Skip to content

Latest commit

 

History

History
199 lines (100 loc) · 2.97 KB

api.md

File metadata and controls

199 lines (100 loc) · 2.97 KB

Overview

carbono is a lightweight neural network library for JavaScript that supports both classification and regression tasks with multiple activation functions, optimizers, and loss functions.

Core API

Creating a Network

const nn = new carbono(debug = true);
  • debug: Enable/disable console logging (default: true)

Adding Layers

nn.layer(inputSize, outputSize, activation)

  • inputSize: Number of input neurons

  • outputSize: Number of output neurons

  • activation: Activation function ("tanh", "sigmoid", "relu", "selu", "softmax")

Returns: this

Training

await nn.train(trainSet, options)

Parameters:

  • trainSet: Array of {input: [], output: []} objects

  • options: Training configuration object:

{

  epochs: 10,               // Training iterations (default: 10)

  learningRate: 0.212,      // Learning rate (default: 0.212)

  printEveryEpochs: 1,      // Logging frequency (default: 1)

  earlyStopThreshold: 1e-6, // Early stopping condition

  testSet: null,            // Validation data

  callback: null,           // Epoch callback function  

  optimizer: "sgd",         // "sgd" or "adam"

  lossFunction: "mse"       // "mse" or "cross-entropy"

}

Prediction

nn.predict(input, tags = true)

  • input: Input array

  • tags: Return labeled predictions (default: true)

Model Persistence

// Save model (supports .json and binary .uai format)

await nn.save(name = "model", useBinary = false)

// Load model (supports .json and binary .uai format) 

await nn.load(callback, useBinary = false)
  • useBinary: Use binary .uai format instead of JSON (default: false)

Model Metadata

nn.info({

  name: string,     // Model name

  author: string,   // Author name

  license: string,  // License type

  note: string,     // Description

  date: string      // Creation date

})

Supported Functions

Activation Functions

  • tanh: Hyperbolic tangent

  • sigmoid: Sigmoid function

  • relu: Rectified Linear Unit

  • selu: Scaled Exponential Linear Unit

  • softmax: Softmax (for classification)

Loss Functions

  • mse: Mean Squared Error

  • cross-entropy: Cross Entropy Loss

Optimizers

  • sgd: Stochastic Gradient Descent

  • adam: Adam Optimizer

Example Usage

// Create network

const nn = new carbono();

// Add layers

nn.layer(2, 4, "tanh")

  .layer(4, 1, "sigmoid");

// Training data

const data = [

  {input: [0,0], output: [0]},

  {input: [0,1], output: [1]},

  {input: [1,0], output: [1]}, 

  {input: [1,1], output: [0]}

];

// Train model

await nn.train(data, {

  epochs: 1000,

  learningRate: 0.1

});

// Make prediction

const prediction = nn.predict([0,1]);

// Save model

await nn.save("xor_model");

Additional updates from previous version:

  • Updated default epochs to 10 and printEveryEpochs to 1

  • Clarified return types and default values

  • Corrected default learning rate to 0.212