Re: [Scikit-learn-general] "Improved" GLMNET

Hey Immanuelle, Thanks for the summary of your reading of the paper. This is useful. On Thu, Jul 19, 2012 at 05:37:31PM +0200, iBayer wrote: >What's the main point of implementing regularized log loss in >scikit-learn? I think that you and I have the same needs here: we need some code th

Re: [Scikit-learn-general] LinearRegression() doesn't work

On 07/19/2012 10:30 PM, Zach Bastick wrote: Thanks Sure enough making those changes fixes it. Are those going to be inclued in the next release? Yes, sure, all fixes in the master branch will be in the next release. The release is basically the master branch at a state when we decide it loo

Re: [Scikit-learn-general] LinearRegression() doesn't work

Thanks Sure enough making those changes fixes it. Are those going to be inclued in the next release? Zach On 19/07/2012 16:42, Warren Weckesser wrote: On Thu, Jul 19, 2012 at 3:41 PM, Warren Weckesser > wrote: On Thu, Jul 19, 2012 at 2:45 PM

Re: [Scikit-learn-general] LinearRegression() doesn't work

Thanks. I knew there was something like that. - Ursprüngliche Mail - > Von: "Warren Weckesser" > An: scikit-learn-general@lists.sourceforge.net > Gesendet: Donnerstag, 19. Juli 2012 21:42:47 > Betreff: Re: [Scikit-learn-general] LinearRegression() doesn't work > > > > > > On Thu, Ju

Re: [Scikit-learn-general] LinearRegression() doesn't work

On Thu, Jul 19, 2012 at 3:41 PM, Warren Weckesser < warren.weckes...@enthought.com> wrote: > > > On Thu, Jul 19, 2012 at 2:45 PM, Andreas Müller > wrote: > >> Oh, I'm on 0.11 btw on work and this is windows so I won't bother >> checking out. >> Was this fixed recently? It seems a weird bug, thoug

Re: [Scikit-learn-general] LinearRegression() doesn't work

On Thu, Jul 19, 2012 at 2:45 PM, Andreas Müller wrote: > Oh, I'm on 0.11 btw on work and this is windows so I won't bother checking > out. > Was this fixed recently? It seems a weird bug, though. > So for two dim cases we get coef_.shape = (2,), for three-dim we get (1, 3) > Looks like it was fi

Re: [Scikit-learn-general] LinearRegression() doesn't work

Zach: This is more of an inconsistency than a bug, shouldn't be that big a problem. for example you could call .ravel() on the output. - Ursprüngliche Mail - Von: "Zach Bastick" An: scikit-learn-general@lists.sourceforge.net Gesendet: Donnerstag, 19. Juli 2012 21:04:52 Betreff: Re: [Sci

Re: [Scikit-learn-general] LinearRegression() doesn't work

I'm on Windows 7. That makes 3 of us with the bug. Ordinary Least Squares (OLS) Regression is something so common that I don't understand why no one noticed this before. I really need it to work, so if anyone finds a solution, please share! Zach On 19/07/2012 15:45, Andreas Müller wrote: > Oh,

Re: [Scikit-learn-general] LinearRegression() doesn't work

Oh, I'm on 0.11 btw on work and this is windows so I won't bother checking out. Was this fixed recently? It seems a weird bug, though. So for two dim cases we get coef_.shape = (2,), for three-dim we get (1, 3) - Ursprüngliche Mail - Von: "Andreas Müller" An: scikit-learn-general@lists.so

Re: [Scikit-learn-general] LinearRegression() doesn't work

On Thu, Jul 19, 2012 at 2:29 PM, Zach Bastick wrote: > I don't understand how it works on your side. I tried: > Uninstalling ALL python modules and python, and reinstalling only > python, numpy, scipy and Scikit-learn > Uninstalling ALL python modules and python, and installing Enthought > Python,

Re: [Scikit-learn-general] LinearRegression() doesn't work

I can reproduce the error. Weird. Looking into it. - Ursprüngliche Mail - Von: "Zach Bastick" An: scikit-learn-general@lists.sourceforge.net Gesendet: Donnerstag, 19. Juli 2012 20:29:47 Betreff: Re: [Scikit-learn-general] LinearRegression() doesn't work I don't understand how it works on

Re: [Scikit-learn-general] LinearRegression() doesn't work

I don't understand how it works on your side. I tried: Uninstalling ALL python modules and python, and reinstalling only python, numpy, scipy and Scikit-learn Uninstalling ALL python modules and python, and installing Enthought Python, and I still get the same error. This happens, running Python i

[Scikit-learn-general] Scikit-learn for distributed processing?

