Re: [Scikit-learn-general] User Survey

Hi Ian. No, we didn't really include questions in this regard. It seemed to be quite a bit of text mining but other than that, I don't think there were any particular areas mentioned. Cheers, Andy On 07/15/2013 01:53 PM, Ian Ozsvald wrote: > Hi Andreas. I know this thread goes back a little way.

Re: [Scikit-learn-general] User Survey

Hi Denzil. The survey is over ;) Cheers, Andy On 07/15/2013 05:03 PM, Denzil Correa wrote: Hi Andreas, What is the link to the survey? On Fri, Feb 1, 2013 at 11:37 PM, Andreas Mueller mailto:amuel...@ais.uni-bonn.de>> wrote: Hey everybody. A quick status update on the user survey:

Re: [Scikit-learn-general] Free rackspace cloud account for scikit-learn

I received the confirmation by Jesse that our account is activated. If you need the credentials for scikit-learn related OSS dev, please send me a private email and I will send them to you. Currently there is no limitation on the number of machines we can provision, so it can be useful for experim

[Scikit-learn-general] Jenkins build on Python 2.6 / NumPy 1.3 disabled

Hi all, The recently merged new version of the cython code for the KD Tree / Ball Tree neighbor search requires the numpy buffer interface which was introduced in NumPy 1.5. Hence the test do no longer pass on NumPy 1.3 https://jenkins.shiningpanda-ci.com/scikit-learn/job/python-2.6-numpy-1.3.0-

Re: [Scikit-learn-general] User Survey

Hi Andreas, What is the link to the survey? On Fri, Feb 1, 2013 at 11:37 PM, Andreas Mueller wrote: > Hey everybody. > A quick status update on the user survey: > > I think the number of entries to the user survey is converging (no new > entries today). > I think it would be ok to close it afte

Re: [Scikit-learn-general] My talk from pycon AU is up

2013/7/14 Harshal : > Thanks for that awesome talk. I have recently started to explore > scikit-learn and find it pretty amazing. > > In that talk you mention that there are many request for having neural > network and deep learning support. I'm eagerly waiting to get my hands dirty > on it once av

Re: [Scikit-learn-general] Python 3 port

2013/7/9 Arnaud Joly : > Is the py3k branch https://github.com/scikit-learn/scikit-learn/tree/py3k > still useful? I had completely forgotten about that one. I picked off the one commit that was still useful and pushed that to master, then deleted it. -- Lars Buitinck Scientific programmer, ILP

Re: [Scikit-learn-general] My talk from pycon AU is up

Thanks for that awesome talk. I have recently started to explore scikit-learn and find it pretty amazing. In that talk you mention that there are many request for having neural network and deep learning support. I'm eagerly waiting to get my hands dirty on it once available. Thanks, Harshal On Ju

Re: [Scikit-learn-general] Python 3 port

Is the py3k branch https://github.com/scikit-learn/scikit-learn/tree/py3k still useful? Arnaud On 09 Jul 2013, at 16:26, Olivier Grisel wrote: > The REAMDE-Py3k.rst was not reflecting the current situation. I just > updated it. We don't use 2to3 anymore but a single code base with > helpers in

Re: [Scikit-learn-general] mean(scores) vs score(concatenation). E.g. AUC with LOO validation

The 0.13.X version of scikit-learn doesn't support grid search with an aux score. In the master branch, this is possible thanks to Andreas (see https://github.com/scikit-learn/scikit-learn/pull/1381) However, there is still work in progress on this subject see https://github.com/scikit-learn/sci

Re: [Scikit-learn-general] Paris Sprint location

Hi, I'd like to improve ICA as discussed in #2113. Also I'd like to inquire memory behavior of the decomposition classes to better support their combined application on bog data sets with more than 100k samples. Also I'd like to take a look at API inconsistencies between those classes, e.g with r

[Scikit-learn-general] Paris sprint: Various informations

Hi everyone, Here are some information about the sprint and Paris in general, in 4 points: 1. Traveling in paris: tickets and costs 2. Direction to the venue 3. Ethernet cables 4. My contact information. *1. Traveling in Paris: tickets and costs* The easiest way to go from one point to ano

Re: [Scikit-learn-general] User Survey

Hi Andreas. I know this thread goes back a little way...do you remember if any of the feedback you got in the survey indicates the type of problems that people are solving with scikit-learn? E.g. NLP for classification, recommendation in ecommerce (two examples I'm familiar with here in London)? Pe

Re: [Scikit-learn-general] Text processing using nltk, sklearn and pandas

2013/7/14 Lars Buitinck : > 2013/7/12 Olivier Grisel : >> 2013/7/12 Lars Buitinck : >>> 2013/7/12 Antonio Manuel Macías Ojeda : >>> Pretty good results actually. I was clustering these words to get >>> extra features for a NER tagger, which immediately got a boost in F1 >>> score. >> >> Interesting