Skip to content

Commit

Permalink
website
Browse files Browse the repository at this point in the history
  • Loading branch information
@vincentarelbundock
vincentarelbundock committed Oct 6, 2023
1 parent d263bb1 commit 07c1081
Show file tree
Hide file tree
Showing 5 changed files with 69 additions and 60 deletions.
125 changes: 67 additions & 58 deletions docs/articles/machine_learning.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -585,11 +585,11 @@ <h1 class="title">
<pre><code>
Term Contrast Estimate
bill_length_mm +1 0.000
island Dream - Biscoe 0.154
island Torgersen - Biscoe -1.900
sex male - female 2.724
species Chinstrap - Adelie 6.035
species Gentoo - Adelie 5.933
island Dream - Biscoe 0.227
island Torgersen - Biscoe -2.209
sex male - female 2.605
species Chinstrap - Adelie 6.088
species Gentoo - Adelie 6.112

Columns: term, contrast, estimate
Type: numeric </code></pre>
Expand Down Expand Up @@ -645,43 +645,52 @@ <h1 class="title">
<div class="cell">
<div class="sourceCode" id="cb12"><pre class="downlit sourceCode r code-with-copy"><code class="sourceCode R"><span><span class="kw"><a href="https://rdrr.io/r/base/library.html">library</a></span><span class="op">(</span><span class="va"><a href="https://modelsummary.com">modelsummary</a></span><span class="op">)</span></span>
<span></span>
<span><span class="va">recipe</span> <span class="op">&lt;-</span> <span class="va">penguins</span> <span class="op">|&gt;</span></span>
<span><span class="co"># pre-processing</span></span>
<span><span class="va">pre</span> <span class="op">&lt;-</span> <span class="va">penguins</span> <span class="op">|&gt;</span></span>
<span> <span class="fu">recipe</span><span class="op">(</span><span class="va">sex</span> <span class="op">~</span> <span class="va">.</span>, data <span class="op">=</span> <span class="va">_</span><span class="op">)</span> <span class="op">|&gt;</span></span>
<span> <span class="fu">step_ns</span><span class="op">(</span><span class="va">bill_length_mm</span>, deg_free <span class="op">=</span> <span class="fl">4</span><span class="op">)</span> <span class="op">|&gt;</span></span>
<span> <span class="fu">step_dummy</span><span class="op">(</span><span class="fu">all_nominal_predictors</span><span class="op">(</span><span class="op">)</span><span class="op">)</span></span>
<span></span>
<span><span class="co"># modelling strategies</span></span>
<span><span class="va">models</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/list.html">list</a></span><span class="op">(</span></span>
<span> logit <span class="op">=</span> <span class="fu">logistic_reg</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"glm"</span><span class="op">)</span>,</span>
<span> forest <span class="op">=</span> <span class="fu">rand_forest</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"ranger"</span><span class="op">)</span>,</span>
<span> xgb <span class="op">=</span> <span class="fu">boost_tree</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"xgboost"</span><span class="op">)</span></span>
<span> <span class="st">"Logit"</span> <span class="op">=</span> <span class="fu">logistic_reg</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"glm"</span><span class="op">)</span>,</span>
<span> <span class="st">"Random Forest"</span> <span class="op">=</span> <span class="fu">rand_forest</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"ranger"</span><span class="op">)</span>,</span>
<span> <span class="st">"XGBooost"</span> <span class="op">=</span> <span class="fu">boost_tree</span><span class="op">(</span>mode <span class="op">=</span> <span class="st">"classification"</span>, engine <span class="op">=</span> <span class="st">"xgboost"</span><span class="op">)</span></span>
<span><span class="op">)</span></span>
<span></span>
<span><span class="fu"><a href="https://rdrr.io/r/base/lapply.html">lapply</a></span><span class="op">(</span><span class="va">models</span>, \<span class="op">(</span><span class="va">x</span><span class="op">)</span> <span class="op">{</span></span>
<span> <span class="va">recipe</span> <span class="op">|&gt;</span></span>
<span><span class="co"># fit to data</span></span>
<span><span class="va">fits</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/lapply.html">lapply</a></span><span class="op">(</span><span class="va">models</span>, \<span class="op">(</span><span class="va">x</span><span class="op">)</span> <span class="op">{</span></span>
<span> <span class="va">pre</span> <span class="op">|&gt;</span></span>
<span> <span class="fu">workflow</span><span class="op">(</span>spec <span class="op">=</span> <span class="va">x</span><span class="op">)</span> <span class="op">|&gt;</span></span>
<span> <span class="fu">fit</span><span class="op">(</span><span class="va">penguins</span><span class="op">)</span> <span class="op">|&gt;</span></span>
<span> <span class="fu"><a href="https://rdrr.io/pkg/marginaleffects/man/comparisons.html">avg_comparisons</a></span><span class="op">(</span>newdata <span class="op">=</span> <span class="va">penguins</span>, type <span class="op">=</span> <span class="st">"prob"</span><span class="op">)</span> <span class="op">}</span><span class="op">)</span> <span class="op">|&gt;</span></span>
<span> <span class="fu"><a href="https://modelsummary.com/reference/modelsummary.html">modelsummary</a></span><span class="op">(</span></span>
<span> shape <span class="op">=</span> <span class="va">term</span> <span class="op">+</span> <span class="va">contrast</span> <span class="op">+</span> <span class="va">group</span> <span class="op">~</span> <span class="va">model</span>,</span>
<span> coef_omit <span class="op">=</span> <span class="st">"sex"</span>,</span>
<span> coef_rename <span class="op">=</span> <span class="va">coef_rename</span><span class="op">)</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<span> <span class="fu">fit</span><span class="op">(</span><span class="va">penguins</span><span class="op">)</span></span>
<span><span class="op">}</span><span class="op">)</span></span>
<span></span>
<span><span class="co"># marginaleffects</span></span>
