From 07c4b4e20082ce1b6c6b898ec30f5595f481a47a Mon Sep 17 00:00:00 2001
From: GitHub Actions <actions@github.com>
Date: Mon, 22 Jul 2024 12:26:38 +0000
Subject: [PATCH] site deploy

Auto-generated via {sandpaper}
Source  : 654fe7a09bd634621452ce93e7ee21a661671ab9
Branch  : md-outputs
Author  : GitHub Actions <actions@github.com>
Time    : 2024-07-22 12:26:21 +0000
Message : markdown source builds

Auto-generated via {sandpaper}
Source  : 219589148319f3a83ad5ae81a2832bcc9253cb77
Branch  : main
Author  : Carsten Schnober <c.schnober@esciencecenter.nl>
Time    : 2024-07-22 12:25:30 +0000
Message : Merge pull request #495 from carpentries-incubator/fix-header-structure-episode-4

Fix header structure in episode 4
---
 2-keras.html                           | 10 +--
 3-monitor-the-model.html               | 16 ++---
 4-advanced-layer-types.html            | 43 ++++++-------
 5-transfer-learning.html               |  4 +-
 aio.html                               | 73 +++++++++++-----------
 index.html                             | 10 +--
 instructor/2-keras.html                | 16 ++---
 instructor/3-monitor-the-model.html    | 22 +++----
 instructor/4-advanced-layer-types.html | 42 ++++++-------
 instructor/5-transfer-learning.html    |  4 +-
 instructor/aio.html                    | 84 +++++++++++++-------------
 instructor/index.html                  | 14 ++---
 md5sum.txt                             |  2 +-
 pkgdown.yml                            |  2 +-
 14 files changed, 174 insertions(+), 168 deletions(-)

diff --git a/2-keras.html b/2-keras.html
index df059e4c..d41866fb 100644
--- a/2-keras.html
+++ b/2-keras.html
@@ -583,7 +583,7 @@ <h3 class="callout-title">Pairplot<a class="anchor" aria-label="anchor" href="#p
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ul><li>The plots show that the green class, Gentoo is somewhat more easily
 distinguishable from the other two.</li>
@@ -710,7 +710,7 @@ <h3 class="callout-title">One-hot encoding<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <p>C: 3, one for each output variable class</p>
 </div>
@@ -921,7 +921,7 @@ <h4 id="optional-keras-sequential-vs-functional-api">(optional) Keras Sequential
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>Have a look at the output of <code>model.summary()</code>:</p>
 <div class="codewrapper sourceCode" id="cb18">
@@ -1098,7 +1098,7 @@ <h3 class="callout-title">The Training Curve<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>The training loss decreases quickly. It drops in a smooth line with
 little jitter. This is ideal for a training curve.</li>
@@ -1296,7 +1296,7 @@ <h3 class="callout-title">Confusion Matrix<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The confusion matrix shows that the predictions for Adelie and Gentoo
 are decent, but could be improved. However, Chinstrap is not predicted
diff --git a/3-monitor-the-model.html b/3-monitor-the-model.html
index dc358d03..1ceb4692 100644
--- a/3-monitor-the-model.html
+++ b/3-monitor-the-model.html
@@ -620,7 +620,7 @@ <h3 class="callout-title">Exercise: Architecture of the network<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>The shape of the input layer has to correspond to the number of
 features in our data: 89</li>
@@ -781,7 +781,7 @@ <h3 id="when-the-batch-size-is-increased">3. When the batch size is increased:<a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li><p>Correct answer: B. To find the weights that minimize the loss
 function. The loss function quantifies the total error of the network,
@@ -1008,7 +1008,7 @@ <h3 class="callout-title">Exercise: Reflecting on our results<a class="anchor" a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>While the performance on the train set seems reasonable, the
 performance on the test set is much worse. This is a common problem
@@ -1130,7 +1130,7 @@ <h3 class="callout-title">Exercise: Baseline<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>This really depends on your definition of hard! The baseline gives a
 more accurate prediction than just randomly predicting a number, so the
@@ -1227,7 +1227,7 @@ <h3 class="callout-title">Exercise: plot the training progress.<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The difference in the two curves shows that something is not
 completely right here. The error for the model predictions on the
@@ -1275,7 +1275,7 @@ <h3 class="callout-title">Exercise: Try to reduce the degree of overfitting by l
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution6" aria-expanded="false" aria-controls="collapseSolution6">
   <h4 class="accordion-header" id="headingSolution6"> Show me the solution </h4>
 </button>
-<div id="collapseSolution6" class="accordion-collapse collapse" aria-labelledby="headingSolution6" data-bs-parent="#accordionSolution6">
+<div id="collapseSolution6" class="accordion-collapse collapse" data-bs-parent="#accordionSolution6" aria-labelledby="headingSolution6">
 <div class="accordion-body">
 <p>Let’s first adapt our <code>create_nn</code> function so that we can
 tweak the number of nodes in the 2 layers by passing arguments to the
@@ -1564,7 +1564,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution7" aria-expanded="false" aria-controls="collapseSolution7">
   <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </button>
-<div id="collapseSolution7" class="accordion-collapse collapse" aria-labelledby="headingSolution7" data-bs-parent="#accordionSolution7">
+<div id="collapseSolution7" class="accordion-collapse collapse" data-bs-parent="#accordionSolution7" aria-labelledby="headingSolution7">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="use-9-years-out-of-the-dataset">1. Use 9 years out of the dataset<a class="anchor" aria-label="anchor" href="#use-9-years-out-of-the-dataset"></a></h3>
@@ -1758,7 +1758,7 @@ <h3 class="callout-title">Open question: What could be next steps to further imp
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution8" aria-expanded="false" aria-controls="collapseSolution8">
   <h4 class="accordion-header" id="headingSolution8"> Show me the solution </h4>
 </button>
-<div id="collapseSolution8" class="accordion-collapse collapse" aria-labelledby="headingSolution8" data-bs-parent="#accordionSolution8">
+<div id="collapseSolution8" class="accordion-collapse collapse" data-bs-parent="#accordionSolution8" aria-labelledby="headingSolution8">
 <div class="accordion-body">
 <p>This is an open question. And we don’t actually know how far one
 could push this sunshine hour prediction (try it out yourself if you
diff --git a/4-advanced-layer-types.html b/4-advanced-layer-types.html
index 4829e7eb..8fea7d19 100644
--- a/4-advanced-layer-types.html
+++ b/4-advanced-layer-types.html
@@ -235,20 +235,15 @@ <h2 class="accordion-header chapters" id="flush-headingEleven">
       <div class="accordion-body">
         <ul><li><a href="#different-types-of-layers">Different types of layers</a></li>
 <li><a href="#formulate-outline-the-problem-image-classification">1. Formulate / Outline the problem: Image classification</a></li>
-<li><a href="#dollar-street-10-dataset">Dollar Street 10 dataset</a></li>
 <li><a href="#identify-inputs-and-outputs">2. Identify inputs and outputs</a></li>
 <li><a href="#prepare-data">3. Prepare data</a></li>
 <li><a href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch">4. Choose a pretrained model or start building architecture from
 scratch</a></li>
-<li><a href="#convolutional-layers">Convolutional layers</a></li>
-<li><a href="#pooling-layers">Pooling layers</a></li>
 <li><a href="#choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer</a></li>
 <li><a href="#train-the-model">6. Train the model</a></li>
 <li><a href="#perform-a-predictionclassification">7. Perform a Prediction/Classification</a></li>
 <li><a href="#measure-performance">8. Measure performance</a></li>
 <li><a href="#refine-the-model">9. Refine the model</a></li>
-<li><a href="#dropout">Dropout</a></li>
-<li><a href="#hyperparameter-tuning">Hyperparameter tuning</a></li>
 <li><a href="#share-model">10. Share model</a></li>
 <li><a href="#conclusion-and-next-steps">Conclusion and next steps</a></li>
         </ul></div><!--/div.accordion-body-->
@@ -436,9 +431,9 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a></h2>
-<hr class="half-width"><p>The Dollar Street 10 dataset consists of images of 10 different
+<div class="section level3">
+<h3 id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a></h3>
+<p>The Dollar Street 10 dataset consists of images of 10 different
 classes, this is the mapping of the categories:</p>
 <table class="table"><thead><tr class="header"><th>Category</th>
 <th>label</th>
@@ -465,7 +460,8 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 </tr></tbody></table><figure><img src="fig/04_dollar_street_10.png" alt="A 5 by 5 grid of 25 sample images from the dollar street 10 data-set. Each image is labelled with a category, for example: 'street sign' or 'soap dispenser'." class="figure mx-auto d-block"><div class="figcaption">Sample images from the Dollar Street 10 dataset.
