A simple neural network with a fixed architecture: 3 fully connected layers, each followed by a square activation layer. More...

Public Member Functions
SimpleNeuralNetPlain ()
Construct a network.

void loadh5 (const H5Parser &h5p, std::vector< std::string > names, std::vector< int > dims, int numFilledSlots)
Loads weights. More...

void predict (const DoubleMatrixArray &input, DoubleMatrixArray &output) const
Run prediction. More...

Public Attributes
SimpleFcPlainLayer fpl1

SimpleFcPlainLayer fpl2

SimpleFcPlainLayer fpl3

SimpleSquareActivationPlainLayer sapl

Detailed Description

A simple neural network with a fixed architecture: 3 fully connected layers, each followed by a square activation layer.

It works on plaintext data, stored as DoubleMatrixArray.

Member Function Documentation

void helayers::SimpleNeuralNetPlain::loadh5	(	const H5Parser &	h5p,
		std::vector< std::string >	names,
		std::vector< int >	dims,
		int	numFilledSlots
	)

Loads weights.

Parameters

[in]	h5p	H5Parser used to read file
[in]	names	names of layers
[in]	dims	size of input vector for each layer, and size of final output
[in]	numFilledSlots	batch size

void helayers::SimpleNeuralNetPlain::predict	(	const DoubleMatrixArray &	input,
		DoubleMatrixArray &	output
	)		const

Run prediction.

Parameters

[in]	input	input data
[out]	output	output prediction

The documentation for this class was generated from the following files: