Skip to content

Commit

Permalink
Separate sites for different materials
Browse files Browse the repository at this point in the history
  • Loading branch information
@ondrejmokry
ondrejmokry committed Sep 4, 2024
1 parent cc02ec8 commit 8381a1b
Show file tree
Hide file tree
Showing 6 changed files with 720 additions and 198 deletions.
1 change: 1 addition & 0 deletions docs/_layouts/default.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ <h2 class="project-tagline">{{ page.description | default: site.description | de
<a href="{{ site.github.zip_url }}" class="btn">Download .zip</a>
<a href="{{ site.github.tar_url }}" class="btn">Download .tar.gz</a>
{% endif %}
<a href="https://arxiv.org/abs/2403.04433" class="btn">Read the paper</a>
</header>

<main id="content" class="main-content" role="main">
Expand Down
90 changes: 90 additions & 0 deletions docs/assets/css/style.scss
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,4 +4,94 @@
@import "{{ site.theme }}";
table {
margin: 15px;
}
.pUnderAlpha{
background-color: #aaffaa;
color: #000000;
}
.pAboveAlpha{
background-color: #ffaaaa;
color: #000000;
}

.cudlik {
display: inline-block;
margin-bottom: 1rem;
color: rgba(0, 0, 0, 0.7);
background-color: rgba(0, 0, 0, 0.08);
border-color: rgba(0, 0, 0, 0.2);
border-style: solid;
border-width: 1px;
border-radius: 0.3rem;
transition: color 0.2s, background-color 0.2s, border-color 0.2s;

&:hover {
color: rgba(0, 0, 0, 0.8);
text-decoration: none;
background-color: rgba(0, 0, 0, 0.2);
border-color: rgba(0, 0, 0, 0.3);
}

+ .cudlik {
margin-left: 1rem;
}

@include large {
padding: 0.75rem 1rem;
}

@include medium {
padding: 0.6rem 0.9rem;
font-size: 0.9rem;
}

@include small {
display: block;
width: 100%;
padding: 0.75rem;
font-size: 0.9rem;

+ .cudlik {
margin-top: 1rem;
margin-left: 0;
}
}
}

.playButton {
display: inline-block;
margin-bottom: 0rem;
color: rgba(0, 0, 0, 0.5);
background-color: rgba(255, 255, 255, 0.1);
border-color: rgba(0, 0, 0, 0.2);
border-style: solid;
border-width: 1px;
border-radius: 0.3rem;
transition: color 0.2s, background-color 0.2s, border-color 0.2s;

&:hover {
color: rgba(0, 0, 0, 0.8);
text-decoration: none;
background-color: rgba(0, 0, 0, 0.2);
border-color: rgba(0, 0, 0, 0.3);
}
}

.pauseButton {
display: inline-block;
margin-bottom: 0rem;
color: rgba(0, 64, 0, 0.5);
background-color: rgba(0, 255, 0, 0.1);
border-color: rgba(0, 0, 0, 0.2);
border-style: solid;
border-width: 1px;
border-radius: 0.3rem;
transition: color 0.2s, background-color 0.2s, border-color 0.2s;

&:hover {
color: rgba(0, 64, 0, 0.8);
text-decoration: none;
background-color: rgba(0, 255, 0, 0.2);
border-color: rgba(0, 0, 0, 0.3);
}
}
204 changes: 6 additions & 198 deletions docs/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -8,203 +8,11 @@
</blockquote>
<p>The preprint is available at <a href="http://arxiv.org/abs/2403.04433">arXiv</a>.</p>

<h2 id="statistical-testing-of-lpc-versus-burg-algorithm">Statistical testing of LPC versus Burg algorithm</h2>
<a href="./pvals.html" class="cudlik">Statistical testing of LPC versus Burg algorithm</a>
<br>

