class SHAInet::Layer

Overview


This is an experimental file for pointer matrix implementation

Included Modules

Defined in:

shainet/basic/exprimental.cr
shainet/basic/layer.cr

Constant Summary

Log = ::Log.for(self)

Constructors

Instance Method Summary

Constructor Detail

def self.new(n_type : String, l_size : Int32, activation_function : ActivationFunction = SHAInet.sigmoid) #

[View source]

Instance Method Detail

def activation_function : Float32 | Float64 | Int32 | Int64 -> {Float64, Float64} #

[View source]
def activations : Matrix(Float64) #

[View source]
def biases : Matrix(Float64) #

[View source]
def biases=(biases : Matrix(Float64)) #

[View source]
def clone #

[View source]
def input_sums : Matrix(Float64) #

[View source]
def input_sums=(input_sums : Matrix(Float64)) #

[View source]
def inspect #
Description copied from class Object

Returns an unambiguous and information-rich string representation of this object, typically intended for developers.

This method should usually not be overridden. It delegates to #inspect(IO) which can be overridden for custom implementations.

Also see #to_s.


[View source]
def l_size : Int32 #

[View source]
def n_type : String #

[View source]
def n_type=(n_type : String) #

[View source]
def neurons : Array(SHAInet::Neuron) #

[View source]
def neurons=(neurons : Array(SHAInet::Neuron)) #

[View source]
def propagate_forward_exp(prev_layer : Layer) #

[View source]
def random_seed #

If you don't want neurons to have a blank memory of builds


[View source]
def sigma_primes : Matrix(Float64) #

[View source]
def size : Int32 #

[View source]
def type_change(new_neuron_type : String) #

If you want to change the type of layer including all neuron types within it


[View source]
def weights : Matrix(Float64) #

[View source]
def weights=(weights : Matrix(Float64)) #

[View source]