Re: [Scikit-learn-general] Loss documented in the GradientBoostedTrees

2012-09-03 Thread Gael Varoquaux
Hi Peter, Sorry for missing the PR: it was high on my list of things to review, but I hadn't gotten to it. Thanks, Gael -- Live Security Virtual Conference Exclusive live event will cover all the ways today's security a

Re: [Scikit-learn-general] Loss documented in the GradientBoostedTrees

2012-09-03 Thread Peter Prettenhofer
Hi Gael, the following PR contains better docstrings as well as some API consistency issues: https://github.com/scikit-learn/scikit-learn/pull/1088 thanks, Peter 2012/9/3 Gael Varoquaux : > Hi tree growers, > > I was looking at the Gradient Boosted Tree docstring, and it seems to me > that the

[Scikit-learn-general] Loss documented in the GradientBoostedTrees

2012-09-03 Thread Gael Varoquaux
Hi tree growers, I was looking at the Gradient Boosted Tree docstring, and it seems to me that the docstring for the 'loss' argument is incomplete: it only lists 'ls' and 'lad' as possible losses, and with a quick glance at the code, I have the feeling that there are more choices for the loss. Ga