<span><span class="va">cmp</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/lapply.html">lapply</a></span><span class="op">(</span><span class="va">fits</span>, <span class="va">avg_comparisons</span>, newdata <span class="op">=</span> <span class="va">penguins</span>, type <span class="op">=</span> <span class="st">"prob"</span><span class="op">)</span></span>
<span></span>
<span><span class="co"># summary table</span></span>
<span><span class="fu"><a href="https://modelsummary.com/reference/modelsummary.html">modelsummary</a></span><span class="op">(</span></span>
<span> <span class="va">cmp</span>,</span>
<span> shape <span class="op">=</span> <span class="va">term</span> <span class="op">+</span> <span class="va">contrast</span> <span class="op">+</span> <span class="va">group</span> <span class="op">~</span> <span class="va">model</span>,</span>
<span> coef_omit <span class="op">=</span> <span class="st">"sex"</span>,</span>
<span> coef_rename <span class="op">=</span> <span class="va">coef_rename</span><span class="op">)</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output-display">
<table class="table table-sm table-striped small" data-quarto-postprocess="true">
<thead><tr class="header">
<th style="text-align: left;" data-quarto-table-cell-role="th"></th>
<th style="text-align: left;" data-quarto-table-cell-role="th"></th>
<th style="text-align: left;" data-quarto-table-cell-role="th"></th>
<th style="text-align: center;" data-quarto-table-cell-role="th">logit</th>
<th style="text-align: center;" data-quarto-table-cell-role="th">forest</th>
<th style="text-align: center;" data-quarto-table-cell-role="th">xgb</th>
<th style="text-align: center;" data-quarto-table-cell-role="th">Logit</th>
<th style="text-align: center;" data-quarto-table-cell-role="th">Random Forest</th>
<th style="text-align: center;" data-quarto-table-cell-role="th">&nbsp;XGBooost</th>
</tr></thead>
<tbody>
<tr class="odd">
<td style="text-align: left;">Bill Length Mm</td>
<td style="text-align: left;">+1</td>
<td style="text-align: left;">female</td>
<td style="text-align: center;">−0.101</td>
<td style="text-align: center;">−0.059</td>
<td style="text-align: center;">−0.062</td>
<td style="text-align: center;">−0.075</td>
</tr>
<tr class="even">
Expand All @@ -697,7 +706,7 @@ <h1 class="title">
<td style="text-align: left;"></td>
<td style="text-align: left;">male</td>
<td style="text-align: center;">0.101</td>
<td style="text-align: center;">0.059</td>
<td style="text-align: center;">0.062</td>
<td style="text-align: center;">0.075</td>
</tr>
<tr class="even">
Expand All @@ -713,7 +722,7 @@ <h1 class="title">
<td style="text-align: left;">Dream - Biscoe</td>
<td style="text-align: left;">female</td>
<td style="text-align: center;">−0.044</td>
<td style="text-align: center;">0.011</td>
<td style="text-align: center;">0.007</td>
<td style="text-align: center;">−0.004</td>
</tr>
<tr class="even">
Expand All @@ -729,7 +738,7 @@ <h1 class="title">
<td style="text-align: left;"></td>
<td style="text-align: left;">male</td>
<td style="text-align: center;">0.044</td>
<td style="text-align: center;">−0.011</td>
<td style="text-align: center;">−0.007</td>
<td style="text-align: center;">0.004</td>
</tr>
<tr class="even">
Expand All @@ -745,7 +754,7 @@ <h1 class="title">
<td style="text-align: left;">Torgersen - Biscoe</td>
<td style="text-align: left;">female</td>
<td style="text-align: center;">0.015</td>
<td style="text-align: center;">−0.056</td>
<td style="text-align: center;">−0.063</td>
<td style="text-align: center;">0.008</td>
</tr>
<tr class="even">
Expand All @@ -761,7 +770,7 @@ <h1 class="title">
<td style="text-align: left;"></td>
<td style="text-align: left;">male</td>
<td style="text-align: center;">−0.015</td>
<td style="text-align: center;">0.056</td>
<td style="text-align: center;">0.063</td>
<td style="text-align: center;">−0.008</td>
</tr>
<tr class="even">
Expand All @@ -777,7 +786,7 @@ <h1 class="title">
<td style="text-align: left;">Chinstrap - Adelie</td>
<td style="text-align: left;">female</td>
<td style="text-align: center;">0.562</td>
<td style="text-align: center;">0.162</td>
<td style="text-align: center;">0.166</td>
<td style="text-align: center;">0.441</td>
</tr>
<tr class="even">
Expand All @@ -793,7 +802,7 @@ <h1 class="title">
<td style="text-align: left;"></td>
<td style="text-align: left;">male</td>
<td style="text-align: center;">−0.562</td>
<td style="text-align: center;">−0.162</td>
<td style="text-align: center;">−0.166</td>
<td style="text-align: center;">−0.441</td>
</tr>
<tr class="even">
Expand All @@ -809,7 +818,7 @@ <h1 class="title">
<td style="text-align: left;">Gentoo - Adelie</td>
<td style="text-align: left;">female</td>
<td style="text-align: center;">0.453</td>
<td style="text-align: center;">0.119</td>
<td style="text-align: center;">0.110</td>
<td style="text-align: center;">0.361</td>
</tr>
<tr class="even">
Expand All @@ -825,7 +834,7 @@ <h1 class="title">
<td style="text-align: left;"></td>
<td style="text-align: left;">male</td>
<td style="text-align: center;">−0.453</td>
<td style="text-align: center;">−0.119</td>
<td style="text-align: center;">−0.110</td>
<td style="text-align: center;">−0.361</td>
</tr>
<tr class="even">
Expand Down Expand Up @@ -890,37 +899,37 @@ <h1 class="title">
<pre><code>
Term Contrast Estimate
count +1 0.000
holiday False - True 12.556
humidity +1 -23.124
month +1 3.892
season spring - fall -31.058
season summer - fall -8.917
season winter - fall 2.506
temp +1 3.409
weather misty - clear -7.332
weather rain - clear -58.989
weekday Fri - Sun 74.481
weekday Mon - Sun 82.925
weekday Sat - Sun 23.244
weekday Thu - Sun 90.207
weekday Tue - Sun 88.818
weekday Wed - Sun 91.096
windspeed +1 0.221
workingday False - True -187.642
year 1 - 0 99.448
holiday False - True 13.101
humidity +1 -23.369
month +1 3.902
season spring - fall -29.587
season summer - fall -8.385
season winter - fall 3.458
temp +1 3.701
weather misty - clear -7.973
weather rain - clear -59.693
weekday Fri - Sun 69.059
weekday Mon - Sun 77.247
weekday Sat - Sun 20.396
weekday Thu - Sun 84.326
weekday Tue - Sun 83.109
weekday Wed - Sun 85.338
windspeed +1 0.136
workingday False - True -190.696
year 1 - 0 97.664

