Comments on: Decision Tree Learning in Ruby http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/ A goal is a dream with a deadline. Sun, 14 Mar 2010 08:01:10 +0000 http://wordpress.org/?v=2.7.1 hourly 1 By: Ilya Grigorik http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-213856 Ilya Grigorik Thu, 17 Sep 2009 02:49:49 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-213856 Jay, thanks for the heads up. No idea how it got in there, but I pushed out a fix for that. Grab the latest <a href="http://github.com/igrigorik/decisiontree" rel="nofollow">gem from GitHub</a>. Jay, thanks for the heads up. No idea how it got in there, but I pushed out a fix for that. Grab the latest gem from GitHub.

]]> By: Jay Packard http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-213758 Jay Packard Wed, 16 Sep 2009 20:39:58 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-213758 I'm trying out this library and wondering why dec_tree.predict() returns an array with two elements? As a result I get output: Predicted: 11 ... True decision: 0 I’m trying out this library and wondering why dec_tree.predict() returns an array with two elements? As a result I get output:
Predicted: 11 … True decision: 0

]]> By: Collaborative Filtering with Ensembles - igvita.com http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-210809 Collaborative Filtering with Ensembles - igvita.com Tue, 01 Sep 2009 18:08:25 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-210809 [...] techniques is to make each predictor as unique as possible: use a different learning algorithm (decision trees, svm, svd), or use a different feature (random subspace method). Then, once you have many [...] [...] techniques is to make each predictor as unique as possible: use a different learning algorithm (decision trees, svm, svd), or use a different feature (random subspace method). Then, once you have many [...]

]]> By: Ilya Grigorik http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-167275 Ilya Grigorik Sun, 08 Feb 2009 16:39:53 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-167275 Amir, I'm sure you can find alternative implementations, but the question remains: why? The formulas are hidden and should be generic for any dataset, given that the dataset can be classified using a decision tree. Amir, I’m sure you can find alternative implementations, but the question remains: why? The formulas are hidden and should be generic for any dataset, given that the dataset can be classified using a decision tree.

]]> By: amir http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-166603 amir Fri, 06 Feb 2009 10:04:42 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-166603 Is there any way other than these entropy formulas to build this decision tree? Is there any way other than these entropy formulas to build this decision tree?

]]> By: Wouter http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-158775 Wouter Fri, 09 Jan 2009 10:48:24 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-158775 It worked. I dragged the 'graph' folder into my project and now I can use Graphviz and Ruby together. Thank you very much! It worked. I dragged the ‘graph’ folder into my project and now I can use Graphviz and Ruby together. Thank you very much!

]]> By: Ilya Grigorik http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-158634 Ilya Grigorik Fri, 09 Jan 2009 03:57:22 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-158634 Wouter, I think you may have a different graphviz gem installed. The one my gem references is: http://rockit.sourceforge.net/subprojects/graphr/ I haven't looked at other gems, but I'll see if I can migrate the code to a more common library. Wouter, I think you may have a different graphviz gem installed. The one my gem references is: http://rockit.sourceforge.net/subprojects/graphr/

I haven’t looked at other gems, but I’ll see if I can migrate the code to a more common library.

]]> By: Wouter http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-157839 Wouter Wed, 07 Jan 2009 08:31:15 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-157839 Great Article! Since I read your article I want to use Graphviz in a ruby application. Still i can't get graphviz to work. require 'graphviz' results in: no such file to load -- graphviz (LoadError) I've installed Eclipse (with ruby), GraphViz, libgraphviz4 and libgv-ruby. (ubuntu 8.10) How can i make the combination of Ruby & Graphviz in Eclipse work? Great Article! Since I read your article I want to use Graphviz in a ruby application. Still i can’t get graphviz to work.
require ‘graphviz’ results in: no such file to load — graphviz (LoadError)
I’ve installed Eclipse (with ruby), GraphViz, libgraphviz4 and libgv-ruby. (ubuntu 8.10)
How can i make the combination of Ruby & Graphviz in Eclipse work?

]]> By: Ilya Grigorik http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-102114 Ilya Grigorik Tue, 18 Mar 2008 12:37:34 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-102114 Thanks Raf! Amir, Pireland, <a href="http://www.igvita.com/2008/01/07/support-vector-machines-svm-in-ruby/" rel="nofollow">svm article</a>, as Raf suggested. Thanks Raf!

Amir, Pireland, svm article, as Raf suggested.

]]> By: Raf http://www.igvita.com/2007/04/16/decision-tree-learning-in-ruby/comment-page-1/#comment-101901 Raf Mon, 10 Mar 2008 16:53:55 +0000 http://www.igvita.com/blog/2007/04/16/decision-tree-learning-in-ruby/#comment-101901 Privet, nu ti postaralsa! Great article! To Amir and Pireland, you can try SVM as well for your purposes. Privet, nu ti postaralsa!

Great article!

To Amir and Pireland, you can try SVM as well for your purposes.

]]>