rtemis

rtemis-package

Schema-driven machine learning for validated, observable, auditable, and reproducible
machine learning workflows.

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

Schema-Driven Machine Learning and Visualization

E.D. Gennatas

rtemis-package function

<a href="https://docs.rtemis.org/r/ml">https://docs.rtemis.org/r/ml</a>

Online Documentation and Vignettes

Model plots (<code>plot_true_pred()</code>, <code>plot_roc()</code>, <code>plot_varimp()</code>, <code>plot_metric()</code>,
<code>plot_learning()</code>, <code>plot_session()</code>, <code>plot_manhattan()</code>) and <code>present()</code> render
interactive charts through 'rtemis.draw', which also provides the <code>draw_*()</code> functions for
data. <code>draw_linad()</code> draws a LINAD tree.

Visualization

By convention, the last column of the data is the outcome variable, and all other columns are
predictors. Convenience function set_outcome can be used to move a specified column to the
end of the data.
Regression and Classification is performed using <code>train()</code>, which preprocesses, trains,
tunes, and tests models on multiple resamples.
Use available_supervised to get a list of available algorithms

Supervised Learning

For training of binary classification models, the outcome should be provided as a factor,
with the second level of the factor being the 'positive' class.

Classification

Clustering is performed using <code>cluster()</code>.
Use available_clustering to get a list of available algorithms.

Clustering

Decomposition is performed using <code>decomp()</code>.
Use available_decomposition to get a list of available algorithms.

Decomposition

Function documentation includes input type (e.g. "Character", "Integer",
"Float"/"Numeric", etc).
When applicable, value ranges are provided in interval notation. For example, Float: [0, 1)
means floats between 0 and 1 including 0, but excluding 1.
Categorical variables may include set of allowed values using curly braces.
For example, Character: {"future", "mirai", "none"}.

Type Documentation

rtemis internally uses methods for efficient handling of tabular data, with support for
<code>data.frame</code>, <code>data.table</code>, and <code>tibble</code>. If a function is documented as accepting
"tabular data", it accepts any of these data structures. A function documented as accepting
only one of these accepts only that structure.
For example, some optimized <code>data.table</code> operations that perform in-place modifications only
work with <code>data.table</code> objects.

Tabular Data

Errors are signalled as classed conditions that can be handled with
<code><a href="/link/tryCatch()?package=rtemis&version=1.4.3" data-mini-rdoc="rtemis::tryCatch()">tryCatch()</a></code>. See rtemis_conditions for the class hierarchy.

Error Conditions

Maintainer: E.D. Gennatas <a href="/link/gennatas%40gmail.com?package=rtemis&version=1.4.3" data-mini-rdoc="rtemis::gennatas@gmail.com">gennatas@gmail.com</a> (<a href="https://orcid.org/0000-0001-9280-3609">ORCID</a>) [copyright holder]
Authors:<ul>
<li>E.D. Gennatas <a href="/link/gennatas%40gmail.com?package=rtemis&version=1.4.3" data-mini-rdoc="rtemis::gennatas@gmail.com">gennatas@gmail.com</a> (<a href="https://orcid.org/0000-0001-9280-3609">ORCID</a>) [copyright holder]</li>
</ul>

Author

rtemis: Advanced Machine Learning and Visualization — rtemis-package

<a href='https://docs.rtemis.org/r/ml'>https://docs.rtemis.org/r/ml</a>

Model plots (<code>plot_true_pred()</code>, <code>plot_roc()</code>, <code>plot_varimp()</code>, <code>plot_metric()</code>,
<code>plot_learning()</code>, <code>plot_session()</code>, <code>plot_manhattan()</code>) and <code>present()</code> render
interactive charts through 'rtemis.draw', which also provides the <code>draw_*()</code> functions for
data. <code>draw_linad()</code> draws a LINAD tree.

By convention, the last column of the data is the outcome variable, and all other columns are
predictors. Convenience function set_outcome can be used to move a specified column to the
end of the data.
Regression and Classification is performed using <code>train()</code>, which preprocesses, trains,
tunes, and tests models on multiple resamples.
Use available_supervised to get a list of available algorithms

For training of binary classification models, the outcome should be provided as a factor,
with the second level of the factor being the 'positive' class.

Clustering is performed using <code>cluster()</code>.
Use available_clustering to get a list of available algorithms.

Decomposition is performed using <code>decomp()</code>.
Use available_decomposition to get a list of available algorithms.

Function documentation includes input type (e.g. "Character", "Integer",
"Float"/"Numeric", etc).
When applicable, value ranges are provided in interval notation. For example, Float: [0, 1)
means floats between 0 and 1 including 0, but excluding 1.
Categorical variables may include set of allowed values using curly braces.
For example, Character: {"future", "mirai", "none"}.

rtemis internally uses methods for efficient handling of tabular data, with support for
<code>data.frame</code>, <code>data.table</code>, and <code>tibble</code>. If a function is documented as accepting
"tabular data", it accepts any of these data structures. A function documented as accepting
only one of these accepts only that structure.
For example, some optimized <code>data.table</code> operations that perform in-place modifications only
work with <code>data.table</code> objects.

Errors are signalled as classed conditions that can be handled with
<code><a href='https://rdrr.io/r/base/conditions.html'>tryCatch()</a></code>. See rtemis_conditions for the class hierarchy.

Maintainer: E.D. Gennatas <a href='mailto:gennatas@gmail.com'>gennatas@gmail.com</a> (<a href='https://orcid.org/0000-0001-9280-3609'>ORCID</a>) [copyright holder]
Authors:<ul>
<li>E.D. Gennatas <a href='mailto:gennatas@gmail.com'>gennatas@gmail.com</a> (<a href='https://orcid.org/0000-0001-9280-3609'>ORCID</a>) [copyright holder]</li>
</ul>

rtemis-package: rtemis: Advanced Machine Learning and Visualization

Description

Arguments

Online Documentation and Vignettes

Visualization

Supervised Learning

Classification

Clustering

Decomposition

Type Documentation

Tabular Data

Error Conditions

Author

See Also