 Each image is labelled with a category, for example: ‘street sign’ or
 ‘soap dispenser’</div>
-</figure></section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
+</figure></div>
+</section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 <a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a></h2>
 <hr class="half-width"><div class="section level3">
 <h3 id="explore-the-data">Explore the data<a class="anchor" aria-label="anchor" href="#explore-the-data"></a></h3>
@@ -571,9 +567,9 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 </section><section><h2 class="section-heading" id="choose-a-pretrained-model-or-start-building-architecture-from-scratch">4. Choose a pretrained model or start building architecture from
 scratch<a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a></h2>
-<hr class="half-width"></section><section><h2 class="section-heading" id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a></h2>
-<hr class="half-width"><p>In the previous episodes, we used ‘fully connected layers’ , that
+<hr class="half-width"><div class="section level3">
+<h3 id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a></h3>
+<p>In the previous episodes, we used ‘fully connected layers’ , that
 connected all input values of a layer to all outputs of a layer. This
 results in many connections, and thus many weights to be learned, in the
 network. Note that our input dimension is now quite high (even with
@@ -882,9 +878,10 @@ <h3 class="callout-title">Search for existing architectures or pretrained models
 </div>
 </div>
 
-</section><section><h2 class="section-heading" id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-<a class="anchor" aria-label="anchor" href="#pooling-layers"></a></h2>
-<hr class="half-width"><p>Often in convolutional neural networks, the convolutional layers are
+</div>
+<div class="section level3">
+<h3 id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a></h3>
+<p>Often in convolutional neural networks, the convolutional layers are
 intertwined with <strong>Pooling layers</strong>. As opposed to the
 convolutional layer, the pooling layer actually alters the dimensions of
 the image and reduces it by a scaling factor. It is basically decreasing
@@ -944,6 +941,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 Non-trainable params: 0 (0.00 Byte)
 _________________________________________________________________</code></pre>
 </div>
+</div>
 </section><section><h2 class="section-heading" id="choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer<a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a></h2>
 <hr class="half-width"><p>We compile the model using the adam optimizer (other optimizers could
@@ -1271,9 +1269,9 @@ <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
-<a class="anchor" aria-label="anchor" href="#dropout"></a></h2>
-<hr class="half-width"><p>Note that the training loss continues to decrease, while the
+<div class="section level3">
+<h3 id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a></h3>
+<p>Note that the training loss continues to decrease, while the
 validation loss stagnates, and even starts to increase over the course
 of the epochs. Similarly, the accuracy for the validation set does not
 improve anymore after some epochs. This means we are overfitting on our
@@ -1464,9 +1462,11 @@ <h3 id="term-associated-to-this-procedure">2. Term associated to this procedure<
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a></h2>
-<hr class="half-width"><p>Recall that hyperparameters are model configuration settings that are
+</div>
+<div class="section level3">
+<h3 id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a></h3>
+
+<p>Recall that hyperparameters are model configuration settings that are
 chosen before the training process and affect the model’s learning
 behavior and performance, for example the dropout rate. In general, if
 you are varying hyperparameters to find the combination of
@@ -1732,6 +1732,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 hyperparameter tuning strategies, including <a href="https://en.wikipedia.org/wiki/Neural_architecture_search" class="external-link">neural
 architecture search</a> where a separate neural network is trained to
 find the best architecture for a model!</p>
+</div>
 </section><section><h2 class="section-heading" id="share-model">10. Share model<a class="anchor" aria-label="anchor" href="#share-model"></a>
 <a class="anchor" aria-label="anchor" href="#share-model"></a></h2>
 <hr class="half-width"><p>Let’s save our model</p>
diff --git a/5-transfer-learning.html b/5-transfer-learning.html
index 2388cf38..23d28722 100644
--- a/5-transfer-learning.html
+++ b/5-transfer-learning.html
@@ -525,7 +525,7 @@ <h3 id="max-pooling">3. Max pooling<a class="anchor" aria-label="anchor" href="#
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Solutions </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="trainable-parameters-1">1. Trainable parameters<a class="anchor" aria-label="anchor" href="#trainable-parameters-1"></a></h3>
@@ -589,7 +589,7 @@ <h3 id="optional-try-out-other-pre-trained-neural-networks">4. (Optional) Try ou
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="compile-the-model-1">1. Compile the model<a class="anchor" aria-label="anchor" href="#compile-the-model-1"></a></h3>
diff --git a/aio.html b/aio.html
index 412aa02c..182047fc 100644
--- a/aio.html
+++ b/aio.html
@@ -1612,7 +1612,7 @@ <h3 class="callout-title">Pairplot<a class="anchor" aria-label="anchor" href="#p
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ul>
 <li>The plots show that the green class, Gentoo is somewhat more easily
@@ -1748,7 +1748,7 @@ <h3 class="callout-title">One-hot encoding<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <p>C: 3, one for each output variable class</p>
 </div>
@@ -1971,7 +1971,7 @@ <h4 id="optional-keras-sequential-vs-functional-api">(optional) Keras Sequential
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>Have a look at the output of <code>model.summary()</code>:</p>
 <div class="codewrapper sourceCode" id="cb18">
@@ -2163,7 +2163,7 @@ <h3 class="callout-title">The Training Curve<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>The training loss decreases quickly. It drops in a smooth line with
@@ -2394,7 +2394,7 @@ <h3 class="callout-title">Confusion Matrix<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The confusion matrix shows that the predictions for Adelie and Gentoo
 are decent, but could be improved. However, Chinstrap is not predicted
@@ -2839,7 +2839,7 @@ <h3 class="callout-title">Exercise: Architecture of the network<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>The shape of the input layer has to correspond to the number of
@@ -3016,7 +3016,7 @@ <h3 id="when-the-batch-size-is-increased">3. When the batch size is increased:<a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li><p>Correct answer: B. To find the weights that minimize the loss
@@ -3262,7 +3262,7 @@ <h3 class="callout-title">Exercise: Reflecting on our results<a class="anchor" a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>While the performance on the train set seems reasonable, the
 performance on the test set is much worse. This is a common problem
@@ -3388,7 +3388,7 @@ <h3 class="callout-title">Exercise: Baseline<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>This really depends on your definition of hard! The baseline gives a
@@ -3492,7 +3492,7 @@ <h3 class="callout-title">Exercise: plot the training progress.<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The difference in the two curves shows that something is not
 completely right here. The error for the model predictions on the
@@ -3543,7 +3543,7 @@ <h3 class="callout-title">Exercise: Try to reduce the degree of overfitting by l
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution6" aria-expanded="false" aria-controls="collapseSolution6">
   <h4 class="accordion-header" id="headingSolution6"> Show me the solution </h4>
 </button>
-<div id="collapseSolution6" class="accordion-collapse collapse" aria-labelledby="headingSolution6" data-bs-parent="#accordionSolution6">
+<div id="collapseSolution6" class="accordion-collapse collapse" data-bs-parent="#accordionSolution6" aria-labelledby="headingSolution6">
 <div class="accordion-body">
 <p>Let’s first adapt our <code>create_nn</code> function so that we can
 tweak the number of nodes in the 2 layers by passing arguments to the
