inputs

<p>Model inputs are the predictor and response variables whose relationship is
determined by a model fit.  Input specifications supported by
<span class="pkg">MachineShop</span> are summarized in the table below.</p><table><tr><td>
  <code><a rd-options="" href="/link/formula?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::formula">formula</a></code>         </td>
<td> Traditional model formula </td>
</tr><tr><td>
  <code><a rd-options="" href="/link/matrix?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::matrix">matrix</a></code>          </td>
<td> Design matrix of predictors </td>
</tr><tr><td>
  <code><a rd-options="" href="/link/ModelFrame?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::ModelFrame">ModelFrame</a></code>      </td>
<td> Model frame </td>
</tr><tr><td>
  <code><a rd-options="recipes" href="/link/recipe?package=MachineShop&version=2.2.0&to=recipes" data-mini-rdoc="recipes::recipe">recipe</a></code> </td>
<td> Preprocessing recipe roles and steps </td>
</tr></table><p>Response variable types in the input specifications are defined by the user
with the functions and recipe roles:</p><table><tr><td>
  <strong>Response Functions</strong>
    </td>
<td> <code><a rd-options="" href="/link/BinomialVariate?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::BinomialVariate">BinomialVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/DiscreteVariate?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::DiscreteVariate">DiscreteVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/factor?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::factor">factor</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/matrix?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::matrix">matrix</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/NegBinomialVariate?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::NegBinomialVariate">NegBinomialVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/numeric?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::numeric">numeric</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/ordered?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::ordered">ordered</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/PoissonVariate?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::PoissonVariate">PoissonVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="survival" href="/link/Surv?package=MachineShop&version=2.2.0&to=survival" data-mini-rdoc="survival::Surv">Surv</a></code> </td>
</tr><tr><td>
  <strong>Recipe Roles</strong>
    </td>
<td> <code><a rd-options="" href="/link/role_binom?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::role_binom">role_binom</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options="" href="/link/role_surv?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::role_surv">role_surv</a></code> </td>
</tr></table><p>Inputs may be combined, selected, or tuned with the following meta-input
functions.</p><table><tr><td>
  <code><a rd-options="" href="/link/ModeledInput?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::ModeledInput">ModeledInput</a></code>  </td>
<td> Input with a prespecified model </td>
</tr><tr><td>
  <code><a rd-options="" href="/link/SelectedInput?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::SelectedInput">SelectedInput</a></code> </td>
<td> Input selection from a candidate set </td>
</tr><tr><td>
  <code><a rd-options="" href="/link/TunedInput?package=MachineShop&version=2.2.0" data-mini-rdoc="MachineShop::TunedInput">TunedInput</a></code>    </td>
<td> Input tuning over a parameter grid </td>
</tr></table>

Meta-package for statistical and machine learning with a unified
interface for model fitting, prediction, performance assessment, and
presentation of results.  Approaches for model fitting and prediction of
numerical, categorical, or censored time-to-event outcomes include
traditional regression models, regularization methods, tree-based methods,
support vector machines, neural networks, ensembles, data preprocessing,
filtering, and model tuning and selection.  Performance metrics are provided
for model assessment and can be estimated with independent test sets, split
sampling, cross-validation, or bootstrap resampling.  Resample estimation
can be executed in parallel for faster processing and nested in cases of
model tuning and selection.  Modeling results can be summarized with
descriptive statistics; calibration curves; variable importance; partial
dependence plots; confusion matrices; and ROC, lift, and other performance
curves.

Brian Smith

MachineShop

Machine Learning Models and Tools

inputs function

<p>Model inputs are the predictor and response variables whose relationship is
determined by a model fit.  Input specifications supported by
<span class="pkg">MachineShop</span> are summarized in the table below.</p><table><tr><td>
  <code><a rd-options='' href='formula'>formula</a></code>         </td>
<td> Traditional model formula </td>
</tr><tr><td>
  <code><a rd-options='' href='matrix'>matrix</a></code>          </td>
<td> Design matrix of predictors </td>
</tr><tr><td>
  <code><a rd-options='' href='ModelFrame'>ModelFrame</a></code>      </td>
<td> Model frame </td>
</tr><tr><td>
  <code><a rd-options='recipes' href='recipe'>recipe</a></code> </td>
<td> Preprocessing recipe roles and steps </td>
</tr></table><p>Response variable types in the input specifications are defined by the user
with the functions and recipe roles:</p><table><tr><td>
  <strong>Response Functions</strong>
    </td>
<td> <code><a rd-options='' href='BinomialVariate'>BinomialVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='DiscreteVariate'>DiscreteVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='factor'>factor</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='matrix'>matrix</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='NegBinomialVariate'>NegBinomialVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='numeric'>numeric</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='ordered'>ordered</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='PoissonVariate'>PoissonVariate</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='survival' href='Surv'>Surv</a></code> </td>
</tr><tr><td>
  <strong>Recipe Roles</strong>
    </td>
<td> <code><a rd-options='' href='role_binom'>role_binom</a></code> </td>
</tr><tr><td>
    </td>
<td> <code><a rd-options='' href='role_surv'>role_surv</a></code> </td>
</tr></table><p>Inputs may be combined, selected, or tuned with the following meta-input
functions.</p><table><tr><td>
  <code><a rd-options='' href='ModeledInput'>ModeledInput</a></code>  </td>
<td> Input with a prespecified model </td>
</tr><tr><td>
  <code><a rd-options='' href='SelectedInput'>SelectedInput</a></code> </td>
<td> Input selection from a candidate set </td>
</tr><tr><td>
  <code><a rd-options='' href='TunedInput'>TunedInput</a></code>    </td>
<td> Input tuning over a parameter grid </td>
</tr></table>

Response Functions	`BinomialVariate`
	`DiscreteVariate`
	`factor`
	`matrix`
	`NegBinomialVariate`
	`numeric`
	`ordered`
	`PoissonVariate`
	`Surv`
Recipe Roles	`role_binom`
	`role_surv`

inputs: Model Inputs

Description

Arguments

See Also

`formula`	Traditional model formula
`matrix`	Design matrix of predictors
`ModelFrame`	Model frame
`recipe`	Preprocessing recipe roles and steps

`ModeledInput`	Input with a prespecified model
`SelectedInput`	Input selection from a candidate set
`TunedInput`	Input tuning over a parameter grid