index.html

---
layout: landing-page
---

<div class="maintenance">
    <span class="maintenance-badge">Important</span>
    Chainer is under the maintenance phase. <a href="https://chainer.org/announcement/2019/12/05/released-v7.html">Read the blog.</a>
</div>

<!-- One -->
<section id="one" class="wrapper style1">
    <div class="container">
        <div class="row">
            <header class="major">
                <h2>Bridge the gap between algorithms and implementations of deep learning</h2>
            </header>
            <div class="4u">
                <section class="special box">
                    <i class="icon fa-rocket major"></i>
                    <h3>Powerful</h3>
                    <p>
                        Chainer supports CUDA computation.
                        It only requires a few lines of code to leverage a GPU.
                        It also runs on multiple GPUs with little effort.
                    </p>
                </section>
            </div>
            <div class="4u">
                <section class="special box">
                    <i class="icon fa-puzzle-piece major"></i>
                    <h3>Flexible</h3>
                    <p>
                        Chainer supports various network architectures including feed-forward nets, convnets, recurrent nets and recursive nets. It also supports per-batch architectures.
                    </p>
                </section>
            </div>
            <div class="4u">
                <section class="special box">
                    <i class="icon fa-bolt major"></i>
                    <h3>Intuitive</h3>
                    <p>
                        Forward computation can include any control flow statements of Python without lacking the ability of backpropagation. It makes code intuitive and easy to debug.
                    </p>
                </section>
            </div>
        </div>
    </div>
</section>

<section id="two" class="style1">
    <div class="container">
		<div class="row double">
	  		<div id="quickstart" class="7u">
	    		<header class="major">
	      			<h2>Quick Start</h2>
	    		</header>
	    		<p>
	      			Install Chainer:<br>
	      			<pre><code>pip install chainer</code></pre>
	    		</p>
	    		<p>
	      			Run the MNIST example:<br>
	      			<pre><code>wget https://github.com/chainer/chainer/archive/v{{ site.chainer_version }}.tar.gz
tar xzf v{{ site.chainer_version }}.tar.gz
python chainer-{{ site.chainer_version }}/examples/mnist/train_mnist.py</code></pre>
                </p>
                <p>
                    Learn more from <a href="http://docs.chainer.org">the official documentation</a>.
                </p>
            </div>
        </div>
    </div>
</section>

<section id="two-2" class="style1">
    <div class="container">
        <header class="major">
            <h2>Extension Libraries</h2>
        </header>
        <div class="row extensions">
            <div class="4u">
                <div class="extension-item" id="chainerrl">
                    <a href="https://github.com/chainer/chainerrl">
                        <div class="extension-cover"></div>
                    </a>
                    <div class="extension-icon">
                        <img src="images/RL1-2.png" alt="ChainerRL" />
                    </div>
                    <div class="extension-content">
                        A library that implements various state-of-the-art deep reinforcement algorithms.
                    </div>
                </div>
            </div>
            <div class="4u">
                <div class="extension-item" id="chainercv">
                    <a href="https://github.com/chainer/chainercv">
                        <div class="extension-cover"></div>
                    </a>
                    <div class="extension-icon">
                        <img src="images/CV1-2.png" alt="ChainerCV" />
                    </div>
                    <div class="extension-content">
                        A collection of tools to train and run neural networks for computer vision tasks.
                    </div>
                </div>
            </div>
        </div>
    </div>
</section>

<section id="three" class="wrapper style1">
    <header class="major">
        <h2>Slides</h2>
    </header>
    <div class="container">
        <strong>
            <a href="//www.slideshare.net/pfi/introduction-to-chainer-11-may2018-96768990" title="Introduction to Chainer 11 may,2018" target="_blank">Introduction to Chainer</a>
        </strong>
    </div>
</section>

<section id="parters" class="wrapper style1">
    <header class="major">
        <h2>Companies supporting Chainer</h2>
    </header>
    <div class="container small">
        <section class="special">
            <ul class="actions fit logo">
                <li><img src="images/ibm.png" alt="IBM" class="logo"/></li>
                <li><img src="images/intel.png" alt="Intel" class="logo"/></li>
                <li><img src="images/microsoft.png" alt="Microsoft" class="logo"/></li>
                <li><img src="images/nvidia.png" alt="NVIDIA" class="logo"/></li>
                <li><img src="images/aws.png" alt="AWS" class="logo"/></li>
            </ul>
        </section>
    </div>
</section>