@@ -3841,7 +3841,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution7" aria-expanded="false" aria-controls="collapseSolution7">
   <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </button>
-<div id="collapseSolution7" class="accordion-collapse collapse" aria-labelledby="headingSolution7" data-bs-parent="#accordionSolution7">
+<div id="collapseSolution7" class="accordion-collapse collapse" data-bs-parent="#accordionSolution7" aria-labelledby="headingSolution7">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="use-9-years-out-of-the-dataset">1. Use 9 years out of the dataset<a class="anchor" aria-label="anchor" href="#use-9-years-out-of-the-dataset"></a>
@@ -4045,7 +4045,7 @@ <h3 class="callout-title">Open question: What could be next steps to further imp
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution8" aria-expanded="false" aria-controls="collapseSolution8">
   <h4 class="accordion-header" id="headingSolution8"> Show me the solution </h4>
 </button>
-<div id="collapseSolution8" class="accordion-collapse collapse" aria-labelledby="headingSolution8" data-bs-parent="#accordionSolution8">
+<div id="collapseSolution8" class="accordion-collapse collapse" data-bs-parent="#accordionSolution8" aria-labelledby="headingSolution8">
 <div class="accordion-body">
 <p>This is an open question. And we don’t actually know how far one
 could push this sunshine hour prediction (try it out yourself if you
@@ -4207,10 +4207,9 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-</h2>
-<hr class="half-width">
+<div class="section level3">
+<h3 id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
+</h3>
 <p>The Dollar Street 10 dataset consists of images of 10 different
 classes, this is the mapping of the categories:</p>
 <table class="table">
@@ -4264,7 +4263,9 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 <figure><img src="fig/04_dollar_street_10.png" alt="A 5 by 5 grid of 25 sample images from the dollar street 10 data-set. Each image is labelled with a category, for example: 'street sign' or 'soap dispenser'." class="figure mx-auto d-block"><div class="figcaption">Sample images from the Dollar Street 10 dataset.
 Each image is labelled with a category, for example: ‘street sign’ or
 ‘soap dispenser’</div>
-</figure></section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
+</figure>
+</div>
+</section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 <a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 </h2>
 <hr class="half-width">
@@ -4378,10 +4379,10 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 scratch<a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 </h2>
-<hr class="half-width"></section><section><h2 class="section-heading" id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-</h2>
 <hr class="half-width">
+<div class="section level3">
+<h3 id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
+</h3>
 <p>In the previous episodes, we used ‘fully connected layers’ , that
 connected all input values of a layer to all outputs of a layer. This
 results in many connections, and thus many weights to be learned, in the
@@ -4699,10 +4700,10 @@ <h3 class="callout-title">Search for existing architectures or pretrained models
 </div>
 </div>
 
-</section><section><h2 class="section-heading" id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-</h2>
-<hr class="half-width">
+</div>
+<div class="section level3">
+<h3 id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
+</h3>
 <p>Often in convolutional neural networks, the convolutional layers are
 intertwined with <strong>Pooling layers</strong>. As opposed to the
 convolutional layer, the pooling layer actually alters the dimensions of
@@ -4763,6 +4764,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 Non-trainable params: 0 (0.00 Byte)
 _________________________________________________________________</code></pre>
 </div>
+</div>
 </section><section><h2 class="section-heading" id="choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer<a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 </h2>
@@ -5107,10 +5109,9 @@ <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
-<a class="anchor" aria-label="anchor" href="#dropout"></a>
-</h2>
-<hr class="half-width">
+<div class="section level3">
+<h3 id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
+</h3>
 <p>Note that the training loss continues to decrease, while the
 validation loss stagnates, and even starts to increase over the course
 of the epochs. Similarly, the accuracy for the validation set does not
@@ -5307,10 +5308,11 @@ <h3 id="term-associated-to-this-procedure">2. Term associated to this procedure<
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-</h2>
-<hr class="half-width">
+</div>
+<div class="section level3">
+<h3 id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
+</h3>
+
 <p>Recall that hyperparameters are model configuration settings that are
 chosen before the training process and affect the model’s learning
 behavior and performance, for example the dropout rate. In general, if
@@ -5581,6 +5583,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 hyperparameter tuning strategies, including <a href="https://en.wikipedia.org/wiki/Neural_architecture_search" class="external-link">neural
 architecture search</a> where a separate neural network is trained to
 find the best architecture for a model!</p>
+</div>
 </section><section><h2 class="section-heading" id="share-model">10. Share model<a class="anchor" aria-label="anchor" href="#share-model"></a>
 <a class="anchor" aria-label="anchor" href="#share-model"></a>
 </h2>
@@ -5843,7 +5846,7 @@ <h3 id="max-pooling">3. Max pooling<a class="anchor" aria-label="anchor" href="#
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Solutions </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="trainable-parameters-1">1. Trainable parameters<a class="anchor" aria-label="anchor" href="#trainable-parameters-1"></a>
@@ -5913,7 +5916,7 @@ <h3 id="optional-try-out-other-pre-trained-neural-networks">4. (Optional) Try ou
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="compile-the-model-1">1. Compile the model<a class="anchor" aria-label="anchor" href="#compile-the-model-1"></a>
diff --git a/index.html b/index.html
index ee6b051d..34a4e023 100644
--- a/index.html
+++ b/index.html
@@ -388,7 +388,7 @@ <h3 class="accordion-header" id="headingSpoiler1">
   Windows
   </h3>
 </button>
-<div id="collapseSpoiler1" class="accordion-collapse collapse" aria-labelledby="headingSpoiler1" data-bs-parent="#accordionSpoiler1">
+<div id="collapseSpoiler1" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler1" aria-labelledby="headingSpoiler1">
 <div class="accordion-body">
 <p>Checkout the <a href="https://www.youtube.com/watch?v=xxQ0mzZ8UvA" class="external-link">video tutorial</a>
 or:</p>
@@ -410,7 +410,7 @@ <h3 class="accordion-header" id="headingSpoiler2">
   MacOS
   </h3>
 </button>
-<div id="collapseSpoiler2" class="accordion-collapse collapse" aria-labelledby="headingSpoiler2" data-bs-parent="#accordionSpoiler2">
+<div id="collapseSpoiler2" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler2" aria-labelledby="headingSpoiler2">
 <div class="accordion-body">
 <p>Checkout the <a href="https://www.youtube.com/watch?v=TcSAln46u9U" class="external-link">video tutorial</a>
 or:</p>
@@ -432,7 +432,7 @@ <h3 class="accordion-header" id="headingSpoiler3">
   Linux
   </h3>
 </button>
-<div id="collapseSpoiler3" class="accordion-collapse collapse" aria-labelledby="headingSpoiler3" data-bs-parent="#accordionSpoiler3">
+<div id="collapseSpoiler3" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler3" aria-labelledby="headingSpoiler3">
 <div class="accordion-body">
 <p>Note that the following installation steps require you to work from
 the shell. If you run into any difficulties, please request help before
@@ -515,7 +515,7 @@ <h3 class="accordion-header" id="headingSpoiler4">
   Troubleshooting for Windows
   </h3>
 </button>
-<div id="collapseSpoiler4" class="accordion-collapse collapse" aria-labelledby="headingSpoiler4" data-bs-parent="#accordionSpoiler4">
+<div id="collapseSpoiler4" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler4" aria-labelledby="headingSpoiler4">
 <div class="accordion-body">
 <p>It is possible that Windows users will run into version conflicts. If
 you are on Windows and get errors running the command, you can try
@@ -539,7 +539,7 @@ <h3 class="accordion-header" id="headingSpoiler5">
