Skip to content

Commit

Permalink
deploy: 4bc38be
Browse files Browse the repository at this point in the history
  • Loading branch information
@JulioContrerasH
JulioContrerasH committed Dec 2, 2024
1 parent 17f8823 commit a5badb3
Show file tree
Hide file tree
Showing 3 changed files with 54 additions and 24 deletions.
76 changes: 53 additions & 23 deletions index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -415,6 +415,33 @@
Download Sentinel-2 L2A cube
</a>

</li>

<li class="md-nav__item">
<a href="#prepare-the-data-cpu-and-gpu-usage" class="md-nav__link">
Prepare the data (CPU and GPU usage)
</a>

<nav class="md-nav" aria-label="Prepare the data (CPU and GPU usage)">
<ul class="md-nav__list">

<li class="md-nav__item">
<a href="#default-model-setup" class="md-nav__link">
Default model setup
</a>

</li>

<li class="md-nav__item">
<a href="#plot-explanation" class="md-nav__link">
Plot explanation
</a>

</li>

</ul>
</nav>

</li>

<li class="md-nav__item">
Expand Down Expand Up @@ -567,11 +594,13 @@ <h2 id="installation"><strong>Installation</strong> ⚙️</h2>
<p>Install the latest version from PyPI:</p>
<div class="highlight"><pre><span></span><code>pip<span class="w"> </span>install<span class="w"> </span>supers2
</code></pre></div>
<p>From GitHub:</p>
<div class="highlight"><pre><span></span><code>pip<span class="w"> </span>install<span class="w"> </span>git+https://github.com/IPL-UV/supers2.git
</code></pre></div>
<h2 id="how-to-use"><strong>How to use</strong> 🛠️</h2>
<h3 id="load-libraries"><strong>Load libraries</strong></h3>
<div class="highlight"><pre><span></span><code><span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="kn">import</span> <span class="nn">supers2</span>
<span class="kn">import</span> <span class="nn">torch</span>
<span class="kn">import</span> <span class="nn">cubo</span>

Expand All @@ -589,36 +618,37 @@ <h3 id="download-sentinel-2-l2a-cube"><strong>Download Sentinel-2 L2A cube</stro
<span class="n">edge_size</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
<span class="n">resolution</span><span class="o">=</span><span class="mi">10</span>
<span class="p">)</span>

<span class="c1">### **Prepare the data (CPU and GPU usage)**</span>

<span class="n">When</span> <span class="n">converting</span> <span class="n">the</span> <span class="n">NumPy</span> <span class="n">array</span> <span class="n">to</span> <span class="n">a</span> <span class="n">PyTorch</span> <span class="n">tensor</span><span class="p">,</span> <span class="n">the</span> <span class="n">use</span> <span class="n">of</span> <span class="err">`</span><span class="n">cuda</span><span class="p">()</span><span class="err">`</span> <span class="ow">is</span> <span class="n">optional</span> <span class="ow">and</span> <span class="n">depends</span> <span class="n">on</span> <span class="n">whether</span> <span class="n">the</span> <span class="n">user</span> <span class="n">has</span> <span class="n">access</span> <span class="n">to</span> <span class="n">a</span> <span class="n">GPU</span><span class="o">.</span> <span class="n">Below</span> <span class="ow">is</span> <span class="n">the</span> <span class="n">explanation</span> <span class="k">for</span> <span class="n">both</span> <span class="n">cases</span><span class="p">:</span>

<span class="o">-</span> <span class="o">**</span><span class="n">GPU</span><span class="p">:</span><span class="o">**</span> <span class="n">If</span> <span class="n">a</span> <span class="n">GPU</span> <span class="ow">is</span> <span class="n">available</span> <span class="ow">and</span> <span class="n">CUDA</span> <span class="ow">is</span> <span class="n">installed</span><span class="p">,</span> <span class="n">you</span> <span class="n">can</span> <span class="n">transfer</span> <span class="n">the</span> <span class="n">tensor</span> <span class="n">to</span> <span class="n">the</span> <span class="n">GPU</span> <span class="n">using</span> <span class="err">`</span><span class="o">.</span><span class="n">cuda</span><span class="p">()</span><span class="err">`</span><span class="o">.</span> <span class="n">This</span> <span class="n">improves</span> <span class="n">the</span> <span class="n">processing</span> <span class="n">speed</span><span class="p">,</span> <span class="n">especially</span> <span class="k">for</span> <span class="n">large</span> <span class="n">datasets</span> <span class="ow">or</span> <span class="n">deep</span> <span class="n">learning</span> <span class="n">models</span><span class="o">.</span>

<span class="o">-</span> <span class="o">**</span><span class="n">CPU</span><span class="p">:</span><span class="o">**</span> <span class="n">If</span> <span class="n">no</span> <span class="n">GPU</span> <span class="ow">is</span> <span class="n">available</span><span class="p">,</span> <span class="n">the</span> <span class="n">tensor</span> <span class="n">will</span> <span class="n">be</span> <span class="n">processed</span> <span class="n">on</span> <span class="n">the</span> <span class="n">CPU</span><span class="p">,</span> <span class="n">which</span> <span class="ow">is</span> <span class="n">the</span> <span class="n">default</span> <span class="n">behavior</span> <span class="ow">in</span> <span class="n">PyTorch</span><span class="o">.</span> <span class="n">In</span> <span class="n">this</span> <span class="n">case</span><span class="p">,</span> <span class="n">simply</span> <span class="n">omit</span> <span class="n">the</span> <span class="err">`</span><span class="o">.</span><span class="n">cuda</span><span class="p">()</span><span class="err">`</span> <span class="n">call</span><span class="o">.</span>

