update pages

bayesian-process-models.html

@@ -560,7 +560,7 @@ <h2 data-number="19.1" class="anchored" data-anchor-id="running-mcmc-on-the-fore
 <div class="cell">
 <div class="sourceCode cell-code" id="cb7"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a>accept <span class="sc">/</span> (num_iter <span class="sc">*</span> num_pars)</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] 0.3535556</code></pre>
+<pre><code>[1] 0.3555556</code></pre>
 </div>
 </div>
 <p>Process MCMC chain by removing the first 500 iterations and pivoting to a long format</p>

evaluating-forecasts.html

@@ -483,7 +483,7 @@ <h3 data-number="11.3.2" class="anchored" data-anchor-id="example-of-a-crps-calc
 <div class="sourceCode cell-code" id="cb4"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a><span class="fu">library</span>(scoringRules)</span>
 <span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a><span class="fu">crps_sample</span>(<span class="at">y =</span> y, <span class="at">dat =</span> x)</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] 0.2384334</code></pre>
+<pre><code>[1] 0.237732</code></pre>
 </div>
 </div>
 </section>

index.html

@@ -7,7 +7,7 @@
 <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
 
 <meta name="author" content="R. Quinn Thomas">
-<meta name="dcterms.date" content="2025-02-19">
+<meta name="dcterms.date" content="2025-02-21">
 
 <title>A Practical Guide to Ecosystem Forecasting</title>
 <style>
@@ -405,7 +405,7 @@ <h1 class="title">A Practical Guide to Ecosystem Forecasting</h1>
     <div>
     <div class="quarto-title-meta-heading">Published</div>
     <div class="quarto-title-meta-contents">
-      <p class="date">February 19, 2025</p>
+      <p class="date">February 21, 2025</p>
     </div>
   </div>
 

process-model-forecast-evaluation.html

@@ -436,7 +436,7 @@ <h2 data-number="22.2" class="anchored" data-anchor-id="aggregated-scores"><span
 <pre><code># A tibble: 1 × 1
   mean_crps
       &lt;dbl&gt;
-1     0.872</code></pre>
+1     0.870</code></pre>
 </div>
 </div>
 </section>

process-model-forecast.html

@@ -576,7 +576,7 @@ <h2 data-number="21.4" class="anchored" data-anchor-id="step-4-save-forecast-and
 <pre><code>validating that file matches required standard</code></pre>
 </div>
 <div class="cell-output cell-output-stderr">
-<pre><code>data/terrestrial_daily-2025-02-19-bookcast_forest.csv</code></pre>
+<pre><code>data/terrestrial_daily-2025-02-21-bookcast_forest.csv</code></pre>
 </div>
 <div class="cell-output cell-output-stderr">
 <pre><code>✔ file has model_id column