   Troubleshooting for Macs with Apple silicon chip
   </h3>
 </button>
-<div id="collapseSpoiler5" class="accordion-collapse collapse" aria-labelledby="headingSpoiler5" data-bs-parent="#accordionSpoiler5">
+<div id="collapseSpoiler5" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler5" aria-labelledby="headingSpoiler5">
 <div class="accordion-body">
 <p>Newer Macs (from 2020 onwards) often have a different kind of chip,
 manufactured by Apple instead of Intel. This can lead to problems
diff --git a/instructor/2-keras.html b/instructor/2-keras.html
index 91ff2342..e2a2839a 100644
--- a/instructor/2-keras.html
+++ b/instructor/2-keras.html
@@ -405,7 +405,7 @@ <h3 class="card-title">Objectives</h3>
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor1" aria-expanded="false" aria-controls="collapseInstructor1">
   <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> Instructor Note </h3>
 </button>
-<div id="collapseInstructor1" class="accordion-collapse collapse" aria-labelledby="headingInstructor1" data-bs-parent="#accordionInstructor1">
+<div id="collapseInstructor1" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor1" aria-labelledby="headingInstructor1">
 <div class="accordion-body">
 <p>This episode really aims to go through the whole process once, as
 quickly as possible. In episode 3 we will expand on all the concepts
@@ -604,7 +604,7 @@ <h3 class="callout-title">Pairplot<a class="anchor" aria-label="anchor" href="#p
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ul><li>The plots show that the green class, Gentoo is somewhat more easily
 distinguishable from the other two.</li>
@@ -731,7 +731,7 @@ <h3 class="callout-title">One-hot encoding<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <p>C: 3, one for each output variable class</p>
 </div>
@@ -783,7 +783,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor2" aria-expanded="false" aria-controls="collapseInstructor2">
   <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor2" class="accordion-collapse collapse" aria-labelledby="headingInstructor2" data-bs-parent="#accordionInstructor2">
+<div id="collapseInstructor2" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor2" aria-labelledby="headingInstructor2">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -953,7 +953,7 @@ <h4 id="optional-keras-sequential-vs-functional-api">(optional) Keras Sequential
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>Have a look at the output of <code>model.summary()</code>:</p>
 <div class="codewrapper sourceCode" id="cb18">
@@ -1130,7 +1130,7 @@ <h3 class="callout-title">The Training Curve<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>The training loss decreases quickly. It drops in a smooth line with
 little jitter. This is ideal for a training curve.</li>
@@ -1253,7 +1253,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor3" aria-expanded="false" aria-controls="collapseInstructor3">
   <h3 class="accordion-header" id="headingInstructor3">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor3" class="accordion-collapse collapse" aria-labelledby="headingInstructor3" data-bs-parent="#accordionInstructor3">
+<div id="collapseInstructor3" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor3" aria-labelledby="headingInstructor3">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -1339,7 +1339,7 @@ <h3 class="callout-title">Confusion Matrix<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The confusion matrix shows that the predictions for Adelie and Gentoo
 are decent, but could be improved. However, Chinstrap is not predicted
diff --git a/instructor/3-monitor-the-model.html b/instructor/3-monitor-the-model.html
index c16dce95..edfd764f 100644
--- a/instructor/3-monitor-the-model.html
+++ b/instructor/3-monitor-the-model.html
@@ -380,7 +380,7 @@ <h3 class="card-title">Objectives</h3>
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor1" aria-expanded="false" aria-controls="collapseInstructor1">
   <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> Copy-pasting code </h3>
 </button>
-<div id="collapseInstructor1" class="accordion-collapse collapse" aria-labelledby="headingInstructor1" data-bs-parent="#accordionInstructor1">
+<div id="collapseInstructor1" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor1" aria-labelledby="headingInstructor1">
 <div class="accordion-body">
 <p>In this episode we first introduce a simple approach to the problem,
 then we iterate on that a few times to, step-by-step, working towards a
@@ -599,7 +599,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor2" aria-expanded="false" aria-controls="collapseInstructor2">
   <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor2" class="accordion-collapse collapse" aria-labelledby="headingInstructor2" data-bs-parent="#accordionInstructor2">
+<div id="collapseInstructor2" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor2" aria-labelledby="headingInstructor2">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -651,7 +651,7 @@ <h3 class="callout-title">Exercise: Architecture of the network<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>The shape of the input layer has to correspond to the number of
 features in our data: 89</li>
@@ -812,7 +812,7 @@ <h3 id="when-the-batch-size-is-increased">3. When the batch size is increased:<a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li><p>Correct answer: B. To find the weights that minimize the loss
 function. The loss function quantifies the total error of the network,
@@ -987,7 +987,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor3" aria-expanded="false" aria-controls="collapseInstructor3">
   <h3 class="accordion-header" id="headingInstructor3">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor3" class="accordion-collapse collapse" aria-labelledby="headingInstructor3" data-bs-parent="#accordionInstructor3">
+<div id="collapseInstructor3" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor3" aria-labelledby="headingInstructor3">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -1050,7 +1050,7 @@ <h3 class="callout-title">Exercise: Reflecting on our results<a class="anchor" a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>While the performance on the train set seems reasonable, the
 performance on the test set is much worse. This is a common problem
@@ -1172,7 +1172,7 @@ <h3 class="callout-title">Exercise: Baseline<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal"><li>This really depends on your definition of hard! The baseline gives a
 more accurate prediction than just randomly predicting a number, so the
@@ -1269,7 +1269,7 @@ <h3 class="callout-title">Exercise: plot the training progress.<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The difference in the two curves shows that something is not
 completely right here. The error for the model predictions on the
@@ -1317,7 +1317,7 @@ <h3 class="callout-title">Exercise: Try to reduce the degree of overfitting by l
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution6" aria-expanded="false" aria-controls="collapseSolution6">
   <h4 class="accordion-header" id="headingSolution6"> Show me the solution </h4>
 </button>
-<div id="collapseSolution6" class="accordion-collapse collapse" aria-labelledby="headingSolution6" data-bs-parent="#accordionSolution6">
+<div id="collapseSolution6" class="accordion-collapse collapse" data-bs-parent="#accordionSolution6" aria-labelledby="headingSolution6">
 <div class="accordion-body">
 <p>Let’s first adapt our <code>create_nn</code> function so that we can
 tweak the number of nodes in the 2 layers by passing arguments to the
@@ -1606,7 +1606,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution7" aria-expanded="false" aria-controls="collapseSolution7">
   <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </button>
-<div id="collapseSolution7" class="accordion-collapse collapse" aria-labelledby="headingSolution7" data-bs-parent="#accordionSolution7">
+<div id="collapseSolution7" class="accordion-collapse collapse" data-bs-parent="#accordionSolution7" aria-labelledby="headingSolution7">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="use-9-years-out-of-the-dataset">1. Use 9 years out of the dataset<a class="anchor" aria-label="anchor" href="#use-9-years-out-of-the-dataset"></a></h3>
@@ -1800,7 +1800,7 @@ <h3 class="callout-title">Open question: What could be next steps to further imp
