setup_DBSCAN

Machine learning and visualization package with an 'S7' backend featuring
comprehensive type checking and validation, observable runs, and auditable outputs.
The functional API provides one-call access to supervised and unsupervised learning
through train(), cluster(), and decomp(). All configuration steps are performed using
setup functions and validated. A single call to train() can handle preprocessing,
decomposition, hyperparameter tuning, model stacking, and testing with nested resampling.
Supports 'data.frame', 'data.table', and 'tibble' inputs and parallel execution. Plot
methods for trained and resampled supervised models draw interactive figures through
'rtemis.draw'. Includes native implementations of SuperLearner, RuleFit, and Linear
Additive Trees, and supports conformal prediction. The package first appeared in
E.D. Gennatas (2017)
<https://repository.upenn.edu/entities/publication/d81892ea-3087-4b71-a6f5-739c58626d64>.

Efstathios D. Gennatas

rtemis

Schema-Driven Machine Learning and Visualization

E.D. Gennatas

setup_DBSCAN function

<dl><dt>eps</dt>
<dd>Numeric (0, Inf): Radius of neighborhood.</dd>
<dt>min_points</dt>
<dd>Integer [1, Inf): Minimum number of points in a neighborhood to form a cluster.</dd>
<dt>weights</dt>
<dd>Optional Numeric vector: Weights for data points.</dd>
<dt>border_points</dt>
<dd>Logical: If TRUE, assign border points to clusters.</dd>
<dt>search</dt>
<dd>Character {"kdtree", "linear", "dist"}: Nearest neighbor search strategy.</dd>
<dt>bucket_size</dt>
<dd>Integer [1, Inf): Size of buckets for k-d tree search.</dd>
<dt>split_rule</dt>
<dd>Character {"SUGGEST", "STD", "MIDPT", "FAIR", "SL_MIDPT", "SL_FAIR"}: Rule for splitting the k-d tree.</dd>
<dt>approx</dt>
<dd>Numeric [0, Inf): Approximation factor for the nearest neighbor
search (<code>dbscan::frNN(approx =)</code>); 0 searches exactly. A positive value
guarantees only the neighbors within <code>eps / (1 + approx)</code> and may omit
others within <code>eps</code>, which can produce spurious clusters and noise points
in exchange for speed.</dd>
<dt>features</dt>
<dd>Optional Character vector: Names of at least 2 distinct
feature columns to cluster on. <code>NULL</code> clusters on all numeric features.</dd></dl>

Arguments

Author

Setup DBSCANConfig — setup_DBSCAN

<dl>

<dt>eps</dt>
<dd>Numeric (0, Inf): Radius of neighborhood.</dd>


<dt>min_points</dt>
<dd>Integer [1, Inf): Minimum number of points in a neighborhood to form a cluster.</dd>


<dt>weights</dt>
<dd>Optional Numeric vector: Weights for data points.</dd>


<dt>border_points</dt>
<dd>Logical: If TRUE, assign border points to clusters.</dd>


<dt>search</dt>
<dd>Character {"kdtree", "linear", "dist"}: Nearest neighbor search strategy.</dd>


<dt>bucket_size</dt>
<dd>Integer [1, Inf): Size of buckets for k-d tree search.</dd>


<dt>split_rule</dt>
<dd>Character {"SUGGEST", "STD", "MIDPT", "FAIR", "SL_MIDPT", "SL_FAIR"}: Rule for splitting the k-d tree.</dd>


<dt>approx</dt>
<dd>Numeric [0, Inf): Approximation factor for the nearest neighbor
search (<code>dbscan::frNN(approx =)</code>); 0 searches exactly. A positive value
guarantees only the neighbors within <code>eps / (1 + approx)</code> and may omit
others within <code>eps</code>, which can produce spurious clusters and noise points
in exchange for speed.</dd>


<dt>features</dt>
<dd>Optional Character vector: Names of at least 2 distinct
feature columns to cluster on. <code>NULL</code> clusters on all numeric features.</dd>

</dl>

setup_DBSCAN: Setup DBSCANConfig

Description

Usage

Value

Arguments

Author

Examples