Columns: term, contrast, estimate
Type: response </code></pre>
</div>
</div>
<p>These results are easy to interpret: An increase of 1 degree Celsius in the temperature is associated with an increase of 3.409 bikes rented per hour.</p>
<p>These results are easy to interpret: An increase of 1 degree Celsius in the temperature is associated with an increase of 3.701 bikes rented per hour.</p>
<p>We could obtain the same result manually as follows:</p>
<div class="cell">
<div class="sourceCode" id="cb16"><pre class="downlit sourceCode r code-with-copy"><code class="sourceCode R"><span><span class="va">lo</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/transform.html">transform</a></span><span class="op">(</span><span class="va">bikes</span>, temp <span class="op">=</span> <span class="va">temp</span> <span class="op">-</span> <span class="fl">0.5</span><span class="op">)</span></span>
<span><span class="va">hi</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/transform.html">transform</a></span><span class="op">(</span><span class="va">bikes</span>, temp <span class="op">=</span> <span class="va">temp</span> <span class="op">+</span> <span class="fl">0.5</span><span class="op">)</span></span>
<span><span class="fu"><a href="https://rdrr.io/r/base/mean.html">mean</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/stats/predict.html">predict</a></span><span class="op">(</span><span class="va">forest</span>, newdata <span class="op">=</span> <span class="va">hi</span><span class="op">)</span> <span class="op">-</span> <span class="fu"><a href="https://rdrr.io/r/stats/predict.html">predict</a></span><span class="op">(</span><span class="va">forest</span>, newdata <span class="op">=</span> <span class="va">lo</span><span class="op">)</span><span class="op">)</span></span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output cell-output-stdout">
<pre><code>[1] 3.40943</code></pre>
<pre><code>[1] 3.70069</code></pre>
</div>
</div>
<section id="fmeffects-forward-or-centered-effects" class="level3" data-number="19.2.1"><h3 data-number="19.2.1" class="anchored" data-anchor-id="fmeffects-forward-or-centered-effects">
Expand All @@ -934,7 +943,7 @@ <h1 class="title">
<div class="cell-output cell-output-stdout">
<pre><code>
Term Contrast Estimate
temp custom 2.3
temp custom 2.43