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution8" aria-expanded="false" aria-controls="collapseSolution8">
   <h4 class="accordion-header" id="headingSolution8"> Show me the solution </h4>
 </button>
-<div id="collapseSolution8" class="accordion-collapse collapse" aria-labelledby="headingSolution8" data-bs-parent="#accordionSolution8">
+<div id="collapseSolution8" class="accordion-collapse collapse" data-bs-parent="#accordionSolution8" aria-labelledby="headingSolution8">
 <div class="accordion-body">
 <p>This is an open question. And we don’t actually know how far one
 could push this sunshine hour prediction (try it out yourself if you
diff --git a/instructor/4-advanced-layer-types.html b/instructor/4-advanced-layer-types.html
index 4f5f379f..e57d9022 100644
--- a/instructor/4-advanced-layer-types.html
+++ b/instructor/4-advanced-layer-types.html
@@ -235,20 +235,15 @@ <h2 class="accordion-header chapters" id="flush-headingEleven">
       <div class="accordion-body">
         <ul><li><a href="#different-types-of-layers">Different types of layers</a></li>
 <li><a href="#formulate-outline-the-problem-image-classification">1. Formulate / Outline the problem: Image classification</a></li>
-<li><a href="#dollar-street-10-dataset">Dollar Street 10 dataset</a></li>
 <li><a href="#identify-inputs-and-outputs">2. Identify inputs and outputs</a></li>
 <li><a href="#prepare-data">3. Prepare data</a></li>
 <li><a href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch">4. Choose a pretrained model or start building architecture from
 scratch</a></li>
-<li><a href="#convolutional-layers">Convolutional layers</a></li>
-<li><a href="#pooling-layers">Pooling layers</a></li>
 <li><a href="#choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer</a></li>
 <li><a href="#train-the-model">6. Train the model</a></li>
 <li><a href="#perform-a-predictionclassification">7. Perform a Prediction/Classification</a></li>
 <li><a href="#measure-performance">8. Measure performance</a></li>
 <li><a href="#refine-the-model">9. Refine the model</a></li>
-<li><a href="#dropout">Dropout</a></li>
-<li><a href="#hyperparameter-tuning">Hyperparameter tuning</a></li>
 <li><a href="#share-model">10. Share model</a></li>
 <li><a href="#conclusion-and-next-steps">Conclusion and next steps</a></li>
         </ul></div><!--/div.accordion-body-->
@@ -438,9 +433,9 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a></h2>
-<hr class="half-width"><p>The Dollar Street 10 dataset consists of images of 10 different
+<div class="section level3">
+<h3 id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a></h3>
+<p>The Dollar Street 10 dataset consists of images of 10 different
 classes, this is the mapping of the categories:</p>
 <table class="table"><thead><tr class="header"><th>Category</th>
 <th>label</th>
@@ -490,7 +485,8 @@ <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square">
 <figure><img src="../fig/04_dollar_street_10.png" alt="A 5 by 5 grid of 25 sample images from the dollar street 10 data-set. Each image is labelled with a category, for example: 'street sign' or 'soap dispenser'." class="figure mx-auto d-block"><div class="figcaption">Sample images from the Dollar Street 10 dataset.
 Each image is labelled with a category, for example: ‘street sign’ or
 ‘soap dispenser’</div>
-</figure></section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
+</figure></div>
+</section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 <a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a></h2>
 <hr class="half-width"><div class="section level3">
 <h3 id="explore-the-data">Explore the data<a class="anchor" aria-label="anchor" href="#explore-the-data"></a></h3>
@@ -596,9 +592,9 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 </section><section><h2 class="section-heading" id="choose-a-pretrained-model-or-start-building-architecture-from-scratch">4. Choose a pretrained model or start building architecture from
 scratch<a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a></h2>
-<hr class="half-width"></section><section><h2 class="section-heading" id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a></h2>
-<hr class="half-width"><p>In the previous episodes, we used ‘fully connected layers’ , that
+<hr class="half-width"><div class="section level3">
+<h3 id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a></h3>
+<p>In the previous episodes, we used ‘fully connected layers’ , that
 connected all input values of a layer to all outputs of a layer. This
 results in many connections, and thus many weights to be learned, in the
 network. Note that our input dimension is now quite high (even with
@@ -930,9 +926,10 @@ <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square">
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-<a class="anchor" aria-label="anchor" href="#pooling-layers"></a></h2>
-<hr class="half-width"><p>Often in convolutional neural networks, the convolutional layers are
+</div>
+<div class="section level3">
+<h3 id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a></h3>
+<p>Often in convolutional neural networks, the convolutional layers are
 intertwined with <strong>Pooling layers</strong>. As opposed to the
 convolutional layer, the pooling layer actually alters the dimensions of
 the image and reduces it by a scaling factor. It is basically decreasing
@@ -992,6 +989,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 Non-trainable params: 0 (0.00 Byte)
 _________________________________________________________________</code></pre>
 </div>
+</div>
 </section><section><h2 class="section-heading" id="choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer<a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a></h2>
 <hr class="half-width"><p>We compile the model using the adam optimizer (other optimizers could
@@ -1352,9 +1350,9 @@ <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
-<a class="anchor" aria-label="anchor" href="#dropout"></a></h2>
-<hr class="half-width"><p>Note that the training loss continues to decrease, while the
+<div class="section level3">
+<h3 id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a></h3>
+<p>Note that the training loss continues to decrease, while the
 validation loss stagnates, and even starts to increase over the course
 of the epochs. Similarly, the accuracy for the validation set does not
 improve anymore after some epochs. This means we are overfitting on our
@@ -1545,9 +1543,10 @@ <h3 id="term-associated-to-this-procedure">2. Term associated to this procedure<
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a></h2>
-<hr class="half-width"><div id="accordionInstructor5" class="accordion instructor-note accordion-flush">
+</div>
+<div class="section level3">
+<h3 id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a></h3>
+<div id="accordionInstructor5" class="accordion instructor-note accordion-flush">
 <div class="accordion-item">
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor5" aria-expanded="false" aria-controls="collapseInstructor5">
   <h3 class="accordion-header" id="headingInstructor5">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> Do a live demo instead of live coding </h3>
@@ -1831,6 +1830,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 hyperparameter tuning strategies, including <a href="https://en.wikipedia.org/wiki/Neural_architecture_search" class="external-link">neural
 architecture search</a> where a separate neural network is trained to
 find the best architecture for a model!</p>
+</div>
 </section><section><h2 class="section-heading" id="share-model">10. Share model<a class="anchor" aria-label="anchor" href="#share-model"></a>
 <a class="anchor" aria-label="anchor" href="#share-model"></a></h2>
 <hr class="half-width"><p>Let’s save our model</p>
diff --git a/instructor/5-transfer-learning.html b/instructor/5-transfer-learning.html
index 51955fe7..b4dcf11e 100644
--- a/instructor/5-transfer-learning.html
+++ b/instructor/5-transfer-learning.html
@@ -527,7 +527,7 @@ <h3 id="max-pooling">3. Max pooling<a class="anchor" aria-label="anchor" href="#
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Solutions </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="trainable-parameters-1">1. Trainable parameters<a class="anchor" aria-label="anchor" href="#trainable-parameters-1"></a></h3>
@@ -591,7 +591,7 @@ <h3 id="optional-try-out-other-pre-trained-neural-networks">4. (Optional) Try ou