Dear all, I am trying to find information on the use of scikit-learn on very large datasets, i.e. if and how it can be used in a distributed processing setup. I saw that PiCloud has scikit-learn installed in their environment, and this post suggests it can be deployed on PiCloud: http://stackoverf

Re: [Scikit-learn-general] "Improved" GLMNET

I went through the paper, for me the take home points are the following: The comparison has to be taken with a grain of salt. The GLMNET implementation used in the comparison was v.1.5.3 while the current version (1.6) has new stopping rules and uses strong rules to early discard features. [2] Bet

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Ah, i see! Thanks a ton, Brian!:) I really appreciate it :) regards, shankar. On Thu, Jul 19, 2012 at 6:02 PM, Brian Holt wrote: > Hi Shankar, > > I can reproduce your error and it turns out to be quite a subtle bug > in your class. > > self.reg=RandomForestRegressor(n_estimators, max

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Hi Shankar, I can reproduce your error and it turns out to be quite a subtle bug in your class. self.reg=RandomForestRegressor(n_estimators, max_depth, min_samples_split, random_state, n_jobs=1, **kwargs) self.clf=RandomForestClassifier(n_estimators, max_depth, min_samples_split,

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Sure! Here is X_train[0:5] : [[ 0.65179688 -0.9582746 1.04295266 -0.26265019 -0.14655937 1.58062601] [-1.53422034 -0.9582746 1.04295266 -0.26265019 -0.14655937 -0.31644633] [ 0.65179688 1.04354227 -0.95881623 2.22137356 -0.67541134 0.3159] [ 0.65179688 -0.9582746 1.04295266 -0.883

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Hi Shankar, Can you paste a small snippet of your data (X_train, Y_train) that reproduces this behaviour? Cheers Brian -- Live Security Virtual Conference Exclusive live event will cover all the ways today's security and

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Done! Updated to last commit, and tried again. I still get the same error message. regards shankar. On Thu, Jul 19, 2012 at 3:53 PM, Gilles Louppe wrote: > Can you try updating your version then to the latest commit and then > check if this still happens? The tree module has been updated very

Re: [Scikit-learn-general] LinearRegression() doesn't work

Works on my side as well.. Can't reproduce your error. 2012/7/19 Gael Varoquaux > On Wed, Jul 18, 2012 at 05:50:16PM -0400, Zach Bastick wrote: > > But this doesn't: > > clf.fit([[0, 0, 0], [1, 1, 1]], [0, 1]) > > Works for me. > > I don't understand what excatly is your problem. > > Gael > > >

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Can you try updating your version then to the latest commit and then check if this still happens? The tree module has been updated very recently (yesterday). Gilles On 19 July 2012 09:34, Shankar Satish wrote: > Hi Gilles, > > I'm using 0.12. > > regards > shankar. > > > > On Thu, Jul 19, 2012 a

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Hi Gilles, I'm using 0.12. regards shankar. On Thu, Jul 19, 2012 at 3:28 PM, Gilles Louppe wrote: > Hi, > > What version of scikit-learn are you using? 0.11 or dev? > > Best, > > Gilles > > On 19 July 2012 06:34, Shankar Satish wrote: > > Hello everyone, > > > > I have a custom prediction c

Re: [Scikit-learn-general] Unable to call fit() on random forest classifier when it is encapsulated in separate class

Hi, What version of scikit-learn are you using? 0.11 or dev? Best, Gilles On 19 July 2012 06:34, Shankar Satish wrote: > Hello everyone, > > I have a custom prediction class which in fact consists of a random forest > regressor+classifier. The class implements a fit() method, which in turn > c

Re: [Scikit-learn-general] Textbook advice

On 07/19/2012 06:25 AM, Gael Varoquaux wrote: On Thu, Jul 19, 2012 at 12:42:25AM +0200, federico vaggi wrote: I was looking for suggestions for a good ML textbook - ideally one that has a heavy emphasis on applications and is based on Python. I don't know personally any that is based on Python.