project1.html

@@ -641,7 +641,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <div class="sourceCode cell-code" id="cb14"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb14-1"><a href="#cb14-1" aria-hidden="true" tabindex="-1"></a>split</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>&lt;Training/Testing/Total&gt;
-&lt;4159/1040/5199&gt;</code></pre>
+&lt;4189/1048/5237&gt;</code></pre>
 </div>
 </div>
 <p>To get the training and testing data we need to apply the <code>training()</code> and <code>testing()</code> functions to the split.</p>
@@ -653,7 +653,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <div class="cell">
 <div class="sourceCode cell-code" id="cb17"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb17-1"><a href="#cb17-1" aria-hidden="true" tabindex="-1"></a>train_data</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># A tibble: 4,159 × 5
+<pre><code># A tibble: 4,189 × 5
    datetime   site_id temperature air_temperature   doy
    &lt;date&gt;     &lt;chr&gt;         &lt;dbl&gt;           &lt;dbl&gt; &lt;dbl&gt;
  1 2017-08-27 BARC           31.5              NA   239
@@ -666,7 +666,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
  8 2017-09-03 BARC           31.1              NA   246
  9 2017-09-06 BARC           30.8              NA   249
 10 2017-09-07 BARC           30.5              NA   250
-# ℹ 4,149 more rows</code></pre>
+# ℹ 4,179 more rows</code></pre>
 </div>
 </div>
 <p><em>Feature engineering using a recipe</em></p>
@@ -796,7 +796,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 
 Coefficients:
     (Intercept)      site_idSUGG  air_temperature              doy  
-       5.142973        -0.785496         0.880690         0.002305  </code></pre>
+       5.108053        -0.725307         0.875235         0.002883  </code></pre>
 </div>
 </div>
 <p><em>Step 5: Predict Test Data</em></p>
@@ -808,7 +808,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <div class="cell">
 <div class="sourceCode cell-code" id="cb41"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb41-1"><a href="#cb41-1" aria-hidden="true" tabindex="-1"></a>predictions</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># A tibble: 1,040 × 1
+<pre><code># A tibble: 1,048 × 1
    .pred
    &lt;dbl&gt;
  1    NA
@@ -821,7 +821,7 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
  8    NA
  9    NA
 10    NA
-# ℹ 1,030 more rows</code></pre>
+# ℹ 1,038 more rows</code></pre>
 </div>
 </div>
 <p>We need to combine the <code>.pred</code> column with the testing data using the <code>bind_cols</code> function</p>
@@ -832,20 +832,20 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <div class="cell">
 <div class="sourceCode cell-code" id="cb44"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb44-1"><a href="#cb44-1" aria-hidden="true" tabindex="-1"></a>pred_test</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># A tibble: 1,040 × 6
+<pre><code># A tibble: 1,048 × 6
    datetime   site_id temperature air_temperature   doy .pred
    &lt;date&gt;     &lt;chr&gt;         &lt;dbl&gt;           &lt;dbl&gt; &lt;dbl&gt; &lt;dbl&gt;
  1 2017-09-04 BARC           31.4              NA   247    NA
  2 2017-09-05 BARC           31.0              NA   248    NA
  3 2017-09-14 BARC           28.5              NA   257    NA
- 4 2017-09-19 BARC           28.8              NA   262    NA
- 5 2017-09-22 BARC           29.0              NA   265    NA
- 6 2017-09-29 BARC           30.1              NA   272    NA
- 7 2017-10-04 BARC           27.6              NA   277    NA
- 8 2017-10-06 BARC           27.6              NA   279    NA
- 9 2017-10-11 BARC           29.6              NA   284    NA
-10 2017-10-14 BARC           28.5              NA   287    NA
-# ℹ 1,030 more rows</code></pre>
+ 4 2017-09-15 BARC           29.3              NA   258    NA
+ 5 2017-09-19 BARC           28.8              NA   262    NA
+ 6 2017-09-22 BARC           29.0              NA   265    NA
+ 7 2017-09-29 BARC           30.1              NA   272    NA
+ 8 2017-10-04 BARC           27.6              NA   277    NA
+ 9 2017-10-06 BARC           27.6              NA   279    NA
+10 2017-10-11 BARC           29.6              NA   284    NA
+# ℹ 1,038 more rows</code></pre>
 </div>
 </div>
 <p><em>Step 6: Evaluate model</em></p>
@@ -863,8 +863,8 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <pre><code># A tibble: 2 × 3
   .metric .estimator .estimate
   &lt;chr&gt;   &lt;chr&gt;          &lt;dbl&gt;
-1 rmse    standard       2.37 
-2 rsq     standard       0.829</code></pre>
+1 rmse    standard       2.42 
+2 rsq     standard       0.819</code></pre>
 </div>
 </div>
 <p><em>Step 7: Deploy model</em></p>
@@ -884,16 +884,16 @@ <h4 data-number="12.2.2.2" class="anchored" data-anchor-id="step-2-pre-process-d
 <pre><code># A tibble: 68 × 5
    datetime   site_id air_temperature temperature   doy
    &lt;date&gt;     &lt;chr&gt;             &lt;dbl&gt; &lt;lgl&gt;       &lt;dbl&gt;
- 1 2025-02-19 BARC               17.1 NA             50
- 2 2025-02-20 BARC               16.5 NA             51
- 3 2025-02-21 BARC               13.4 NA             52
- 4 2025-02-22 BARC               12.9 NA             53
- 5 2025-02-23 BARC               15.0 NA             54
- 6 2025-02-24 BARC               19.0 NA             55
- 7 2025-02-25 BARC               14.1 NA             56
- 8 2025-02-26 BARC               11.9 NA             57
- 9 2025-02-27 BARC               12.3 NA             58
-10 2025-02-28 BARC               14.7 NA             59
+ 1 2025-02-21 BARC               7.26 NA             52
+ 2 2025-02-22 BARC               9.64 NA             53
+ 3 2025-02-23 BARC              11.7  NA             54
+ 4 2025-02-24 BARC              10.3  NA             55
+ 5 2025-02-25 BARC              12.0  NA             56
+ 6 2025-02-26 BARC              15.1  NA             57
+ 7 2025-02-27 BARC              16.0  NA             58
+ 8 2025-02-28 BARC              14.8  NA             59
+ 9 2025-03-01 BARC              11.3  NA             60
+10 2025-03-02 BARC              15.1  NA             61
 # ℹ 58 more rows</code></pre>
 </div>
 </div>
@@ -998,7 +998,7 @@ <h3 data-number="12.3.1" class="anchored" data-anchor-id="step-1-set-up-targets"
 <pre><code># A tibble: 1 × 2
   start_date     h
   &lt;date&gt;     &lt;dbl&gt;
-1 2025-01-30    55</code></pre>
+1 2025-02-18    38</code></pre>
 </div>
 </div>
 <p>The targets have to be converted into a time-series tibble (called a <code>tsibble</code>). To make a <code>tsibble</code> you need to define the grouping variables (<code>key</code>) and the common that defines the time-series (<code>index</code>). Many modeling approaches require the datetime to be continuous (no gaps) but allow na values for values in the gaps. The <code>fill_gaps()</code> function fills the gaps with <code>NA</code>s.</p>