Lightweight Feedforward Artificial Neural Network (ANN) library tailored for microcontrollers. More...

#include <diwa.h>

Public Member Functions
	Diwa ()
	Default constructor for the Diwa class.

	~Diwa ()
	Destructor for the Diwa class.

DiwaError	initialize (int inputNeurons, int hiddenLayers, int hiddenNeurons, int outputNeurons, bool randomizeWeights=true)
	Initializes the Diwa neural network with specified parameters.

double *	inference (double *inputs)
	Perform inference on the neural network.

void	train (double learningRate, double inputNeurons, double outputNeurons)
	Train the neural network using backpropagation.

DiwaError	loadFromFile (T annFile)
	Load neural network model from file in Arduino environment.

DiwaError	saveToFile (T annFile)
	Save neural network model to file in Arduino environment.

double	calculateAccuracy (double testInput, double testExpectedOutput, int epoch)
	Calculates the accuracy of the neural network on test data.

double	calculateLoss (double testInput, double testExpectedOutput, int epoch)
	Calculates the loss of the neural network on test data.

void	setActivationFunction (diwa_activation activation)
	Sets the activation function for the neural network.

diwa_activation	getActivationFunction () const
	Retrieves the current activation function used by the neural network.

int	recommendedHiddenNeuronCount ()
	Calculates the recommended number of hidden neurons based on the input and output neurons.

int	recommendedHiddenLayerCount (int numSamples, int alpha)
	Calculates the recommended number of hidden layers based on the dataset size and complexity.

int	getInputNeurons () const
	Get the number of input neurons in the neural network.

int	getHiddenNeurons () const
	Get the number of neurons in the hidden layer.

int	getHiddenLayers () const
	Get the number of hidden layers in the neural network.

int	getOutputNeurons () const
	Get the number of output neurons in the neural network.

int	getWeightCount () const
	Get the total number of weights in the neural network.

int	getNeuronCount () const
	Get the total number of neurons in the neural network.

void	getWeights (double *weights)
	Retrieve the weights of the neural network.

void	getOutputs (double *outputs)
	Retrieve the outputs of the neural network.

Detailed Description

Lightweight Feedforward Artificial Neural Network (ANN) library tailored for microcontrollers.

The Diwa library is designed to provide a simple yet effective implementation of a Feedforward Artificial Neural Network (ANN) for resource-constrained microcontroller environments such as ESP8266, ESP32, and similar development boards.

Note: This library is primarily intended for lightweight applications. For more intricate tasks, consider using advanced machine learning libraries on more powerful platforms.

Constructor & Destructor Documentation

◆ Diwa()

Diwa::Diwa ( )

Default constructor for the Diwa class.

This constructor initializes a new instance of the Diwa class. It sets up the neural network with default value 0 on parameters.

◆ ~Diwa()

Diwa::~Diwa ( )

Destructor for the Diwa class.

This destructor releases resources associated with the Diwa object upon its destruction. It ensures proper cleanup to prevent memory leaks.

Member Function Documentation

◆ calculateAccuracy()

double Diwa::calculateAccuracy	(	double *	testInput,
		double *	testExpectedOutput,
		int	epoch
	)

Calculates the accuracy of the neural network on test data.

This function calculates the accuracy of the neural network on a given set of test data. It compares the inferred output with the expected output for each test sample and calculates the percentage of correct inferences.

Parameters

testInput	Pointer to the input values of the test data.
testExpectedOutput	Pointer to the expected output values of the test data.
epoch	Total number of test samples in the test data.

Returns: The accuracy of the neural network on the test data as a percentage.

◆ calculateLoss()

double Diwa::calculateLoss	(	double *	testInput,
		double *	testExpectedOutput,
		int	epoch
	)

Calculates the loss of the neural network on test data.

This function calculates the loss of the neural network on a given set of test data. It computes the percentage of test samples for which the inferred output does not match the expected output.

Parameters

testInput	Pointer to the input values of the test data.
testExpectedOutput	Pointer to the expected output values of the test data.
epoch	Total number of test samples in the test data.

Returns: The loss of the neural network on the test data as a percentage.

◆ getActivationFunction()

diwa_activation Diwa::getActivationFunction ( ) const

Retrieves the current activation function used by the neural network.

This method returns the activation function currently set for the neurons in the neural network. It allows the user to query the current activation function being used for inference and training purposes. The activation function determines the output of a neuron based on its input. Different activation functions can be used depending on the nature of the problem being solved and the characteristics of the dataset. Common activation functions include sigmoid, ReLU, and tanh.

Returns: The activation function currently set for the neural network.

See also: Diwa::setActivationFunction()

◆ getHiddenLayers()

int Diwa::getHiddenLayers ( ) const

Get the number of hidden layers in the neural network.

This function returns the total number of hidden layers in the neural network. It helps to understand the depth of the network architecture.

Returns: The number of hidden layers.

◆ getHiddenNeurons()

int Diwa::getHiddenNeurons ( ) const

Get the number of neurons in the hidden layer.

This function returns the number of neurons in a single hidden layer of the neural network. If there are multiple hidden layers, this value represents the number of neurons per hidden layer.

Returns: The number of neurons in each hidden layer.

◆ getInputNeurons()

int Diwa::getInputNeurons ( ) const

Get the number of input neurons in the neural network.

This function returns the number of input neurons in the neural network. It is useful for understanding the network's input size.

Returns: The number of input neurons.

◆ getNeuronCount()

int Diwa::getNeuronCount ( ) const

Get the total number of neurons in the neural network.

This function calculates and returns the total number of neurons in the neural network, including input neurons, hidden neurons (across all layers), and output neurons.

Returns: The total number of neurons.

◆ getOutputNeurons()

int Diwa::getOutputNeurons ( ) const

Get the number of output neurons in the neural network.