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="compile-the-model-1">1. Compile the model<a class="anchor" aria-label="anchor" href="#compile-the-model-1"></a></h3>
diff --git a/instructor/aio.html b/instructor/aio.html
index b6c2b63a..a6ab18ae 100644
--- a/instructor/aio.html
+++ b/instructor/aio.html
@@ -1420,7 +1420,7 @@ <h3 class="card-title">Objectives</h3>
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor1" aria-expanded="false" aria-controls="collapseInstructor1">
   <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> Instructor Note </h3>
 </button>
-<div id="collapseInstructor1" class="accordion-collapse collapse" aria-labelledby="headingInstructor1" data-bs-parent="#accordionInstructor1">
+<div id="collapseInstructor1" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor1" aria-labelledby="headingInstructor1">
 <div class="accordion-body">
 <p>This episode really aims to go through the whole process once, as
 quickly as possible. In episode 3 we will expand on all the concepts
@@ -1654,7 +1654,7 @@ <h3 class="callout-title">Pairplot<a class="anchor" aria-label="anchor" href="#p
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ul>
 <li>The plots show that the green class, Gentoo is somewhat more easily
@@ -1790,7 +1790,7 @@ <h3 class="callout-title">One-hot encoding<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <p>C: 3, one for each output variable class</p>
 </div>
@@ -1843,7 +1843,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor2" aria-expanded="false" aria-controls="collapseInstructor2">
   <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor2" class="accordion-collapse collapse" aria-labelledby="headingInstructor2" data-bs-parent="#accordionInstructor2">
+<div id="collapseInstructor2" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor2" aria-labelledby="headingInstructor2">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -2024,7 +2024,7 @@ <h4 id="optional-keras-sequential-vs-functional-api">(optional) Keras Sequential
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>Have a look at the output of <code>model.summary()</code>:</p>
 <div class="codewrapper sourceCode" id="cb18">
@@ -2216,7 +2216,7 @@ <h3 class="callout-title">The Training Curve<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>The training loss decreases quickly. It drops in a smooth line with
@@ -2367,7 +2367,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor3" aria-expanded="false" aria-controls="collapseInstructor3">
   <h3 class="accordion-header" id="headingInstructor3">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor3" class="accordion-collapse collapse" aria-labelledby="headingInstructor3" data-bs-parent="#accordionInstructor3">
+<div id="collapseInstructor3" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor3" aria-labelledby="headingInstructor3">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -2458,7 +2458,7 @@ <h3 class="callout-title">Confusion Matrix<a class="anchor" aria-label="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The confusion matrix shows that the predictions for Adelie and Gentoo
 are decent, but could be improved. However, Chinstrap is not predicted
@@ -2623,7 +2623,7 @@ <h3 class="card-title">Objectives</h3>
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor1" aria-expanded="false" aria-controls="collapseInstructor1">
   <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> Copy-pasting code </h3>
 </button>
-<div id="collapseInstructor1" class="accordion-collapse collapse" aria-labelledby="headingInstructor1" data-bs-parent="#accordionInstructor1">
+<div id="collapseInstructor1" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor1" aria-labelledby="headingInstructor1">
 <div class="accordion-body">
 <p>In this episode we first introduce a simple approach to the problem,
 then we iterate on that a few times to, step-by-step, working towards a
@@ -2877,7 +2877,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor2" aria-expanded="false" aria-controls="collapseInstructor2">
   <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor2" class="accordion-collapse collapse" aria-labelledby="headingInstructor2" data-bs-parent="#accordionInstructor2">
+<div id="collapseInstructor2" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor2" aria-labelledby="headingInstructor2">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -2934,7 +2934,7 @@ <h3 class="callout-title">Exercise: Architecture of the network<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Show me the solution </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>The shape of the input layer has to correspond to the number of
@@ -3111,7 +3111,7 @@ <h3 id="when-the-batch-size-is-increased">3. When the batch size is increased:<a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li><p>Correct answer: B. To find the weights that minimize the loss
@@ -3299,7 +3299,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor3" aria-expanded="false" aria-controls="collapseInstructor3">
   <h3 class="accordion-header" id="headingInstructor3">  <div class="note-square"><i aria-hidden="true" class="callout-icon" data-feather="edit-2"></i></div> BREAK </h3>
 </button>
-<div id="collapseInstructor3" class="accordion-collapse collapse" aria-labelledby="headingInstructor3" data-bs-parent="#accordionInstructor3">
+<div id="collapseInstructor3" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor3" aria-labelledby="headingInstructor3">
 <div class="accordion-body">
 <p>This is a good time for switching instructor and/or a break.</p>
 </div>
@@ -3368,7 +3368,7 @@ <h3 class="callout-title">Exercise: Reflecting on our results<a class="anchor" a
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution3" aria-expanded="false" aria-controls="collapseSolution3">
   <h4 class="accordion-header" id="headingSolution3"> Show me the solution </h4>
 </button>
-<div id="collapseSolution3" class="accordion-collapse collapse" aria-labelledby="headingSolution3" data-bs-parent="#accordionSolution3">
+<div id="collapseSolution3" class="accordion-collapse collapse" data-bs-parent="#accordionSolution3" aria-labelledby="headingSolution3">
 <div class="accordion-body">
 <p>While the performance on the train set seems reasonable, the
 performance on the test set is much worse. This is a common problem
@@ -3494,7 +3494,7 @@ <h3 class="callout-title">Exercise: Baseline<a class="anchor" aria-label="anchor
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution4" aria-expanded="false" aria-controls="collapseSolution4">
   <h4 class="accordion-header" id="headingSolution4"> Show me the solution </h4>
 </button>
-<div id="collapseSolution4" class="accordion-collapse collapse" aria-labelledby="headingSolution4" data-bs-parent="#accordionSolution4">
+<div id="collapseSolution4" class="accordion-collapse collapse" data-bs-parent="#accordionSolution4" aria-labelledby="headingSolution4">
 <div class="accordion-body">
 <ol style="list-style-type: decimal">
 <li>This really depends on your definition of hard! The baseline gives a
@@ -3598,7 +3598,7 @@ <h3 class="callout-title">Exercise: plot the training progress.<a class="anchor"
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution5" aria-expanded="false" aria-controls="collapseSolution5">
   <h4 class="accordion-header" id="headingSolution5"> Show me the solution </h4>
 </button>
-<div id="collapseSolution5" class="accordion-collapse collapse" aria-labelledby="headingSolution5" data-bs-parent="#accordionSolution5">
+<div id="collapseSolution5" class="accordion-collapse collapse" data-bs-parent="#accordionSolution5" aria-labelledby="headingSolution5">
 <div class="accordion-body">
 <p>The difference in the two curves shows that something is not
 completely right here. The error for the model predictions on the
@@ -3649,7 +3649,7 @@ <h3 class="callout-title">Exercise: Try to reduce the degree of overfitting by l
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution6" aria-expanded="false" aria-controls="collapseSolution6">
   <h4 class="accordion-header" id="headingSolution6"> Show me the solution </h4>
 </button>
-<div id="collapseSolution6" class="accordion-collapse collapse" aria-labelledby="headingSolution6" data-bs-parent="#accordionSolution6">
+<div id="collapseSolution6" class="accordion-collapse collapse" data-bs-parent="#accordionSolution6" aria-labelledby="headingSolution6">
 <div class="accordion-body">
 <p>Let’s first adapt our <code>create_nn</code> function so that we can
