Diwa
Lightweight implementation of Artificial Neural Network for resource-constrained environments
Loading...
Searching...
No Matches
diwa.h File Reference

This file contains the declaration of the Diwa class, a lightweight Feedforward Artificial Neural Network (ANN) library tailored mainly for microcontrollers. More...

#include <diwa_activations.h>
#include <stdint.h>

Go to the source code of this file.

Classes

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

Enumerations

enum  DiwaError {
  NO_ERROR , INVALID_PARAM_VALUES , MODEL_READ_ERROR , MODEL_SAVE_ERROR ,
  INVALID_MAGIC_NUMBER , STREAM_NOT_OPEN , MALLOC_FAILED
}
 Enumeration representing various error codes that may occur during the operation of the Diwa library. More...
 

Detailed Description

This file contains the declaration of the Diwa class, a lightweight Feedforward Artificial Neural Network (ANN) library tailored mainly for microcontrollers.

Author
Nathanne Isip

The Diwa class allows users to initialize, train, and perform inference with neural networks, as well as save and load trained models from files. It supports both Arduino and non-Arduino environments, enabling seamless integration into various projects.

Note
This library is primarily intended for lightweight applications. For more complex tasks requiring advanced machine learning capabilities, consider using more powerful platforms and libraries.

Enumeration Type Documentation

◆ DiwaError

enum DiwaError

Enumeration representing various error codes that may occur during the operation of the Diwa library.

Enumerator
NO_ERROR 

No error

INVALID_PARAM_VALUES 

Invalid parameter values

MODEL_READ_ERROR 

Error reading model

MODEL_SAVE_ERROR 

Error saving model

INVALID_MAGIC_NUMBER 

Invalid magic number

STREAM_NOT_OPEN 

Stream not open

MALLOC_FAILED 

Memory allocation failed