DataTaunew | comments | leaders | submitlogin
A Visual Introduction to Machine Learning (r2d3.us)
48 points by ajinkyakale 3189 days ago | 5 comments


1 point by rahmaniacc 3189 days ago | link

does anyone what kind of library was used to create the visualizations? They look pretty good. this is a really good way to present findings as a data scientist.

-----

4 points by skadamat 3189 days ago | link

You can do all of these using D3, albeit it takes a while. It's a really neat demonstration but it's a little hard to understand / interact with to be completely honest.

If you're using the ipython notebook workflow, I would check out bokeh or seaborn for publishing web-ready graphics / interactions. Vispy has some promise although its super early.

-----

1 point by ajinkyakale 3188 days ago | link

i was introduced to bokeh in a recent data science conference and it looked pretty cool, esp the interactivity and the way you can drill down into the granularity of the plot changing the scales.

-----

1 point by ajinkyakale 3188 days ago | link

http://www.datatau.com/item?id=8767

-----

3 points by seabass 3188 days ago | link

from the source code of the page:

  <!-- jQuery (necessary for Bootstrap's JavaScript plugins) -->
  <script src="/static/app/_bower_components/jquery/dist/jquery.min.js"></script>

  <!-- Include all compiled plugins (below), or include individual files as needed -->
  <script src="/static/app/_bower_components/bootstrap/dist/js/bootstrap.min.js"></script>
  <script src="/static/app/_bower_components/underscore/underscore.js"></script>
  <script src="/static/app/_bower_components/d3/d3.min.js"></script>
...

  <script src="/static/app/_bower_components/backbone/backbone.js"></script>
...

  <script src="/static/pages/decision-trees-part-1/rAF-polyfill.js"></script>
So, the JavaScript libraries used were jQuery, Bootstrap, Underscore, Backbone, Polyfill, and D3.

-----




RSS | Announcements