Mirror/AI-on-the-edge-device-docs
1
0
Fork 0
mirror of https://github.com/jomjol/AI-on-the-edge-device-docs.git synced 2025-12-06 11:36:53 +03:00
Code Issues Packages Projects Releases Wiki Activity

Deployed ec107cb with MkDocs version: 1.6.1

Browse Source
This commit is contained in:
github-actions[bot]
2025-04-13 19:50:25 +00:00
parent c2dee3942d
commit 8b84093ddc
5 changed files with 46 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files

192
Choosing-the-Model/index.html
View File

@@ -267,42 +267,13 @@
dig-class11
</span>
</a>
<nav aria-label="dig-class11" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features">1.1.1.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#dig-class100-dig-cont">1.1.2
<a class="md-nav__link" href="#dig-class100-and-dig-cont">1.1.2
<span class="md-ellipsis">
dig-class100 / dig-cont
dig-class100 and dig-cont
</span>
</a>
<nav aria-label="dig-class100 / dig-cont" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features_1">1.1.2.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#dig-class100-vs-dig-cont">1.1.2.2
<span class="md-ellipsis">
dig-class100 vs. dig-cont
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -316,29 +287,11 @@
<nav aria-label="Analog pointer models" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#ana-class100-ana-cont">1.2.1
<a class="md-nav__link" href="#ana-class100-and-ana-cont">1.2.1
<span class="md-ellipsis">
ana-class100 / ana-cont
ana-class100 and ana-cont
</span>
</a>
<nav aria-label="ana-class100 / ana-cont" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features_2">1.2.1.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#ana-class100-vs-ana-cont">1.2.1.2
<span class="md-ellipsis">
ana-class100 vs. ana-cont
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -349,17 +302,6 @@
Different types of models (normal vs. quantized)
</span>
</a>
<nav aria-label="Different types of models (normal vs. quantized)" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#example">1.3.0.1
<span class="md-ellipsis">
Example:
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -783,42 +725,13 @@
dig-class11
</span>
</a>
<nav aria-label="dig-class11" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features">1.1.1.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#dig-class100-dig-cont">1.1.2
<a class="md-nav__link" href="#dig-class100-and-dig-cont">1.1.2
<span class="md-ellipsis">
dig-class100 / dig-cont
dig-class100 and dig-cont
</span>
</a>
<nav aria-label="dig-class100 / dig-cont" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features_1">1.1.2.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#dig-class100-vs-dig-cont">1.1.2.2
<span class="md-ellipsis">
dig-class100 vs. dig-cont
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -832,29 +745,11 @@
<nav aria-label="Analog pointer models" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#ana-class100-ana-cont">1.2.1
<a class="md-nav__link" href="#ana-class100-and-ana-cont">1.2.1
<span class="md-ellipsis">
ana-class100 / ana-cont
ana-class100 and ana-cont
</span>
</a>
<nav aria-label="ana-class100 / ana-cont" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#main-features_2">1.2.1.1
<span class="md-ellipsis">
Main features
</span>
</a>
</li>
<li class="md-nav__item">
<a class="md-nav__link" href="#ana-class100-vs-ana-cont">1.2.1.2
<span class="md-ellipsis">
ana-class100 vs. ana-cont
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -865,17 +760,6 @@
Different types of models (normal vs. quantized)
</span>
</a>
<nav aria-label="Different types of models (normal vs. quantized)" class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a class="md-nav__link" href="#example">1.3.0.1
<span class="md-ellipsis">
Example:
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
@@ -893,48 +777,62 @@
<p>This page tries to help you on which model to select.
For more technical/deeper explanations have a look on <a href="../Neural-Network-Types/">Neural-Network-Types</a>.</p>
<h2 id="digit-models"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1</span> Digit Models</h2>
<p>For digits on water meters, gas-meters or power meters you can select between two main types of models.</p>
<h3 id="dig-class11"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.1</span> dig-class11</h3>
<p>This model can recognize full digits. It was the first model version. All intermediate states shown a "N" for not a number. But in post process it uses older values to fill up the "N" values if possible.</p>
<p>For digits on water meters, gas-meters or power meters you can select between two main types of models:</p>
<ul>
<li><code>dig-class11</code></li>
<li><code>dig-class100</code> and <code>dig-cont</code></li>
</ul>
<h3 id="dig-class11"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.1</span> <code>dig-class11</code></h3>
<p>This model can recognize <strong>full digits</strong>. It was the first model version. All intermediate states shown a <code>N</code> for not a number. But in post process it uses older values to fill up the <code>N</code> values if possible.</p>
<p><img alt="" src="../img/dig-class11.png" style="width:300px"/></p>
<p>It's possibly a good fallback, if <code>dig-cont/dig-class100</code> results are not good.</p>
<h4 id="main-features"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.1.1</span> Main features</h4>
<p>It's possibly a good fallback, if <code>dig-cont</code> or <code>dig-class100</code> results are not good.</p>
<p>Main features:</p>
<ul>
<li>well suited for LCD digits</li>
<li>the ExtendedResolution option is not supported. (Only in conjunction with ana-class100 / ana-cont)</li>
<li>the ExtendedResolution option is not supported. (Only in conjunction with <code>ana-class100</code> or <code>ana-cont</code>)</li>