This function returns the number of output neurons in the neural network. This is useful for understanding the size of the output layer and the expected output format.

Returns: The number of output neurons.

◆ getOutputs()

void Diwa::getOutputs ( double * outputs )

Retrieve the outputs of the neural network.

This function copies the current outputs of the neural network into the provided array. The array must be pre-allocated with sufficient space to hold all the outputs.

Parameters

outputs Pointer to an array where the outputs will be copied. The size of the array should be at least getOutputNeurons() elements.

◆ getWeightCount()

int Diwa::getWeightCount ( ) const

Get the total number of weights in the neural network.

This function calculates and returns the total number of weights present in the neural network. This includes weights for connections between input neurons, hidden layers, and output neurons.

Returns: The total number of weights.

◆ getWeights()

void Diwa::getWeights ( double * weights )

Retrieve the weights of the neural network.

This function copies the current weights of the neural network into the provided array. The array must be pre-allocated with sufficient space to hold all the weights.

Parameters

weights Pointer to an array where the weights will be copied. The size of the array should be at least getWeightCount() elements.

◆ inference()

double * Diwa::inference ( double * inputs )

Perform inference on the neural network.

Given an array of input values, this method computes and returns an array of output values through the neural network.

Parameters

inputs Array of input values for the neural network.

Returns: Array of output values after inference.

◆ initialize()

DiwaError Diwa::initialize	(	int	inputNeurons,
		int	hiddenLayers,
		int	hiddenNeurons,
		int	outputNeurons,
		bool	randomizeWeights = `true`
	)

Initializes the Diwa neural network with specified parameters.

This method initializes the Diwa neural network with the given parameters, including the number of input neurons, hidden layers, hidden neurons per layer, and output neurons. Additionally, it allows the option to randomize the weights in the network if desired.

Parameters

inputNeurons	Number of input neurons in the neural network.
hiddenLayers	Number of hidden layers in the neural network.
hiddenNeurons	Number of neurons in each hidden layer.
outputNeurons	Number of output neurons in the neural network.
randomizeWeights	Flag indicating whether to randomize weights in the network (default is true).

Returns: DiwaError indicating the initialization status.

◆ loadFromFile()

DiwaError Diwa::loadFromFile ( T annFile )

Load neural network model from file in Arduino environment.

This method loads a previously saved neural network model from the specified file in an Arduino environment. It reads the model data from the given file and initializes the Diwa object with the loaded model parameters and weights.

Parameters

annFile File object representing the neural network model file.

Returns: DiwaError indicating the loading status.

◆ recommendedHiddenLayerCount()

int Diwa::recommendedHiddenLayerCount	(	int	numSamples,
		int	alpha
	)

Calculates the recommended number of hidden layers based on the dataset size and complexity.

This function computes the recommended number of hidden layers for a neural network based on the size and complexity of the dataset. The recommendation is calculated using a heuristic formula that takes into account the number of samples, input neurons, output neurons, and a scaling factor alpha. The recommended number of hidden layers is determined as the total number of samples divided by (alpha times the sum of input and output neurons).

Parameters

numSamples	The total number of samples in the dataset.
alpha	A scaling factor used to adjust the recommendation based on dataset complexity.

Returns: The recommended number of hidden layers, or -1 if any of the input parameters are non-positive.

◆ recommendedHiddenNeuronCount()

int Diwa::recommendedHiddenNeuronCount ( )

Calculates the recommended number of hidden neurons based on the input and output neurons.

This function computes the recommended number of hidden neurons for a neural network based on the number of input and output neurons. The recommendation is calculated using a heuristic formula that aims to strike a balance between model complexity and generalization ability. The recommended number of hidden neurons is determined as the square root of the product of the input and output neurons.

Returns: The recommended number of hidden neurons, or -1 if the input or output neurons are non-positive.

◆ saveToFile()

DiwaError Diwa::saveToFile ( T annFile )

Save neural network model to file in Arduino environment.

This method saves the current state of the neural network model to the specified file in an Arduino environment. It writes the model parameters and weights to the given file, allowing later retrieval and reuse of the trained model.

Parameters

annFile File object representing the destination file for the model.

Returns: DiwaError indicating the saving status.

◆ setActivationFunction()

void Diwa::setActivationFunction ( diwa_activation activation )

Sets the activation function for the neural network.

This method allows the user to set the activation function used by the neurons in the neural network. The activation function determines the output of a neuron based on its input. Different activation functions can be used depending on the nature of the problem being solved and the characteristics of the dataset. Common activation functions include sigmoid, ReLU, and tanh.

Parameters

activation The activation function to be set for the neural network.

See also: Diwa::getActivationFunction()

◆ train()

void Diwa::train	(	double	learningRate,
		double *	inputNeurons,
		double *	outputNeurons
	)

Train the neural network using backpropagation.

This method facilitates the training of the neural network by adjusting its weights based on the provided input and target output values.

Parameters

learningRate	Learning rate for the training process.
inputNeurons	Array of input values for training.
outputNeurons	Array of target output values for training.

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

src/diwa.h
src/diwa.cpp

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ Diwa()

◆ ~Diwa()

Member Function Documentation

◆ calculateAccuracy()

◆ calculateLoss()

◆ getActivationFunction()

◆ getHiddenLayers()

◆ getHiddenNeurons()

◆ getInputNeurons()

◆ getNeuronCount()

◆ getOutputNeurons()

◆ getOutputs()

◆ getWeightCount()

◆ getWeights()

◆ inference()

◆ initialize()

◆ loadFromFile()

◆ recommendedHiddenLayerCount()

◆ recommendedHiddenNeuronCount()

◆ saveToFile()

◆ setActivationFunction()

◆ train()