 tweak the number of nodes in the 2 layers by passing arguments to the
@@ -3947,7 +3947,7 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution7" aria-expanded="false" aria-controls="collapseSolution7">
   <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </button>
-<div id="collapseSolution7" class="accordion-collapse collapse" aria-labelledby="headingSolution7" data-bs-parent="#accordionSolution7">
+<div id="collapseSolution7" class="accordion-collapse collapse" data-bs-parent="#accordionSolution7" aria-labelledby="headingSolution7">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="use-9-years-out-of-the-dataset">1. Use 9 years out of the dataset<a class="anchor" aria-label="anchor" href="#use-9-years-out-of-the-dataset"></a>
@@ -4151,7 +4151,7 @@ <h3 class="callout-title">Open question: What could be next steps to further imp
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution8" aria-expanded="false" aria-controls="collapseSolution8">
   <h4 class="accordion-header" id="headingSolution8"> Show me the solution </h4>
 </button>
-<div id="collapseSolution8" class="accordion-collapse collapse" aria-labelledby="headingSolution8" data-bs-parent="#accordionSolution8">
+<div id="collapseSolution8" class="accordion-collapse collapse" data-bs-parent="#accordionSolution8" aria-labelledby="headingSolution8">
 <div class="accordion-body">
 <p>This is an open question. And we don’t actually know how far one
 could push this sunshine hour prediction (try it out yourself if you
@@ -4314,10 +4314,9 @@ <h3 class="callout-title">A Note About Data Provenance<a class="anchor" aria-lab
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
-</h2>
-<hr class="half-width">
+<div class="section level3">
+<h3 id="dollar-street-10-dataset">Dollar Street 10 dataset<a class="anchor" aria-label="anchor" href="#dollar-street-10-dataset"></a>
+</h3>
 <p>The Dollar Street 10 dataset consists of images of 10 different
 classes, this is the mapping of the categories:</p>
 <table class="table">
@@ -4394,7 +4393,9 @@ <h3 class="accordion-header" id="headingInstructor1">  <div class="note-square">
 <figure><img src="../fig/04_dollar_street_10.png" alt="A 5 by 5 grid of 25 sample images from the dollar street 10 data-set. Each image is labelled with a category, for example: 'street sign' or 'soap dispenser'." class="figure mx-auto d-block"><div class="figcaption">Sample images from the Dollar Street 10 dataset.
 Each image is labelled with a category, for example: ‘street sign’ or
 ‘soap dispenser’</div>
-</figure></section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
+</figure>
+</div>
+</section><section><h2 class="section-heading" id="identify-inputs-and-outputs">2. Identify inputs and outputs<a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 <a class="anchor" aria-label="anchor" href="#identify-inputs-and-outputs"></a>
 </h2>
 <hr class="half-width">
@@ -4508,10 +4509,10 @@ <h3 class="code-label">PYTHON<i aria-hidden="true" data-feather="chevron-left"><
 scratch<a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-pretrained-model-or-start-building-architecture-from-scratch"></a>
 </h2>
-<hr class="half-width"></section><section><h2 class="section-heading" id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
-</h2>
 <hr class="half-width">
+<div class="section level3">
+<h3 id="convolutional-layers">Convolutional layers<a class="anchor" aria-label="anchor" href="#convolutional-layers"></a>
+</h3>
 <p>In the previous episodes, we used ‘fully connected layers’ , that
 connected all input values of a layer to all outputs of a layer. This
 results in many connections, and thus many weights to be learned, in the
@@ -4852,10 +4853,10 @@ <h3 class="accordion-header" id="headingInstructor2">  <div class="note-square">
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
-</h2>
-<hr class="half-width">
+</div>
+<div class="section level3">
+<h3 id="pooling-layers">Pooling layers<a class="anchor" aria-label="anchor" href="#pooling-layers"></a>
+</h3>
 <p>Often in convolutional neural networks, the convolutional layers are
 intertwined with <strong>Pooling layers</strong>. As opposed to the
 convolutional layer, the pooling layer actually alters the dimensions of
@@ -4916,6 +4917,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 Non-trainable params: 0 (0.00 Byte)
 _________________________________________________________________</code></pre>
 </div>
+</div>
 </section><section><h2 class="section-heading" id="choose-a-loss-function-and-optimizer">5. Choose a loss function and optimizer<a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 <a class="anchor" aria-label="anchor" href="#choose-a-loss-function-and-optimizer"></a>
 </h2>
@@ -5295,10 +5297,9 @@ <h4 class="accordion-header" id="headingSolution7"> Show me the solution </h4>
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
-<a class="anchor" aria-label="anchor" href="#dropout"></a>
-</h2>
-<hr class="half-width">
+<div class="section level3">
+<h3 id="dropout">Dropout<a class="anchor" aria-label="anchor" href="#dropout"></a>
+</h3>
 <p>Note that the training loss continues to decrease, while the
 validation loss stagnates, and even starts to increase over the course
 of the epochs. Similarly, the accuracy for the validation set does not
@@ -5495,10 +5496,10 @@ <h3 id="term-associated-to-this-procedure">2. Term associated to this procedure<
 </div>
 </div>
 </div>
-</section><section><h2 class="section-heading" id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
-</h2>
-<hr class="half-width">
+</div>
+<div class="section level3">
+<h3 id="hyperparameter-tuning">Hyperparameter tuning<a class="anchor" aria-label="anchor" href="#hyperparameter-tuning"></a>
+</h3>
 <div id="accordionInstructor5" class="accordion instructor-note accordion-flush">
 <div class="accordion-item">
 <button class="accordion-button instructor-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseInstructor5" aria-expanded="false" aria-controls="collapseInstructor5">
@@ -5787,6 +5788,7 @@ <h3 class="code-label">OUTPUT<i aria-hidden="true" data-feather="chevron-left"><
 hyperparameter tuning strategies, including <a href="https://en.wikipedia.org/wiki/Neural_architecture_search" class="external-link">neural
 architecture search</a> where a separate neural network is trained to
 find the best architecture for a model!</p>
+</div>
 </section><section><h2 class="section-heading" id="share-model">10. Share model<a class="anchor" aria-label="anchor" href="#share-model"></a>
 <a class="anchor" aria-label="anchor" href="#share-model"></a>
 </h2>
@@ -6050,7 +6052,7 @@ <h3 id="max-pooling">3. Max pooling<a class="anchor" aria-label="anchor" href="#
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution1" aria-expanded="false" aria-controls="collapseSolution1">
   <h4 class="accordion-header" id="headingSolution1"> Solutions </h4>
 </button>
-<div id="collapseSolution1" class="accordion-collapse collapse" aria-labelledby="headingSolution1" data-bs-parent="#accordionSolution1">
+<div id="collapseSolution1" class="accordion-collapse collapse" data-bs-parent="#accordionSolution1" aria-labelledby="headingSolution1">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="trainable-parameters-1">1. Trainable parameters<a class="anchor" aria-label="anchor" href="#trainable-parameters-1"></a>
@@ -6120,7 +6122,7 @@ <h3 id="optional-try-out-other-pre-trained-neural-networks">4. (Optional) Try ou
 <button class="accordion-button solution-button collapsed" type="button" data-bs-toggle="collapse" data-bs-target="#collapseSolution2" aria-expanded="false" aria-controls="collapseSolution2">
   <h4 class="accordion-header" id="headingSolution2"> Show me the solution </h4>
 </button>
-<div id="collapseSolution2" class="accordion-collapse collapse" aria-labelledby="headingSolution2" data-bs-parent="#accordionSolution2">
+<div id="collapseSolution2" class="accordion-collapse collapse" data-bs-parent="#accordionSolution2" aria-labelledby="headingSolution2">
 <div class="accordion-body">
 <div class="section level3">
 <h3 id="compile-the-model-1">1. Compile the model<a class="anchor" aria-label="anchor" href="#compile-the-model-1"></a>
diff --git a/instructor/index.html b/instructor/index.html
index 16698673..cf6278c0 100644
--- a/instructor/index.html
+++ b/instructor/index.html
@@ -353,7 +353,7 @@ <h3 class="accordion-header" id="headingInstructor1">
   Looking for Beta Testers!
