ScaleNormalize

<p>scales input user-item interaction matrix as per eq (16) from the paper.
Usage of such rescaled matrix with [PureSVD] model will be equal to running PureSVD
on the scaled cosine-based inter-item similarity matrix.</p>

Implements many algorithms for statistical learning on
sparse matrices - matrix factorizations, matrix completion,
elastic net regressions, factorization machines.
Also 'rsparse' enhances 'Matrix' package by providing methods for
multithreaded <sparse, dense> matrix products and native slicing of
the sparse matrices in Compressed Sparse Row (CSR) format.
List of the algorithms for regression problems:
1) Elastic Net regression via Follow The Proximally-Regularized Leader (FTRL)
Stochastic Gradient Descent (SGD), as per McMahan et al(, <doi:10.1145/2487575.2488200>)
2) Factorization Machines via SGD, as per Rendle (2010, <doi:10.1109/ICDM.2010.127>)
List of algorithms for matrix factorization and matrix completion:
1) Weighted Regularized Matrix Factorization (WRMF) via Alternating Least
Squares (ALS) - paper by Hu, Koren, Volinsky (2008, <doi:10.1109/ICDM.2008.22>)
2) Maximum-Margin Matrix Factorization via ALS, paper by Rennie, Srebro
(2005, <doi:10.1145/1102351.1102441>)
3) Fast Truncated Singular Value Decomposition (SVD), Soft-Thresholded SVD,
Soft-Impute matrix completion via ALS - paper by Hastie, Mazumder
et al. (2014, <arXiv:1410.2596>)
4) Linear-Flow matrix factorization, from 'Practical linear models for
large-scale one-class collaborative filtering' by Sedhain, Bui, Kawale et al
(2016, ISBN:978-1-57735-770-4)
5) GlobalVectors (GloVe) matrix factorization via SGD, paper by Pennington,
Socher, Manning (2014, <https://www.aclweb.org/anthology/D14-1162>)
Package is reasonably fast and memory efficient - it allows to work with large
datasets - millions of rows and millions of columns. This is particularly useful
for practitioners working on recommender systems.

Dmitriy Selivanov

rsparse

Statistical Learning on Sparse Matrices

Drew Schmidt

Wei-Chen Chen

ScaleNormalize function

<p></p><dl class="dl-horizontal">
<dt><code>norm</code></dt><dd><p>which norm model should make equal to one</p></dd><p></p><p><dt><code>scale</code></dt><dd><p>how to rescale norm vector</p></dd>
</p></dl><p></p>

Public fields

<h3>Public methods</h3><p></p><p><ul>
<li><p><a href="/link/ScaleNormalize%24new()?package=rsparse&version=0.4.0" data-mini-rdoc="rsparse::ScaleNormalize$new()"><code>ScaleNormalize$new()</code></a></p></li>
<li><p><a href="/link/ScaleNormalize%24fit()?package=rsparse&version=0.4.0" data-mini-rdoc="rsparse::ScaleNormalize$fit()"><code>ScaleNormalize$fit()</code></a></p></li>
<li><p><a href="/link/ScaleNormalize%24transform()?package=rsparse&version=0.4.0" data-mini-rdoc="rsparse::ScaleNormalize$transform()"><code>ScaleNormalize$transform()</code></a></p></li>
<li><p><a href="/link/ScaleNormalize%24fit_transform()?package=rsparse&version=0.4.0" data-mini-rdoc="rsparse::ScaleNormalize$fit_transform()"><code>ScaleNormalize$fit_transform()</code></a></p></li>
<li><p><a href="/link/ScaleNormalize%24clone()?package=rsparse&version=0.4.0" data-mini-rdoc="rsparse::ScaleNormalize$clone()"><code>ScaleNormalize$clone()</code></a></p></li>
</ul>
</p><p></p><h3>Method <code>new()</code></h3><p></p><p>creates model
</p><h3>Usage</h3><p></p><p><pre>ScaleNormalize$new(scale = 0.5, norm = 2, target = c("rows", "columns"))</pre></p><p></p><h3>Arguments</h3><p></p><p><dl class="dl-horizontal">
<dt><code>scale</code></dt><dd><p>numeric, how to rescale norm vector</p></dd></dl></p><p><dt><code>norm</code></dt><dd><p>numeric, which norm model should make equal to one</p></dd></p><p><dt><code>target</code></dt><dd><p>character, defines whether rows or columns should be rescaled</p></dd>
</p><p></p><p></p><h3>Method <code>fit()</code></h3><p></p><p>fits the modes
</p><h3>Usage</h3><p></p><p><pre>ScaleNormalize$fit(x)</pre></p><p></p><h3>Arguments</h3><p></p><p><dl class="dl-horizontal">
<dt><code>x</code></dt><dd><p>input sparse matrix</p></dd>
</dl></p><p></p><p></p><h3>Method <code>transform()</code></h3><p></p><p>transforms new matrix
</p><h3>Usage</h3><p></p><p><pre>ScaleNormalize$transform(x)</pre></p><p></p><h3>Arguments</h3><p></p><p><dl class="dl-horizontal">
<dt><code>x</code></dt><dd><p>input sparse matrix</p></dd>
</dl></p><p></p><p></p><h3>Method <code>fit_transform()</code></h3><p></p><p>fits the model and transforms input
</p><h3>Usage</h3><p></p><p><pre>ScaleNormalize$fit_transform(x)</pre></p><p></p><h3>Arguments</h3><p></p><p><dl class="dl-horizontal">
<dt><code>x</code></dt><dd><p>input sparse matrix</p></dd>
</dl></p><p></p><p></p><h3>Method <code>clone()</code></h3><p></p><p>The objects of this class are cloneable with this method.
</p><h3>Usage</h3><p></p><p><pre>ScaleNormalize$clone(deep = FALSE)</pre></p><p></p><h3>Arguments</h3><p></p><p><dl class="dl-horizontal">
<dt><code>deep</code></dt><dd><p>Whether to make a deep clone.</p></dd>
</dl></p>

ScaleNormalize: Re-scales input matrix proportinally to item popularity

Description

Arguments

Public fields

Methods

Public methods

Method `new()`

Usage

Arguments

Method `fit()`

Usage

Arguments

Method `transform()`

Usage

Arguments

Method `fit_transform()`

Usage

Arguments

Method `clone()`

Usage

Arguments

References

Description

Arguments

Public fields

Methods

Public methods

Method new()

Usage

Arguments

Method fit()

Usage

Arguments

Method transform()

Usage

Arguments

Method fit_transform()

Usage

Arguments

Method clone()

Usage

Arguments

References

Method `new()`

Method `fit()`

Method `transform()`

Method `fit_transform()`

Method `clone()`