Columns: term, contrast, estimate
Type: response </code></pre>
Expand All @@ -949,7 +958,7 @@ <h1 class="title">
<span> feature <span class="op">=</span> <span class="st">"temp"</span>,</span>
<span> step.size <span class="op">=</span> <span class="fl">1</span><span class="op">)</span><span class="op">$</span><span class="va">ame</span> </span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
<div class="cell-output cell-output-stdout">
<pre><code>[1] 2.30096</code></pre>
<pre><code>[1] 2.434684</code></pre>
</div>
</div>
<p>With <code><a href="https://rdrr.io/pkg/marginaleffects/man/comparisons.html">marginaleffects::avg_comparisons()</a></code>, we can also compute the average effect of a simultaneous change in multiple predictors, using the <code>variables</code> and <code>cross</code> arguments. In this example, we see what happens (on average) to the predicted outcome when the <code>temp</code>, <code>season</code>, and <code>weather</code> predictors all change together:</p>
Expand All @@ -962,12 +971,12 @@ <h1 class="title">
<div class="cell-output cell-output-stdout">
<pre><code>
Estimate C: season C: temp C: weather
-34.75 spring - fall +1 misty - clear
-77.35 spring - fall +1 rain - clear
-12.91 summer - fall +1 misty - clear
-61.90 summer - fall +1 rain - clear
-1.75 winter - fall +1 misty - clear
-55.39 winter - fall +1 rain - clear
-33.256 spring - fall +1 misty - clear
-76.451 spring - fall +1 rain - clear
-12.461 summer - fall +1 misty - clear
-62.248 summer - fall +1 rain - clear
-0.937 winter - fall +1 misty - clear
-54.656 winter - fall +1 rain - clear

Columns: term, contrast_season, contrast_temp, contrast_weather, estimate
Type: response </code></pre>
Expand Down
Binary file modified docs/articles/machine_learning_files/figure-html/unnamed-chunk-15-1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/articles/machine_learning_files/figure-html/unnamed-chunk-18-1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified docs/articles/machine_learning_files/figure-html/unnamed-chunk-19-1.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading

0 comments on commit 07c1081

Please sign in to comment.