   </h3>
 </button>
-<div id="collapseInstructor1" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor1" aria-labelledby="headingInstructor1">
+<div id="collapseInstructor1" class="accordion-collapse collapse" aria-labelledby="headingInstructor1" data-bs-parent="#accordionInstructor1">
 <div class="accordion-body">
 <p><strong>We are currently looking for volunteers to test this
 lesson!</strong> If you would like to teach this lesson in a pilot
@@ -375,7 +375,7 @@ <h3 class="accordion-header" id="headingInstructor2">
   Breaks and Schedule
   </h3>
 </button>
-<div id="collapseInstructor2" class="accordion-collapse collapse" data-bs-parent="#accordionInstructor2" aria-labelledby="headingInstructor2">
+<div id="collapseInstructor2" class="accordion-collapse collapse" aria-labelledby="headingInstructor2" data-bs-parent="#accordionInstructor2">
 <div class="accordion-body">
 <p>Episode 2, 3, and 4 in this workshop are relatively long. We suggest
 to have a break at least every 90 minutes and to switch the instructor
@@ -507,7 +507,7 @@ <h3 class="accordion-header" id="headingSpoiler1">
   Windows
   </h3>
 </button>
-<div id="collapseSpoiler1" class="accordion-collapse collapse" aria-labelledby="headingSpoiler1" data-bs-parent="#accordionSpoiler1">
+<div id="collapseSpoiler1" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler1" aria-labelledby="headingSpoiler1">
 <div class="accordion-body">
 <p>Checkout the <a href="https://www.youtube.com/watch?v=xxQ0mzZ8UvA" class="external-link">video tutorial</a>
 or:</p>
@@ -529,7 +529,7 @@ <h3 class="accordion-header" id="headingSpoiler2">
   MacOS
   </h3>
 </button>
-<div id="collapseSpoiler2" class="accordion-collapse collapse" aria-labelledby="headingSpoiler2" data-bs-parent="#accordionSpoiler2">
+<div id="collapseSpoiler2" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler2" aria-labelledby="headingSpoiler2">
 <div class="accordion-body">
 <p>Checkout the <a href="https://www.youtube.com/watch?v=TcSAln46u9U" class="external-link">video tutorial</a>
 or:</p>
@@ -551,7 +551,7 @@ <h3 class="accordion-header" id="headingSpoiler3">
   Linux
   </h3>
 </button>
-<div id="collapseSpoiler3" class="accordion-collapse collapse" aria-labelledby="headingSpoiler3" data-bs-parent="#accordionSpoiler3">
+<div id="collapseSpoiler3" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler3" aria-labelledby="headingSpoiler3">
 <div class="accordion-body">
 <p>Note that the following installation steps require you to work from
 the shell. If you run into any difficulties, please request help before
@@ -634,7 +634,7 @@ <h3 class="accordion-header" id="headingSpoiler4">
   Troubleshooting for Windows
   </h3>
 </button>
-<div id="collapseSpoiler4" class="accordion-collapse collapse" aria-labelledby="headingSpoiler4" data-bs-parent="#accordionSpoiler4">
+<div id="collapseSpoiler4" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler4" aria-labelledby="headingSpoiler4">
 <div class="accordion-body">
 <p>It is possible that Windows users will run into version conflicts. If
 you are on Windows and get errors running the command, you can try
@@ -658,7 +658,7 @@ <h3 class="accordion-header" id="headingSpoiler5">
   Troubleshooting for Macs with Apple silicon chip
   </h3>
 </button>
-<div id="collapseSpoiler5" class="accordion-collapse collapse" aria-labelledby="headingSpoiler5" data-bs-parent="#accordionSpoiler5">
+<div id="collapseSpoiler5" class="accordion-collapse collapse" data-bs-parent="#accordionSpoiler5" aria-labelledby="headingSpoiler5">
 <div class="accordion-body">
 <p>Newer Macs (from 2020 onwards) often have a different kind of chip,
 manufactured by Apple instead of Intel. This can lead to problems
diff --git a/md5sum.txt b/md5sum.txt
index 84f4d233..ecd250cd 100644
--- a/md5sum.txt
+++ b/md5sum.txt
@@ -8,7 +8,7 @@
 "episodes/1-introduction.Rmd" "70b9b736d80b1f243d82bfae89537b0e" "site/built/1-introduction.md" "2024-07-22"
 "episodes/2-keras.Rmd" "e6fbe6ed2753685642491bbbc53420c5" "site/built/2-keras.md" "2024-07-22"
 "episodes/3-monitor-the-model.Rmd" "d9a73639b67c9c4a149a28d1df067dd2" "site/built/3-monitor-the-model.md" "2024-07-22"
-"episodes/4-advanced-layer-types.Rmd" "b5ead8c6648ecfda05f079e7675d2167" "site/built/4-advanced-layer-types.md" "2024-07-22"
+"episodes/4-advanced-layer-types.Rmd" "71e9dac1dc04adf910d8b7a677bfdf2e" "site/built/4-advanced-layer-types.md" "2024-07-22"
 "episodes/5-transfer-learning.Rmd" "730f5f22ab20ff94fef20fb94f0207e9" "site/built/5-transfer-learning.md" "2024-07-22"
 "episodes/6-outlook.Rmd" "e1cc70a009451a63d39aa1fb4ffe7816" "site/built/6-outlook.md" "2024-07-22"
 "instructors/bonus-material.md" "d5b6aaee56986ab74e33bb95894cdc0e" "site/built/bonus-material.md" "2024-07-22"
diff --git a/pkgdown.yml b/pkgdown.yml
index 23f55f67..35c5fc3b 100644
--- a/pkgdown.yml
+++ b/pkgdown.yml
@@ -2,4 +2,4 @@ pandoc: 3.1.11
 pkgdown: 2.1.0
 pkgdown_sha: ~
 articles: {}
-last_built: 2024-07-22T09:53Z
+last_built: 2024-07-22T12:26Z