createHyperparameterSettings

A user friendly way to create patient level prediction models using
the Observational Medical Outcomes Partnership Common Data Model. Given a cohort
of interest and an outcome of interest, the package can use data in the Common
Data Model to build a large set of features. These features can then be used to
fit a predictive model with a number of machine learning algorithms. This is
further described in Reps (2017) <doi:10.1093/jamia/ocy032>.

Egill Fridgeirsson

PatientLevelPrediction

Develop Clinical Prediction Models Using the Common Data Model

Jenna Reps

Martijn Schuemie

Marc Suchard

Patrick Ryan

Peter Rijnbeek

Observational Health Data Science and Informatics 

createHyperparameterSettings function

<dl><dt>search</dt>
<dd>The type of hyperparameter search to perform. Options are "grid" for grid search, "random" for random search, and "custom" for a user-defined search strategy.</dd>
<dt>tuningMetric</dt>
<dd>The metric to optimize during hyperparameter tuning. Common
choices include <code>aucMetric</code> and <code>auprcMetric</code>.</dd>
<dt>sampleSize</dt>
<dd>Sample size in case of random sampling</dd>
<dt>randomSeed</dt>
<dd>Random seed for random sampling</dd>
<dt>generator</dt>
<dd>Optional custom hyperparameter generator. This can be either
a function with arguments <code>definition</code>, <code>expanded</code>, and <code>settings</code> that
returns a finite list of named parameter lists, or an object with lifecycle
methods. Lifecycle generator objects must provide <code>initialize(definition,
settings)</code> and <code>getNext(history)</code> methods, and may provide an optional
<code>finalize(history)</code> method.</dd></dl>

Arguments

Create Hyperparameter Settings — createHyperparameterSettings

<dl>

<dt>search</dt>
<dd>The type of hyperparameter search to perform. Options are "grid" for grid search, "random" for random search, and "custom" for a user-defined search strategy.</dd>


<dt>tuningMetric</dt>
<dd>The metric to optimize during hyperparameter tuning. Common
choices include <code>aucMetric</code> and <code>auprcMetric</code>.</dd>


<dt>sampleSize</dt>
<dd>Sample size in case of random sampling</dd>


<dt>randomSeed</dt>
<dd>Random seed for random sampling</dd>


<dt>generator</dt>
<dd>Optional custom hyperparameter generator. This can be either
a function with arguments <code>definition</code>, <code>expanded</code>, and <code>settings</code> that
returns a finite list of named parameter lists, or an object with lifecycle
methods. Lifecycle generator objects must provide <code>initialize(definition,
settings)</code> and <code>getNext(history)</code> methods, and may provide an optional
<code>finalize(history)</code> method.</dd>

</dl>

Create Hyperparameter Settings

createHyperparameterSettings: Create Hyperparameter Settings

Description

Usage

Arguments