synaps v1.0.1

Synaps

TypeScript Neural Network Library

Installation

• In the browser

  <script type="text/javascript" src="https://unpkg.com/synaps"></script>

• In Node.js

  $ npm install synaps

  and

  const synaps = require("synaps").default;

  or in es6 or TypeScript

  import synaps from "synaps";

Usage

• Creating a new instance

  • Neural network with 3 input neurons and 1 output neuron

    let network = new synaps.Network.Type.FeedForward(3, [], 1);

  • Neural network with 4 input neurons, 3 hidden neurons and 2 output neurons

    let network = new synaps.Network.Type.FeedForward(4, [ 3 ], 2);

  • Neural network with 6 input neurons, two hidden layers with 4 and 2 neurons, and 3 output neurons

    let network = new synaps.Network.Type.FeedForward(6, [ 4, 2 ], 3);

• Passing any number of additional options to the network

  // pass an object containing the desired options as the fourth parameter
  let network = new synaps.Network.Type.FeedForward(3, [ 4 ], 1, {
      seed: 501935,
      learningRate: 0.3,
      hiddenLayerActivationFunction: new synaps.Activation.HyperbolicTangent(),
      outputLayerActivationFunction: new synaps.Activation.BinaryStep()
  });

• Available activation functions

  new synaps.Activation.ArcTangent();
  new synaps.Activation.BinaryStep();
  new synaps.Activation.GaussianFunction();
  new synaps.Activation.HyperbolicTangent();
  new synaps.Activation.Identity();
  new synaps.Activation.LogisticFunction();
  new synaps.Activation.RectifiedLinearUnit();
  new synaps.Activation.RectifiedLinearUnit(0.01);
  new synaps.Activation.SinusoidFunction();

• Training the network using supervised batch ("all-at-once") learning

  // the first parameter is the array of inputs and the second parameter is the array of desired outputs
  // the third parameter is the optional number of iterations and the fourth parameter is the optional error threshold
  let error = network.trainBatch(
      [
          [0, 0, 1],
          [0, 1, 1],
          [1, 0, 1],
          [1, 1, 1]
      ],
      [
          [ 0 ],
          [ 1 ],
          [ 1 ],
          [ 0 ]
      ],
      60000,
      0.005
  );

• Training the network using supervised online ("single-pattern") learning

  // the first parameter is the input and the second parameter is the desired output
  let error = network.train([0, 0, 1], [ 0 ]);

• Asking the network to predict some output from a supplied input pattern

  // the single parameter is the input to process
  network.predict([ 0, 0, 1 ])

• Saving the network with all its properties to a JSON string

  let jsonStr = JSON.stringify(network);

• Restoring the network with all its properties from a JSON string

  let network = synaps.Network.Type.FeedForward.fromJson(jsonStr);

Development

• Prerequisites

  $ npm install

• Lint the js files

  $ npm lint

  or to fix some errors automatically

  $ npm lint:fix

• Build the js files

  $ npm build

• Running the Node.js examples

  $ node examples/node.js

Contributing

All contributions are welcome! If you wish to contribute, please create an issue first so that your feature, problem or question can be discussed.

Socials Links

• Discord
• Slack