generate_tree_node

It parses a fitted 'R' model object, and returns a formula in
'Tidy Eval' code that calculates the predictions.  It works with
several databases back-ends because it leverages 'dplyr' and 'dbplyr'
for the final 'SQL' translation of the algorithm. It currently
supports lm(), glm(), randomForest(), ranger(), earth(),
xgb.Booster.complete(), cubist(), and ctree() models.

Emil Hvitfeldt

tidypredict

Run Predictions Inside the Database

Edgar Ruiz

Max Kuhn

generate_tree_node function

<dl><dt>node</dt>
<dd>a list with named elements <code>path</code> and <code>prediction</code>. See details
for more.</dd>
<dt>calc_mode</dt>
<dd>character, takes values <code>""</code> and <code>"calc_mode"</code>.
The <code>node</code> list should contain the two lists <code>path</code> and <code>prediction</code>.
The <code>path</code> element has the following structure:
This list can contain 0 or more elemements. The elements but each be of the
following format:<ul>
<li><code>type</code> character, must be <code>"conditional"</code>, <code>"set"</code>, or <code>"all"</code>.</li>
<li><code>op</code> character.
if <code>type == "conditional"</code> must be <code>"more"</code>, <code>"more-equal"</code>, <code>"less"</code>, or
<code>"less-equal"</code>.
if <code>type == "set"</code> must be <code>"in"</code> on <code>not-in</code>.</li>
<li><code>col</code> character.</li>
<li><code>val</code> if <code>type == "conditional"</code> and <code>vals</code> if <code>type == "set"</code>.
Can be character or numeric.</li>
</ul></dd>The <code>prediction</code> list has the following structure:
It can either be a singular value or a list.
If it is a list it will have the following 4 named elements <code>col</code>, <code>val</code>,
<code>op</code>, and <code>is_intercept</code>.<ul>
<li><code>col</code> character, name of column</li>
<li><code>val</code> val, numeric of character</li>
<li><code>op</code> character, known values are <code>"none"</code> and <code>"multiply"</code>. <code>"none"</code> is
used then <code>is_intercept == 1</code>.</li>
<li><code>is_intercept</code>integer, takes values <code>0</code> and <code>1</code>.`</li>
</ul>@keywords internal</dl>

Arguments

Construct a single node of a tree — generate_tree_node

<dl>

<dt>node</dt>
<dd>a list with named elements <code>path</code> and <code>prediction</code>. See details
for more.</dd>


<dt>calc_mode</dt>
<dd>character, takes values <code>""</code> and <code>"calc_mode"</code>.
The <code>node</code> list should contain the two lists <code>path</code> and <code>prediction</code>.
The <code>path</code> element has the following structure:
This list can contain 0 or more elemements. The elements but each be of the
following format:<ul>
<li><code>type</code> character, must be <code>"conditional"</code>, <code>"set"</code>, or <code>"all"</code>.</li>
<li><code>op</code> character.
if <code>type == "conditional"</code> must be <code>"more"</code>, <code>"more-equal"</code>, <code>"less"</code>, or
<code>"less-equal"</code>.
if <code>type == "set"</code> must be <code>"in"</code> on <code>not-in</code>.</li>
<li><code>col</code> character.</li>
<li><code>val</code> if <code>type == "conditional"</code> and <code>vals</code> if <code>type == "set"</code>.
Can be character or numeric.</li>
</ul>

The <code>prediction</code> list has the following structure:
It can either be a singular value or a list.
If it is a list it will have the following 4 named elements <code>col</code>, <code>val</code>,
<code>op</code>, and <code>is_intercept</code>.<ul>
<li><code>col</code> character, name of column</li>
<li><code>val</code> val, numeric of character</li>
<li><code>op</code> character, known values are <code>"none"</code> and <code>"multiply"</code>. <code>"none"</code> is
used then <code>is_intercept == 1</code>.</li>
<li><code>is_intercept</code>integer, takes values <code>0</code> and <code>1</code>.`</li>
</ul>

@keywords internal</dd>

</dl>

generate_tree_node: Construct a single node of a tree

Description

Usage

Arguments