numpy.html

<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="utf-8">
    <title>Configurable runtimes</title>
    <meta name="theme-color" content="#0b0c0c">
    <meta name="viewport" content="width=device-width, initial-scale=1.0, viewport-fit=cover">
    <meta http-equiv="X-UA-Compatible" content="IE=edge">
    <link rel="stylesheet" href="static/gd.css">
  </head>
  <body>
    <py-config>
    {
      "interpreter": "micropython",
      "files": {
        "arrr.py": "/arrr.py"
      }
    }
    </py-config>
    <a href="#main-content" class="hide">Skip to main content</a>
    <a name="top"></a>
    <header>
    <nav role="navigation">
      <a href="/" style="text-decoration: none!important;";>
        <img alt="Anaconda" src="static/logo.svg" width="54" height="54" style="display:inline;">
        <img alt="Anaconda" src="static/anaconda.svg" height="54" style="display: inline; height: 42px;">
      </a>
      <ul>
        <li><a href="/">Home</a></li>
        <li><a href="https://github.com/pyscript/upys">Download</a></li>
      </ul>
    </nav>
  </header>
  <main id="main-content">
    <h2>Native Modules</h2>

    <p>There are numerous pure-Python modules that work with MicroPython.
    However, some of the most interesting involve C-based extensions, with the
    most well known being data science tools like
    <a href="https://numpy.org/">numpy</a> and
    <a href="https://scipy.org/">scipy</a>.</p>

    <p>These C-based extensions can be compiled to WASM, in exactly the same
    way that Python runtimes (written in C) are compiled to WASM.</p>
    
    <p>The version of MicroPython emebedded in the following REPL also
    includes a WASM compiled tool called
    <a href="https://micropython-ulab.readthedocs.io/en/latest/index.html">uLab</a> ("micro lab").
    This is a small reimplementation of the core functions, for effieciently
    working with numeric arrays found, in the numpy
    and scipy modules. It is clearly not a complete re-implementation, but
    serves to prove that such bespoke C-based extensions can be used with the
    MicroPython runtime.</p>

    <py-repl></py-repl>

    <p>Why not try:</p>

    <pre><code>&gt;&gt;&gt; import ulab
&gt;&gt;&gt; np = ulab.numpy
&gt;&gt;&gt; a = np.arange(6)
&gt;&gt;&gt; a
array([0, 1, 2, 3, 4, 5], dtype=int16)
&gt;&gt;&gt; b = a.reshape((2, 3))
&gt;&gt;&gt; b
array([[0, 1, 2],
       [3, 4, 5]], dtype=int16)
&gt;&gt;&gt; np.sqrt(b)
array([[0.0, 1.0, 1.414213562373095],
       [1.732050807568877, 2.0, 2.23606797749979]], dtype=float64)</code></pre>

    <a href="repl.html"><button>Previous: The REPL</button></a>
  </main>
  <footer>
    <main>
      <section>
        <ul>
          <li><a href="/">Home</a></li>
          <li><a href="https://github.com/pyscript/upys">Download</a></li>
        </ul>
        <article>
          <p>Made with ❤️  by folks at <a href="https://anaconda.com" rel="noopener">Anaconda Inc</a>.<br>
            💡Inspired by the great work at <a href="https://micropython.org/" rel="noopener">MicroPython</a>.<br><br>
        </article>
      </section>
    </main>
  </footer>
  </body>
  <script src="customtags.js"></script>
  <script src="pyscript.js" type="module"></script>
</html>