npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

@stdlib/ml-incr-sgd-regression

v0.2.2

Published

Online regression via stochastic gradient descent (SGD).

Downloads

20

Readme

Online Regression

NPM version Build Status Coverage Status

Online regression via Stochastic Gradient Descent.

Installation

npm install @stdlib/ml-incr-sgd-regression

Usage

var incrSGDRegression = require( '@stdlib/ml-incr-sgd-regression' );

incrSGDRegression( [options] )

Creates an online linear regression model fitted via stochastic gradient descent. The module performs L2 regularization of the model coefficients, shrinking them towards zero by penalizing the squared euclidean norm of the coefficients.

var randu = require( '@stdlib/random-base-randu' );
var normal = require( '@stdlib/random-base-normal' );
var accumulator = incrSGDRegression();

var x1;
var x2;
var i;
var y;

// Update model as data comes in...
for ( i = 0; i < 100000; i++ ) {
    x1 = randu();
    x2 = randu();
    y = (3.0 * x1) + (-3.0 * x2) + 2.0 + normal( 0.0, 1.0 );
    accumulator( [ x1, x2 ], y );
}

The function accepts the following options:

  • learningRate: string denoting the learning rate to use. Can be constant, pegasos or basic. Default: basic.
  • loss: string denoting the loss function to use. Can be squaredError, epsilonInsensitive or huber. Default: squaredError.
  • epsilon: insensitivity parameter. Default: 0.1.
  • lambda: regularization parameter. Default: 1e-3.
  • eta0: constant learning rate. Default: 0.02.
  • intercept: boolean indicating whether to include an intercept. Default: true.
var accumulator = incrSGDRegression({
    'loss': 'squaredError',
    'lambda': 1e-4
});

The learningRate decides how fast or slow the weights will be updated towards the optimal weights. Let i denote the current iteration of the algorithm (i.e. the number of data points having arrived). The possible learning rates are:

| Option | Definition | | :-------------: | :---------------------: | | basic (default) | 1000.0 / ( i + 1000.0 ) | | constant | eta0 | | pegasos | 1.0 / ( lambda * i ) |

The used loss function is specified via the loss option. The available options are:

  • epsilonInsensitive: Penalty is the absolute value of the error whenever the absolute error exceeds epsilon and zero otherwise.
  • huber: Squared-error loss for observations with error smaller than epsilon in magnitude, linear loss otherwise. Should be used in order to decrease the influence of outliers on the model fit.
  • squaredError: Squared error loss, i.e. the squared difference of the observed and fitted values.

The lambda parameter determines the amount of shrinkage inflicted on the model coefficients:

var createRandom = require( '@stdlib/random-base-randu' ).factory;

var accumulator;
var coefs;
var opts;
var rand;
var x1;
var x2;
var i;
var y;

opts = {
    'seed': 23
};
rand = createRandom( opts );

accumulator = incrSGDRegression({
    'lambda': 1e-5
});

for ( i = 0; i < 100; i++ ) {
    x1 = rand();
    x2 = rand();
    y = (3.0 * x1) + (-3.0 * x2) + 2.0;
    accumulator( [ x1, x2 ], y );
}

coefs = accumulator.coefs;
// returns [ ~3.007, ~-3.002, ~2 ]

rand = createRandom( opts );
accumulator = incrSGDRegression({
    'lambda': 1e-2
});

for ( i = 0; i < 100; i++ ) {
    x1 = rand();
    x2 = rand();
    y = (3.0 * x1) + (-3.0 * x2) + 2.0;
    accumulator( [ x1, x2 ], y );
}

coefs = accumulator.coefs;
// returns [ ~2.893, ~-2.409, ~1.871 ]

Higher values of lambda reduce the variance of the model coefficient estimates at the expense of introducing bias.

By default, the model contains an intercept term. To omit the intercept, set the corresponding option to false:

var accumulator = incrSGDRegression({
    'intercept': false
});
accumulator( [ 1.4, 0.5 ], 2.0 );

var dim = accumulator.coefs.length;
// returns 2

accumulator = incrSGDRegression();
accumulator( [ 1.4, 0.5 ], 2.0 );

dim = accumulator.coefs.length;
// returns 3

If intercept is true, an element equal to one is implicitly added to each x vector. Hence, this module performs regularization of the intercept term.

accumulator( x, y )

Update the model coefficients in light of incoming data. y must be a numeric response value, x a numeric array of predictors. The number of predictors is decided upon first invocation of this method. All subsequent calls must supply x vectors of the same dimensionality.

accumulator( [ 1.0, 0.0 ], 5.0 );

accumulator.predict( x )

Predict the response for a new feature vector x, where x must be a numeric array of predictors. Given feature vector x = [x_0, x_1, ...] and model coefficients c = [c_0, c_1, ...], the prediction is equal to x_0*c_0 + x_1*c_1 + ... + c_intercept.

var yhat = accumulator.predict( [ 0.5, 2.0 ] );
// returns <number>

accumulator.coefs

Getter for the model coefficients / feature weights stored in an array. The coefficients are ordered as [c_0, c_1,..., c_intercept], where c_0 corresponds to the first feature in x and so on.

var coefs = accumulator.coefs;
// returns <Array>

Notes

  • Stochastic gradient descent is sensitive to the scaling of the features. One is best advised to either scale each attribute to [0,1] or [-1,1] or to transform them into z-scores with zero mean and unit variance. One should keep in mind that the same scaling has to be applied to test vectors in order to obtain accurate predictions.
  • Since this module performs regularization of the intercept term, scaling the response variable to an appropriate scale is also highly recommended.

Examples

var randu = require( '@stdlib/random-base-randu' );
var normal = require( '@stdlib/random-base-normal' );
var incrSGDRegression = require( '@stdlib/ml-incr-sgd-regression' );

var accumulator;
var rnorm;
var x1;
var x2;
var y;
var i;

rnorm = normal.factory( 0.0, 1.0 );

// Create model:
accumulator = incrSGDRegression({
    'lambda': 1e-7,
    'loss': 'squaredError',
    'intercept': true
});

// Update model as data comes in...
for ( i = 0; i < 10000; i++ ) {
    x1 = randu();
    x2 = randu();
    y = (3.0 * x1) + (-3.0 * x2) + 2.0 + rnorm();
    accumulator( [ x1, x2 ], y );
}

// Extract model coefficients:
console.log( accumulator.coefs );

// Predict new observations:
console.log( 'y_hat = %d; x1 = %d; x2 = %d', accumulator.predict( [0.9, 0.1] ), 0.9, 0.1 );
console.log( 'y_hat = %d; x1 = %d; x2 = %d', accumulator.predict( [0.1, 0.9] ), 0.1, 0.9 );
console.log( 'y_hat = %d; x1 = %d; x2 = %d', accumulator.predict( [0.9, 0.9] ), 0.9, 0.9 );

See Also


Notice

This package is part of stdlib, a standard library for JavaScript and Node.js, with an emphasis on numerical and scientific computing. The library provides a collection of robust, high performance libraries for mathematics, statistics, streams, utilities, and more.

For more information on the project, filing bug reports and feature requests, and guidance on how to develop stdlib, see the main project repository.

Community

Chat


License

See LICENSE.

Copyright

Copyright © 2016-2024. The Stdlib Authors.