</ul>
<h3 id="dig-class100-dig-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.2</span> dig-class100 / dig-cont</h3>
<p>These models are used to get a continuous reading with intermediate states. To see what the models are doing, you can go to the Recognition page.</p>
<h3 id="dig-class100-and-dig-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.2</span> <code>dig-class100</code> and <code>dig-cont</code></h3>
<p>These models are used to get a <strong>continuous reading</strong> with intermediate states. To see what the models are doing, you can go to the Recognition page of your device.</p>
<p><img alt="" src="../img/dig-class100.png" style="width:300px"/></p>
<h4 id="main-features_1"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.2.1</span> Main features</h4>
<p>Main features:</p>
<ul>
<li>suitable for all digit displays.</li>
<li>Advantage over dig-class11 that results continue to be calculated in the transition between digits.</li>
<li>Advantage over <code>dig-class11</code> that results continue to be calculated in the transition between digits.</li>
<li>With the ExtendedResolution option, higher accuracy is possible by adding another digit.</li>
</ul>
<p>Look <a href="https://jomjol.github.io/neural-network-digital-counter-readout">here</a> for a list of digit images used for the training.</p>
<h4 id="dig-class100-vs-dig-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.1.2.2</span> dig-class100 vs. dig-cont</h4>
<p>The difference is in the internal processing. </p>
<p>The dig-class100 is a standard classification model. Each tenth step is an output. </p>
<p>dig-cont uses two outputs and arctangent to get the result. You see very complicated. </p>
<p>The difference between <code>dig-class100</code> and <code>dig-cont</code> is in the internal processing.
The <code>dig-class100</code> is a standard classification model. Each tenth step is an output.
<code>dig-cont</code> uses two outputs and arctangent to get the result. You see very complicated. </p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Try both models on your device and take the one that gives you the best results.</p>
</div>
<p>Look <a href="https://jomjol.github.io/neural-network-digital-counter-readout">here</a> for a list of digit images used for the training.</p>
<h2 id="analog-pointer-models"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.2</span> Analog pointer models</h2>
<h3 id="ana-class100-ana-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.2.1</span> ana-class100 / ana-cont</h3>
<p>For pointers on water meters use the analog models. You can only choose between ana-class100 and ana-cont. Both do mainly the same.</p>
<h3 id="ana-class100-and-ana-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.2.1</span> <code>ana-class100</code> and <code>ana-cont</code></h3>
<p>For pointers on water meters use the analog models:</p>
<p><img alt="" src="../img/ana-class100.png" style="width:250px"/></p>
<h4 id="main-features_2"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.2.1.1</span> Main features</h4>
<p>You can choose between two models:</p>
<ul>
<li><code>ana-class100</code> </li>
<li><code>ana-cont</code></li>
</ul>
<p>Both do mainly the same.</p>
<p>Main features:</p>
<ul>
<li>for all analogue pointers, especially for water meters.</li>
<li>With the ExtendedResolution option, higher accuracy is possible by adding another digit.</li>
</ul>
<p>Look <a href="https://jomjol.github.io/neural-network-analog-needle-readout/">here</a> for a list of pointer images used for the training</p>
<h4 id="ana-class100-vs-ana-cont"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.2.1.2</span> ana-class100 vs. ana-cont</h4>
<p>The difference is in the internal processing.</p>
<p>Again, the difference between <code>ana-class100</code> and <code>ana-cont</code> is in the internal processing.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Take the one that gives you the best results. Both models learn from the same data.</p>
</div>
<p>Look <a href="https://jomjol.github.io/neural-network-analog-needle-readout/">here</a> for a list of pointer images used for the training</p>
<h2 id="different-types-of-models-normal-vs-quantized"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.3</span> Different types of models (normal vs. quantized)</h2>
<p>The normally trained network is calculating with internal floating point numbers. The saving of floating point numbers naturally takes more space than an integer type. Often the increased accuracy is not needed. Therefore there is the option, to "quantize" a neural network. In this case the internal values are rescaled to integer values, which is called "quantization". The stored tflite files are usually much smaller and runs faster on the edgeAI-device.
Usually the models are distrusted therefore in both versions. They can be distinguished by a "-q" at the end of the logfile.</p>
<h4 id="example"><span class="enumerate-headings-plugin enumerate-heading-plugin">1.3.0.1</span> Example:</h4>
Usually the models are distrusted therefore in both versions. They can be distinguished by a <code>q</code> at the end of the logfile.</p>
<p>Example:</p>
<table>
<thead>
<tr>

BIN
img/ana-class100.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 71 KiB

After

Width:  |  Height:  |  Size: 49 KiB

BIN
img/dig-class100.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 88 KiB

After

Width:  |  Height:  |  Size: 50 KiB

BIN
img/dig-class11.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 89 KiB

After

Width:  |  Height:  |  Size: 48 KiB

2
search/search_index.json
View File

File diff suppressed because one or more lines are too long