<p>For the assessment of statistical significance of the effect of the AR model estimator (LPC versus Burg algorithm), the <a href="https://www.mathworks.com/help/stats/signrank.html">Wilcoxon signed rank test</a> was employed using the following hypotheses:</p>
<ul>
<li>H0: Burg algorithm and LPC lead to results with the same median,</li>
<li>HA: Burg algorithm leads to results with <em>higher</em> median.</li>
</ul>
<p>This was performed separately for each inpainting method, AR model order, and evaluation metric (SDR, PEMO-Q ODG).</p>
<p>The p-value displayed in the tables below indicate the rejection of the null hypothesis, i.e., p-value &lt; 0.05 implies that the data feature enough evidence to reject the equality of the medians in favor of the alternative hypothesis HA (at the significance level of 5%). On the other hand, p-value &gt; 0.05 means the test is inconclusive.</p>
<table>
<thead>
<tr>
<th>evaluation by SDR</th>
<th>256</th>
<th>512</th>
<th>1024</th>
<th>2048</th>
<th>3072</th>
</tr>
</thead>
<tbody>
<tr>
<td>extrapolation-based</td>
<td><strong>1.06e-21</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>3.58e-20</strong></td>
</tr>
<tr>
<td>Janssen, gap-wise</td>
<td><strong>2.16e-09</strong></td>
<td><strong>3.89e-13</strong></td>
<td><strong>5.54e-10</strong></td>
<td><strong>0.03</strong></td>
<td><em>0.97</em></td>
</tr>
<tr>
<td>Janssen, Hann window</td>
<td><strong>7.82e-05</strong></td>
<td><strong>6.19e-4</strong></td>
<td><em>0.90</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
</tr>
<tr>
<td>Janssen, rect. window</td>
<td><strong>6.69e-11</strong></td>
<td><strong>4.67e-06</strong></td>
<td><em>0.39</em></td>
<td><em>0.99</em></td>
<td><em>1.00</em></td>
</tr>
</tbody>
</table>
<table>
<thead>
<tr>
<th>evaluation by ODG</th>
<th>256</th>
<th>512</th>
<th>1024</th>
<th>2048</th>
<th>3072</th>
</tr>
</thead>
<tbody>
<tr>
<td>extrapolation-based</td>
<td><strong>6.35e-22</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>2.12e-22</strong></td>
<td><strong>3.57e-18</strong></td>
</tr>
<tr>
<td>Janssen, gap-wise</td>
<td><strong>8.94e-13</strong></td>
<td><strong>1.25e-18</strong></td>
<td><strong>1.86e-20</strong></td>
<td><strong>3.42e-16</strong></td>
<td><strong>5.10e-06</strong></td>
</tr>
<tr>
<td>Janssen, Hann window</td>
<td><em>0.20</em></td>
<td><em>0.09</em></td>
<td><em>0.06</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
</tr>
<tr>
<td>Janssen, rect. window</td>
<td><strong>2.11e-22</strong></td>
<td><strong>1.84e-18</strong></td>
<td><strong>0.02</strong></td>
<td><em>0.99</em></td>
<td><em>0.99</em></td>
</tr>
</tbody>
</table>
<a href="./plots.html" class="cudlik">Supplementary plots</a>
<br>

<p>To assess the superiority of LPC in some cases, we performed the same test using a different altarnative hypothesis:</p>
<ul>
<li>H0: Burg algorithm and LPC lead to results with the same median,</li>
<li>HA: Burg algorithm leads to results with <em>lower</em> median.</li>
</ul>
<p>In this case, p-values &lt; 0.05 imply statistical significance of the LPC surpassing the Burg algorithm (at the significance level of 5%).</p>
<table>
<thead>
<tr>
<th>evaluation by SDR</th>
<th>256</th>
<th>512</th>
<th>1024</th>
<th>2048</th>
<th>3072</th>
</tr>
</thead>
<tbody>
<tr>
<td>extrapolation-based</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
</tr>
<tr>
<td>Janssen, gap-wise</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>0.97</em></td>
<td><strong>0.03</strong></td>
</tr>
<tr>
<td>Janssen, Hann window</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>0.10</em></td>
<td><strong>4.06e-18</strong></td>
<td><strong>3.22e-14</strong></td>
</tr>
<tr>
<td>Janssen, rect. window</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>0.61</em></td>
<td><strong>0.01</strong></td>
<td><strong>7.01e-4</strong></td>
</tr>
</tbody>
</table>
<table>
<thead>
<tr>
<th>evaluation by ODG</th>
<th>256</th>
<th>512</th>
<th>1024</th>
<th>2048</th>
<th>3072</th>
</tr>
</thead>
<tbody>
<tr>
<td>extrapolation-based</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
</tr>
<tr>
<td>Janssen, gap-wise</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
</tr>
<tr>
<td>Janssen, Hann window</td>
<td><em>0.80</em></td>
<td><em>0.91</em></td>
<td><em>0.94</em></td>
<td><strong>8.78e-4</strong></td>
<td><strong>4.67e-3</strong></td>
</tr>
<tr>
<td>Janssen, rect. window</td>
<td><em>1.00</em></td>
<td><em>1.00</em></td>
<td><em>0.98</em></td>
<td><strong>0.01</strong></td>
<td><strong>8.69e-3</strong></td>
</tr>
</tbody>
</table>
<a href="./signals.html" class="cudlik">Audio examples</a>
<br>
9 changes: 9 additions & 0 deletions docs/plots.html
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
---
layout: default
---

<h1>Supplementary plots</h1>

<p>
<a href="./index.html" class="cudlik">Back to main page</a>
</p>
Loading

0 comments on commit 8381a1b

Please sign in to comment.