model_performance

Internal function, used downstream in cheem_ls.

Given a non-linear model, calculate the local explanation.
We purpose view the data space, explanation
space, and model residuals as ensemble graphic interactive on a shiny
application. After an observation of interest is identified, the normalized
variable importance of the local explanation is used as a 1D projection
basis. The support of the local explanation is then explored by changing
the basis with the use of the radial tour <doi:10.32614/RJ-2020-027>;
<doi:10.1080/10618600.1997.10474754>.

Nicholas Spyrison

cheem

Interactively Explore Local Explanations with the Radial Tour

model_performance function

<dl><dt>y</dt>
<dd>Observed response, required by ranger models.</dd>
<dt>pred</dt>
<dd>A (n, 1) vector of the predicted values for each observation.
Typically obtainable with <code><a href="/link/stats%3A%3Apredict()?package=cheem&version=0.4.2" data-mini-rdoc="cheem::stats::predict()">stats::predict()</a></code>. Exact syntax and arguments
may change across model types.</dd>
<dt>label</dt>
<dd>Optionally provide a character label to store reminder
text for the type of model and local explanation used.
Defaults to "label".</dd></dl>

Arguments

Extract higher level model performance statistics — model_performance

<dl>

<dt>y</dt>
<dd>Observed response, required by ranger models.</dd>


<dt>pred</dt>
<dd>A (n, 1) vector of the predicted values for each observation.
Typically obtainable with <code><a href='https://rdrr.io/r/stats/predict.html'>stats::predict()</a></code>. Exact syntax and arguments
may change across model types.</dd>


<dt>label</dt>
<dd>Optionally provide a character label to store reminder
text for the type of model and local explanation used.
Defaults to "label".</dd>

</dl>

model_performance: Extract higher level model performance statistics

Description

Usage

Value

Arguments