[%
  title='MyMediaLite: GUI Demo'
  prefix='../'
%]

[% INCLUDE header %]

<h1><a href="index.html">[% title %]</a></h1>

[% INCLUDE menu %]

[% INCLUDE infobox %]

<div id="content">
  <p style="text-align:center">
    <img src="[% prefix %]/img/movie_demo.png"/>
  </p>

  <h2>Setup</h2>
  
  <p>
  Download the source package of MyMediaLite, and open the MyMediaLite solution file <code>src/MyMediaLite.sln</code> in MonoDevelop.
  Hit Ctrl-F8 to build the complete project.
  
  In the MyMediaLite root directory, enter
  <pre>
    make download-imdb
    make download-movielens
  </pre>
  This is what you would do on Linux and OS X.
  </p>

  <p>
  On Windows, you have to perform some steps manually (<a href="https://github.com/zenogantner/MyMediaLite/issues/126">sorry for that, we intend to change this in the future</a>):
  <ol>
    <li>Download <a href="http://www.grouplens.org/system/files/ml-1m.zip">http://www.grouplens.org/system/files/ml-1m.zip</a> and
        <a href="ftp://ftp.fu-berlin.de/pub/misc/movies/database/german-aka-titles.list.gz">ftp://ftp.fu-berlin.de/pub/misc/movies/database/german-aka-titles.list.gz</a></li>
    <li>Unzip the two files with whatever archiving program you have (e.g. WinZip). The IMDB stuff should go into
        <tt>data\imdb</tt>, the MovieLens stuff into <tt>data\ml-1m</tt>.</li>
    <li>then run <tt>perl scripts/import_dataset.pl --separator=:: data/ml-1m/ratings.dat > data/ml-1m/ratings.txt</tt> on the command line</li>
  </ol>
  </p>

  <p>
  To train the initial model, execute
  <pre>
    mkdir data/models
    rating_prediction --training-file=data/ml-1m/ratings.txt --recommender=BiasedMatrixFactorization  --recommender-options="num_iter=60 bias_reg=0.001 regularization=0.045" --save-model=data/models/ml-1m-bmf.model
  </pre>

  </p>
  Then you can run the GUI demo e.g. by opening <code>src/MovieDemo/Program/Main.cs</code> and hitting Ctrl-F5 in MonoDevelop.
</div>

[% INCLUDE footer %]