type(atom() | struct()) :: Annex.Data.type()

Link to this function

feedforward(layer, inputs) View Source
feedforward(struct(), any()) :: {struct(), any()}

Link to this function

forward_shape(layer) View Source
forward_shape(t()) :: Annex.Shape.t() | nil

Link to this function

has_data_type?(module) View Source
has_data_type?(module() | struct()) :: boolean()

Link to this function

has_shapes?(module) View Source
has_shapes?(module() | struct()) :: boolean()

Link to this function

is_layer?(item) View Source
is_layer?(any()) :: boolean()

Link to this section Callbacks