<span class="n">Here</span><span class="err">’</span><span class="n">s</span> <span class="n">how</span> <span class="n">you</span> <span class="n">can</span> <span class="n">handle</span> <span class="n">both</span> <span class="n">scenarios</span> <span class="n">dynamically</span><span class="p">:</span>

<span class="err">```</span><span class="n">python</span>
<span class="c1"># Convert the data array to NumPy and scale</span>
<span class="n">original_s2_numpy</span> <span class="o">=</span> <span class="p">(</span><span class="n">da</span><span class="p">[</span><span class="mi">11</span><span class="p">]</span><span class="o">.</span><span class="n">compute</span><span class="p">()</span><span class="o">.</span><span class="n">to_numpy</span><span class="p">()</span> <span class="o">/</span> <span class="mi">10_000</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="s2">&quot;float32&quot;</span><span class="p">)</span>

<span class="c1"># Check if CUDA is available, use GPU if possible, otherwise fallback to CPU</span>
</code></pre></div>
<h3 id="prepare-the-data-cpu-and-gpu-usage"><strong>Prepare the data (CPU and GPU usage)</strong></h3>
<p>When converting the NumPy array to a PyTorch tensor, the use of <code>cuda()</code> is optional and depends on whether the user has access to a GPU. Below is the explanation for both cases:</p>
<ul>
<li><strong>GPU:</strong> If a GPU is available and CUDA is installed, you can transfer the tensor to the GPU using <code>.cuda()</code>. This improves the processing speed, especially for large datasets or deep learning models.</li>
</ul>
<ul>
<li><strong>CPU:</strong> If no GPU is available, the tensor will be processed on the CPU, which is the default behavior in PyTorch. In this case, simply omit the <code>.cuda()</code> call.</li>
</ul>
<p>Here’s how you can handle both scenarios dynamically:</p>
<p><div class="highlight"><pre><span></span><code><span class="c1"># Check if CUDA is available, use GPU if possible</span>
<span class="n">device</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">&quot;cuda&quot;</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span><span class="p">()</span> <span class="k">else</span> <span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
</code></pre></div>
Converting data to a PyTorch tensor enables efficient computation, especially on GPUs, and ensures compatibility with the neural network. Scaling the data standardizes pixel values for better model performance.</p>
<div class="highlight"><pre><span></span><code><span class="c1"># Convert the data array to NumPy and scale</span>
<span class="n">original_s2_numpy</span> <span class="o">=</span> <span class="p">(</span><span class="n">da</span><span class="p">[</span><span class="mi">11</span><span class="p">]</span><span class="o">.</span><span class="n">compute</span><span class="p">()</span><span class="o">.</span><span class="n">to_numpy</span><span class="p">()</span> <span class="o">/</span> <span class="mi">10_000</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="s2">&quot;float32&quot;</span><span class="p">)</span>

<span class="c1"># Create the tensor and move it to the appropriate device (CPU or GPU)</span>
<span class="n">X</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">original_s2_numpy</span><span class="p">)</span><span class="o">.</span><span class="n">float</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>

<span class="c1"># Set up the model to enhance the spatial resolution</span>
</code></pre></div>
<h4 id="default-model-setup"><strong>Default model setup</strong></h4>
<p>The default model is pre-trained for 2.5m resolution but supports 5m and 10m resolutions via the <code>resolution</code> parameter. It uses lightweight CNN architectures for super-resolution and fusion (<code>sr_model_snippet</code>, <code>fusionx2_model_snippet</code>, <code>fusionx4_model_snippet</code>). Models run on CPU or GPU, configurable via <code>device</code>.</p>
<div class="highlight"><pre><span></span><code><span class="c1"># Set up the model to enhance the spatial resolution</span>
<span class="n">models</span> <span class="o">=</span> <span class="n">supers2</span><span class="o">.</span><span class="n">setmodel</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>

<span class="c1"># Apply spatial resolution enhancement</span>
<span class="n">superX</span> <span class="o">=</span> <span class="n">supers2</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">models</span><span class="o">=</span><span class="n">models</span><span class="p">,</span> <span class="n">resolution</span><span class="o">=</span><span class="s2">&quot;2.5m&quot;</span><span class="p">)</span>

<span class="c1"># Visualize the results</span>
<span class="c1"># Plot the original and enhanced-resolution images</span>
<span class="n">fig</span><span class="p">,</span> <span class="n">ax</span> <span class="o">=</span> <span class="n">plt</span><span class="o">.</span><span class="n">subplots</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">figsize</span><span class="o">=</span><span class="p">(</span><span class="mi">10</span><span class="p">,</span> <span class="mi">5</span><span class="p">))</span>
</code></pre></div>
<h4 id="plot-explanation"><strong>Plot explanation</strong></h4>
<p>The first plot shows the original Sentinel-2 RGB image (10m resolution). The second plot displays the enhanced version with finer spatial details (2.5m resolution) using a lightweight CNN.</p>
<div class="highlight"><pre><span></span><code><span class="n">fig</span><span class="p">,</span> <span class="n">ax</span> <span class="o">=</span> <span class="n">plt</span><span class="o">.</span><span class="n">subplots</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">figsize</span><span class="o">=</span><span class="p">(</span><span class="mi">10</span><span class="p">,</span> <span class="mi">5</span><span class="p">))</span>
<span class="n">ax</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">imshow</span><span class="p">(</span><span class="n">X</span><span class="p">[[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">*</span><span class="mi">4</span><span class="p">)</span>
<span class="n">ax</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">set_title</span><span class="p">(</span><span class="s2">&quot;Original S2&quot;</span><span class="p">)</span>
<span class="n">ax</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">imshow</span><span class="p">(</span><span class="n">superX</span><span class="p">[[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">*</span><span class="mi">4</span><span class="p">)</span>
Expand Down
Loading

0 comments on commit a5badb3

Please sign in to comment.