Re: [Scikit-learn-general] Which scikit-learn contributors share common interests?

On Sat, Sep 28, 2013 at 4:28 AM, Robert Layton wrote: > This is a great idea! > I wonder if my recent PR on licences skewed my result (in which I touched > nearly every file...). > I would try tf-idf. Mathieu -- October

Re: [Scikit-learn-general] [github] How can I track pull-requests for scikit-learn?

Hi Gilles, thanks for the help. It helps, but mlp doesn't track the pull request branch. I would like to be able, when in mlp, to just do: git pull mlp to fetch and merge all the changes in the remote pull request branch! Thanks again for the help! *Luca Cerone* Tel: +447585611951 Skype: luc

Re: [Scikit-learn-general] Which scikit-learn contributors share common interests?

2013/9/28 Mathieu Blondel : > I would try tf-idf. Or at least sublinear frequency weighting: 1 + log(f), implemented by TfidfTransformer(use_idf=False, sublinear_tf=True). But what do we consider the "documents" in this case? As for an example, I was actually considering making a standalone tool

Re: [Scikit-learn-general] error using CountVectorizer with python 2.6.1

This is an error on collections defaultdict in Python 2.6.1 -- October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced prof

[Scikit-learn-general] Possible ways of contributing.

Hi Guys, I'm a newbie here. I want to learn to contribute to scikit learn. I have an understanding of the machine learning concepts that scikit learn implements, so any guidelines would be appreciated! -- Tejas Nikumbh, Fourth Year Undergraduate, Electrical Engineering Department, IIT Bombay. --